Gregory SA Ed The Design Method

THE
DESIGN METROD
THE
DESIGN METHOD
Edited by
S. A. GREGORY
On behalf of Design and lnnovation Group
University of Aston in Birmingham
Springer Science+Business Media, LLC

1966
ISBN 978-1-4899-6169-3 ISBN 978-1-4899-6331-4 (eBook)
DOI 10.1007/978-1-4899-6331-4
Suggested U.D.C. number: 601
©
Springer Science+Business Media New York 1966
Originally published by The severa! contributors named on pages xi to xiii 1966
Softcover reprint of the hardcover 1st edition 1966
PREFACE
THIS book about The Design Method provides a composite statement from different technologies
on the nature of design, covering the principal range of activities of the designer, and a relatively
unified structure of thought against which the work of the designer may be considered. A review
of work and thought on the design process is given, tagether with an appraisal of systematic
methods, and an analytical approach to the process of design, with the possibility of its extension
by scientific research. The relationships between the behavioural sciences and design are also
discussed. Finally, a glossary of design and a substantial, although not comprehensive, biblio-
graphy are included.
This pioneering effort provides practising designers with an opportunity to see their work
in the round and to gain some new insights. Managers are affered a better comprehension of the
problems and outlook of designers, with possible ways to more and better designs. To researchers
some new challenges are given, and to industry and consumers at large the promise of a more
rational approach to design. The attention of educationists is particularly directed to the fact
that there is a design method which is at least as well defined as the scientific method, and
poses more human and personal problems in its application. To the youth of today the book offers
a rather intellectual glimpse of a world of skill, service and excitement, which is able to absorb
to the Iimit the sum of human abilities. It is hoped that it will provide a foundation upon which
students of design can build in the future.
The initial stimulus for this book came in 1964 at Scarborough, where a conference on
education in design was being held. This was a good conference, but it seemed from the dis-
cussion that people were making assumptions about the nature of design which differed widely.
Some people apparently saw design in terms of what went on to a drawing board; others took it to
be something happening inside a designer's head, with the drawing board used only to help
communication. There also seemed to be a preponderance of 'hardware' men and very few 'system'
men.
It was decided to call a further symposium, with the intention of exploring in a relatively
detailed way what the designer does. The symposium was to be open to designers and other
interested people of all persuasions. Indeed, it was hoped to establish a common basis of agree-
ment about the nature of 'the design method', using this phrase in the same way as 'the scientific
method'. The Birmingham College of Advanced Technology (now the University of Aston in
Birmingham) was chosen as the venue, in order that the existing Design and Innovation Group
could be used as a working centre. At the time, this voluntary, but officially recognized, inter-
disciplinary group was unique in Great Britain. It draws its members from the technological
faculties, from industrial administration teaching staff, from the College of Art, and informally
from industry.
An outline of a possible programmewas produced and invitations for contributions published
as widely as possible through the engineering and other institutions, and the technical journals.
Particular use was made of contacts with people connected with specialist 'networks' dealing
with design throughout the country. Several of these have held conferences in the last few years.
V
vi PREFACE
Of great assistance were members of the Conference on Design Methods 1962 and the ad hoc
Committee on Electronics Design of the Institution of Electrical Engineers. Other groups also
helperl but it would be out of place to give a full list here although it is recommended that some
attention should be given to aiding atleast theinformal interchange of ideas between these groups.
On the basis of a number of voluntarily offered papers a pattem began to emerge, within
which it became possible to seek out and obtain others. Certain papers offered were rejected.
The common reasons for rejection tended to be lack of infounation relevant to the design operation,
or the reproduction of what was obvious or known traditional practice. In some cases the papers
were too specialized and therefore of restricted interest relative to the scope of the symposium.
The symposium took place on 21st to 23rd September 1965, and was attended by more than
two hundred people drawn from the mostdiverse branches of technology and design. Papers were
presented very briefly because preprints had been circulated. Discussion was fully recorded by
tape.
This book is derived from the papers and recorded discussion. The papers are very largely
those presented at the symposium with only minor modifications. The discussion has been subject
to a marked amount of editorial work in order to provide readability. Significant points of dis-
cussion now appear as editorial comment and acknowledgment is marle to contributors. Extra
material has been written to give background, continuity, linkage and better coverage. The aim of
the book as it now stands, is to offer to the interested reader co-ordinated information about the
design method in the words of specialists in selected fields, and the possibility of pursuing the
subject further and in depth. Attention is particularly directed to the use of the book for sturlies
in the philosophy of engineering to suit requirements of the Council of Engineering Institutions.
To all who have contributed, whether by word, by action, or by encouragemen t, the editor
gives grateful thanks. Above all thanks are due to Christopher Buck who carried the load of
organization in his capacity of secretary of the Design and Innovation Group. Acknowledgments
are thankfully accorded to the University of Aston in Birmingham for the use of numerous facilities
and for providing a background without which the symposium would not have been able to occur
and the book to appear.
S. A. G.
Birmingham
CONTENTS
Page
Preface . V
List of Contributors . xi
Part I. The Design Method
1. Design and the Design 1VIethod

S.A. Gregory 3
2. The Design Method in Practice

R.]. McCrory 11
3 Definitions and Methodologies

W.E. Eder. 19
Part II. The Human Perspective
4. The Human Perspective, the Design Situation and its Opportunities

S. A. Gregory 35
5. Design and Human Satisfaction

W.H. Mayall 39
6. Invention, Design and Market Research

A.F. Stobart 45
7 Ergonomics and Design
B. Shackel. 49
8. Sociology and Design

.4.M. Penny 59
9. Design Acceptance
M.E. Peplow 65
vii
viii CONTENTS
Part 111. The Elements of Design
10. A More Detailed View of Design

S. A. Gregory 77
11 The Elements of Design
R.D. Watts. 85
12. Infonnation for Design

]. Farradane 97
13. Patterns
A.D. Newman 105
14. Creativity
G.H. Broadbent 111
15. Choosing and Evaluating

A.G. Pleydell-Pearce. 121
16. Design and Decision

S. A. Gregory 131
Part IV. Design Techniques
17. Models in Practical Design

S. A. Gregory 143
18. The Implications of Economics in Engineering Design

A.P. Shahbenderian 157
19. The Use of a Digital Computer in Design Offices

K.C. Parton 167
20. A Practical Design: An Oil Bumer for Large Watertube Boilers

A.M. Needham 175
21. Practical Design Basedon Method (Fundamental Design Method)

E. Matchett and A.H. Briggs 183
22. Design Realization

F.L. Ellis . 201
23. Selection of Materials

A.L. Davies 211
24. ReHability and Maintenance

C.T. Comey 219
CONTENTS ix
25. Design of Materials

S. A. Gregory 235
26. New Ideas in the Drawing Office

P. McMullen 241
Part V. Management and Design
'1:7. Design Policy Fonnutation

B.T. Turner 249
28. Innovative Design as a Policy Function

R. Davis 259
29. Effect of Organizational Procedures on Design - An Outline of the Problems

I.M. Ross 269
30. Communication
C.H. Buck. 279
31. Design, Management, Realization and Change

S. A. Gregory 287
Part VI. Design Research
32. Design Methods Reviewed

].C. ]ones. 295
33. Technologies and Varieties of Design

W.E. Eder . 311
34. Preliminary Research into Electronics Design

H. V. Beck 317
35. Design Science

S. A. Gregory 323
Glossary 331
References and Bibliography 335
Name Index 343
Subject Index 347

LIST OF CONTRIBUTORS
H. V. BECK, M.A., B.Sc., C.Eng., M.I.E.E.

Research Manager, Marconi Instruments Limited, St. Albans, Hertfordshire;
Chairman, Institution of Electrical Engineers Committee on Electronics Design
A.H. BRIGGS, B.Eng. (Hons)

Assistant Chief Draughtsman, Foundry Products Drawing Office,
English Steel Castings Corporation, Sheffield
GEOFFREY H. BROADBENT, B.A., A.R.I.B.A.

Lecturer in Architecture, Department of Architecture, University of Sheffield
C. HEARN BUCK, B.Sc., A.lnst.P., C.Eng., A.M.Inst.Gas E.

Lecturer in Product Design, Department of Industrial Administration,
University of Aston in Birmingham;
Secretary, Design and Innovation Group
C.T. CORNEY, B.Sc., C.Eng., A.M.I.E.E., A.R.C.S.

ReHability Executive, CAV Limited, London W3
A.L. DAVIES, B.Sc., Ph.D., A.I.M.

Senior Consultant, Product Planning Limited, Croydon, Surrey
ROBERT DAVIS, C.Eng., A.M.I.Mech.E., A.M.I.Prod.E.

Research Engineer, Reed Development Services Limited, Research Centre
(Reed Paper Group ), Aylesford, Kent
W.E. EDER, lng. (Vienna)

Lecturer in Mechanical Engineering, University College of Swansea
F.L. ELLIS, C.Eng., A.M.I.Mech.E., A.M.Inst.R.

Lecturer in Mechanical Engineering, Department of Mechanical Engineering,
University of Aston in Birmingham
J. FARRADANE, B.Sc., A.R.C.S., A.R.I.C., F.I.Inf.Sc.

Information Scientist;
Senior Research Fellow, City University, London ECl
S.A. GREGORY, B.Sc., C.Eng., M.I.Chem.E.

Lecturer in Chemical Engineering, Department of Chemical Engineering,
University of Aston in Birmingham;
Industrial Consultant
xi
xii LIST OF CONTRIBUTORS
]. CHRISTOPHER ]ONES
Senior Lecturer in Industrial Design Technology,
Manchester College of Science and Technology
R.J. MCCRORY, Mem.A.S.M.E.

Associate Manager, Mechanical Engineering Department, Battelle Memorial Institute,
Columbus, Ohio
P. MCMULLEN, A.C.G.I., B.Sc., D.I.C.

Formerly of ICI Limited, Plastics Division, Welwyn Garden City, Hertfordshire
E. MATCHETT, C.Eng., A.M.I.Mech.E., A.M.I.Prod.E.

Senior Lecturer and Consultant, Department of Work Study and St'aff Training,
Engineering Employers' West of England Association, Bristol 8
W.H. MAYALL, A.F.R.Ae.S.

Industrial Officer, Capital Goods, Council of Industrial Design, London SWl
A.M. NEEDHAM, B.Sc. (Eng.)

Research Officer, Centrat Electricity Generating Board,
Regional Research and Development Department, Portishead, near Bristol
A.D. NEWMAN, B.Sc., A.Inst.P., C.Eng., A.M.I.E.E.

Senior Principal Scientific Officer, Mini stry of Technology, Autonomics Division,
National Physical LaboratoJ.Y, Teddington, Middlesex
KENNETH CHARLES PARTON, B.Sc.(Eng.), C.Eng., A.M.I.E.E.

Manager, Engineering Services, General Electric Company Limited, Birmingham 6
A.M. PENNY, B.Sc.

Architectural Sociologist, Institute of Education, London WCl;
Convenor, The Design Group, British Sociological Association
M.E. PEPLOW, B.Sc., A.Inst.P., C.Eng., A.M.I.E.E.

Research and Development Manager, Centrat Electricity Generating Board,
Regional Research and Development Department, Porti.shead, near Bristol
A.G. PLEYDELL-PEARCE, B.A.

Staff Tutor in Philosophy, Department of Extra-mural Studies, Uni versity of Birmingham
I.M. Ross
Engineer; Director, Ether Engineering Limited, Bushey, Hertfordshire
B. SHACKEL, M.A.
Research Psychologist;
Head of Ergonomics LaboratoJ.Y, EMI Electronics Limited, Hayes, Middlesex
A.P. SHAHBENDERIAN, M.A., Ph.D., C.Eng., A.M.I.Chem.E.

Consulting Chemical Engineering Economist, 28 Comwall Gardens, London SW7;
Part-time Lecturer, Department of Chemical Engineering and Chemical Technology,
Imperial College of Science and Tedmology, London SW7
LIST OF CONTRIBUTORS xiii
A.F. STOBART, B.Sc., C.Eng., A.M.I.Chem.E.

Engineering Merchant;
Manager, European Operations, Capitol Pipe and Steel Products Incorporated, London Wl
B. T. TURNER, M.Sc.(Eng.), C. Eng., M.I.Mech.E., A.M.l.C.E., A.F.R.Ae.S.

Principal, Staff Training Centre, English Electric Company Limited,
Dunchurch, Warwickshire
RONALD D. WATTS, B.Sc.(Tech.), C.Eng., A.M.I.E.E.

Lecturer (specializing in Instrumentation and Measurement), Department of Production
Technology and Control Engineering, City University, London ECl
PART I
THE DESIGN METHOD

Chapter I
DESIGN AND THE DESIGN METHOD
S. A. Gregory
lntroduction does something with his hands such as draw

This book is about design and designers. on paper, or mould a lump of clay. This is a
It is concemed with the process of design: a design situation in which one man may be
process the pattem of which is the same doing everything: he thinks about the products
whether it deals with the design of a new oil and then forms them. Most practical design
refinery, the construction of a cathedral, or today is split into many stages. Some people
the writing of Dante's Divine Comedy. Much think about the generat scope of the product;
of practical design is humdrum but the same others think about details of the product;
pattem persists. others specify ways in which parts should
This pattem of work, whether conscious be made in workshops, and so on.
or unconscious, is the design method. The The initial work is largely done in the
design method is a way of solving certain head and for this reason remains unseen.
classes of problem: relating product with Only the products which appear in material
situation to give satisfaction. A study of the form come to the eyes of the man in the street.
design method helps one to a better under- It is largely from these products or from
standing of design and to the possibility of activities associated with their use that the
bringing new forces to play in dealing with value of a design may be judged. People
design. For the man in the street a study of therefore tend to think of design as an operation
design method means the potentiality of better strongly related to material things, if not
products and greater satisfaction in those actually something poured into them. Designers,
products. according to this simple view, either make
Although the book cannot be comprehensi ve machines which execute some obvious function,
there is an attempt to deal with most of the or they shape things to be pleasing to the eye.
features of design which appear to have some The former works at a drawing board and is
general significance to practical designers concemed with getting objects made out of
and students of design. cast iron or mild steel; the latter has long hair,
Design is concemed with making things a peculiar taste in clothes, and plays about
that people want: with building up pattems with clay and colours.
which have value. These things have to be It is araund such common conceptions
thought about and made. Design involves a that public discussion, political action, even
thinking activity and an executive activity. sectional promotion, tend to duster. When
This is true whether settees, space satellites, design is scrutinized at high Ievel by an
or sonnets are designed. outsider it means either mechanical design,
where the term mechanical has an ill-defined
Designers meaning, or it stands for 'appearance' design,
The most common understanding of the which its practitioners call industrial design.
activity of a designer seems to be that he These are only certain aspects of design in
3
4 THE DESIGN METHOD
its totality. All kinds of engineers and execution is implicit in the very name.
technologists have an involvement in design. Engineers such as Brindley, Rennie, Telford,
All artists, in whatever medium they may work, Whitworth, became established as craftsmen,
are designers. This is true of architects, skilled in manual operations, before proceeding
electrical engineers, metallurgists, poets, to the stage of master engineers. In this latter
works managers, as well as a host of other capacity they gave instructions to workmen by
people and professions. Fundamental to this drawings or other means so that their visions
general idea of a designer is the building up might be fulfilled in iron or stone. Here, if the
of a structure, pattem, or system within a model was communicated by drawing, this
situation. drawing was an instruction.
It is partly from such a tradition that
The ldea and the Action young people who wish to become designers
People such as painters appear to develop are steered into periods of training in work-
their designs as they put them on canvas. shops and then in drawing offices. Fresh
In fact there seems to be some connection thought is due about the necessary reasons
between the practical development of the for such kinds of activity.
pattern of paint under their hands and the fact From seeing the essential element in
of manual work. But although it is reasonable design skill as residing in the hands of
to expect some interaction between what they craftsmen, people began to see the preparation
are doing and have done practically and their of the instruction to the craftsman as
thoughts about the next strokes, it is recognized containing the vital operation. In architecture
that the painter must be building up a design in and engineering emphasis began to be placed
his mind's eye before committing it to colour. upon the drawing board.
Many painters prepare sketches first. That the drawing board is not an essential
The painter develops a model in his head feature in design is well illustrated by the
before bringing about its realization on canvas. example of J ames Brindley, as recorded by
This is probably true even with 'action' SMILES (1874). His brother-in-law said of the
painting. Instead of using a separate sketch great canal-builder: 'When any extraordinary
some painters make running alterations to difficulty occurred. . . . having little or no
what they have already painted and these assistance from books or the labours of other
alterations constitute the transition from a men, his resources lay within himself. In order,
sketch to the final form. According to this therefore, to be quiet and uninterrupted whilst
view the painter has two kinds of activity at he was in search of the necessary expedients,
least: the construction of a mental model, and he generally retired to his bed; and he has
the transference of this mental model by hand been known to be there one, two, or three days,
on to canvas. To carry these out successfully till he had attained the object in view. He
he needs skills. Comparable sets of skills would then get up and execute his design,
must also be accorded to all artists who without any drawing or model. Indeed, it was
produce material objects with their hands: the never his custom to make either, unless he
sculptor, the creative potter, and so on. was obliged to do so to satisfy his employers.'
Since time immemorial pupils have learned Within the modern engineering industry
skills by working with a master craftsman, a most of the work which is done on drawing
combination in which the pupils begin by boards is to provide instructions to other
practising manual skills directed towards the people about the way in which material is to
execution of ideas developed by the master. be shaped, joined, or assembled. This is
In architecture the separation between the largely carried out by men who have had some
generation of controlling ideas and the practical experience of manufacturing and production
THE DEmGN METHOD 5
methods. In Great Britain they belong to the proceed to the preparation of drawings for
dass of mechanical engineers although in the manufacture and assembly. Here, function
USA they are frequently termed industrial design and production design can be identified.
engineers, the name of mechanical engineer In some other kinds of engineering, in
being used in a more restricted sense. chemical engineering and in electrical
Designers in the engineering industry who use engineering particularly, it is possible to find
drawing boards are likely to be more concerned at least three stages of work. The primary
with finding out the best way to develop the function design is done by one kind of
shape of a part or an assembly rather than specialist engineer; this is then followed by
a mechanical embodiment design; in turn
with preparing drawings for the purpose of
instruction. The drawing is used as a model comes a production design.
for solving problems in design.
It should be noted that the drawing is Change - Evolution and Revolution
likely to become less important in future, To many people in design this view of the
either as a means of communication or as a several stages may not be immediatelyfamiliar.
model for working out shapes or arrangements. This is probably attributable to the fact that
Instead of using drawings which have to be most of practical design activity is invol ved
read by a plant operator, it is already becoming in modification, in making relatively small
worth while to provide machine-toolinstructions changes to products or machines which already
by communicating directly to the machine by exist. This is essentially evolutionary design.
punched tape which provides numerical control. Such changes as take place occur within a
Experiments are being made on working out well-defined situation. Even the designer
shapes on a cathode-ray tube presentation. occupies a well-defined position within a
This kind of shape presentation may be readily closely structured company arrangement, quite
converted into a punched tape communication. typically consisting of vertically defined
But the production of the major ideas - the functional divisions, such as sales, production,
guts of design - is usually done by people at finance, and engineering.
least one remove from the drawing board. The People in design who are associated with
distance from the drawing board depends very major new projects are, on the other band,
much upon the type of design which is being likely to be acutely aware of the inter-
undertaken. In mechanical engineering design, dependence of the various classes of engineer-
which may be concerned with such things as ing design. Indeed, it is fair to say that much
motor cars, machine-tools, refrigerators, of the interest in the new thinking about design
boilers, etc. , there are usually at least two comes from people who have this kind of
stages of design work. In the first stage the involvement. The types of system with which
requirements are worked out for the fulfilment these new projects are largely concerned,
of the function of the product. Thus, for a are the relatively complex arrangements of
motor car, the nurober of seats and the space equipment needed to perfonn some major
needed for getting into the seats and for comfort economic function. Such large systems (the
while in the seats, and the baggage space range of which is considered later in the book)
must be settled; the speed and manoeuverability may include power stations, intercontinental
of the car have to be agreed, and so on. Only telecommunication networks, new towns, etc.
when these functions have been detennined In their own right, these projects are of such
and their balance worked out is it possible magnitude as to demand the most careful
to consider alternative ways of carrying them thinking about their design, so that the best
out in 'hardware'. From the preparation of use may be made of social resources. Thought
schemes for the hardware, it is possible to about specific design leads to thought about
6 THE DESIGN METHOD
the nature of design in general. Many of these systems such as oil refineries, transport
!arge projects are essentially copies of systems, etc., require to be designed afresh
projects which have gone before, but suitably for each case, and furthermore the assembly
adjusted to suit the needs of the local situation. system has to be newly worked out. It is this
Every now and then, however, radically new kind of operation which soaks up design
projects have to be faced. This leads to even capacity.
more determined thought about the fundamentals The study of design is therefore critical
of design. in terms of national economic success and
In the USA there is a clear connection survival.
between the space programme and fresh
thinking about the nature of design. In Great
Britain much of the new approach may be The Design Method and Systematic Procerlure
related to those sections of industry which The conscious identification of the
are undergoing expansion, and the professions design method is quite recent, although
associated with them. There is also, both in precisely when the concept appeared is not
the USA and Great Britain, a considerable clear. It is important for designers to become
impulse from branches of study such as aware, consciously, of the existence of the
ergonomics which were called into being to design method. The design method and its
deal with problems of man- machine interaction. practice distinguish the engineer from the
In Great Britain it is likely that the tasks scientist. Each is a problern-solver but has
of economic expansion, particularly along the different kinds of problern to deal with. Put
lines suggested by the National Plan, may be simply, the scientific method is a pattem of
such as to put considerable strain upon the problem-solving behaviour employed in finding
limited resources of design capacity. According out the nature of what exists, whereas the
to the figures put forward regarding the design method is a pattem of behaviour
preferred extent and directions of expansion, employed in inventing things of value which
the shortage of design capacity is likely to be do not yet exist. Science is analytic; design
felt in the field of consumer durables (cars, is constructive.
houses, etc.) and probably even more in the Problem-solving is a very general kind of
design of !arge projects such as chemical activity; a simple introduction is by HODNETT
plants. The chemical industry is scheduled (1955). It can be regarded as the search for and
as the largest recipient of capital investment discovery of means to achieve or prevent
in the industrial sector as a whole. Power transformation from one state of affairs to
production is the most important manufacturing another, where the affairs may be abstract
activity in the public sector. Behind such or concrete. Although problem-solving is so
requirements for design, expressed primarily general in character there are a number of
in terms of design of the relevant chemical or relatively systematic approaches which may
electrical systems, come the civil engineering be used. An example used in industry is the
problems, the manufacture of the necessary 'systematic appraisal' technique. This is at
hardware, and the assembly and commissioning the basis of what is known as work study.
at site. The most widely-known dass of problem-
Universal items such as electronic solving activity is the scientific method. In
computers and machine-tools, serve a wide dealing with the problern of investigating what
range of industry and in that sense are critical. exists in nature, the method involves the
Their design, once achieved, continues to be generation and testing of models of parts or
reproduced in manufacture for considerable the whole of the universe. The scientific
runs. On the other hand, the complex one-off method has attained such a reputation that it
THE DESIGN METHOD 7
now tends to be seen as the prototype of all which skilled designers use and the
other kinds of problem-solving activity. organization ofthesetechniques into systematic
The human brain is endowed with many methods which may be suggested to other
abilities which may be devoted to problem- designers or those in training.
solving. In particular, GUILFORD (1959), It is important to recognize that systematic
has shown how they may be classified in procedures will not by themselves produce
terms of the categories: perception, memory, outstanding design. While they may be seen
convergent thinking, divergent thinking, and as a way of raising the general level of
judgment. Within these classes of operation competence in journeyman design and as a
the intellect may deal with contents which check-list of procedures for other designers,
are either concrete (as exemplified by the they carry intrinsic danger. This is the danger
person who is a 'visualizer'), symbolic, of routine behaviour, of adherence to a
semantic (in this case the exemplification is disciplined drill. To make prov1s1on for
through .the 'verbalizer'), or behavioural. The dealing with this it is important to emphasize
contents may deal with such products as units, repeatedly the need for freshness and the
classes, relations, systems, changes and creative approach to the situation and the task.
implications.
The central activity of engineering, Human Satisfaction and the Design Situation
technology, and art is the design method. The end of all design is human satisfaction.
Since this is a kind of problem-solving it If a design fails to deliver satisfaction it
will have resemblances to other kinds of fails as a design. The early civil engineers
problem-solving. Technologists, w i t h out who, a century and a half ago, first saw them-
conscious knowledge, frequently switch from selves as professional engineers, described
the design method to the scientific method. their function as the harnessing of the great
Recently, because of the confusion which powers in nature for the service of man. Just
exists, (a confusion which has had undesirable as manufacturing concerns have tended to
educational and other consequences) there become production orientated, so have
have been welcome attempts to differentiate engineers tended to become obsessed by
between the activities of science and those of strictly technical problems. The user and
technology and engineering. Particularly customer have become neglected. Partly this
readable are the little books by SPORN (1964) has come about from the specialization of
and KRICK (1965). functional departments within industrial
Many recent authors have described the companies. An even greater pressure has
design method in considerable detail and later probably come from teaching institutions
chapters of this book pursue the discussion. which have concentrated upon technical
A compact description is given by ASIMOW knowledge and its acquisition.
(1962) and, within a perspective of system This tendency to disregard the consumer
engineering, by HALL (1962). These is not only a retreat from the spirit of
descriptions are based upon much discussion engineering but it carries the possibility of
within a background of practice. national disaster. In a competitive world it is
This concentration of attention upon the mastery of the market which prevails, and
design has brought about a considerable the mastery of the market means design for
interest in systematic procedures for design. the consumer. The market not only represents
Part of this interest in systematic procedures the consumer; to the designer it is also the
has come directly from work study; but the fact challenge and the opportunity. The need of
that the design method may be identified the times is for the designer to be turned
leads to the study of the particular techniques towards the market. All those means which
8 THE DESIGN METHOD
may be legitimately exploited to increase communicate because of the inadequacy of the

satisfaction should be at the disposal and concepts which are currently used. It is likely
command of the designer, who should see the that in some of the finer detail it will only be
consumer as his primary target. For this possible for one designer to talk to another
reason the reader is directed towards that rather than to people at large. A stage is
section of this book which deals with human reached in design where a series of moves has
satisfaction and the opportunities presented. to be made rather like those undertaken by
skilled players of chess. Strategies must
be adopted.
The Elements of Design The study of design strategies is only
When the practice of design is first looked recent and is not dealt with in Chapters 10 to
into in some detail and with some intellectual 16, although it undergoes some examination in
refinement, it can be a disturbing experience. Chapter 32 in particular. There is a more
People who have not previously taken part in detailed discussion of the matter by GREGORY
practical design may wonder what is being (1966). These strategies, which are procedures
discussed. Practical designers of the 'no for dealing with complex and uncertain
nonsense' variety may also have distinct situations, are phenomena which may be
misglVmgs. Indeed, some of them may be clearly observed in practical design. A number
provoked to scom and ridicule. Chapters 10 to of them has already been recorded as the
16 of this book may prove difficult reading to result either of reflection upon practice or,
newcomers and it might be advisable for them in the past few years, as the result of specific
to avoid detailed examination of this section observation of design. Same of the methodical
on the first reading. The difficulty lies in the design procedures already have a selection of
fact that it is concerned with an examination the more obvious strategies built in.
of the process of design and its successive The study of the elements of the design
stages in a theoretical manner. It is concemed process, although at first appearing rather
with advancing a model of the process of abstract, is, in its outcome, severely practical.
design and with investigating the principles
and basis of each of the stages.
The practical value of this kind of approach The Practice of Design
lies in the production of a model or set of A person may build up his knowledge of
models of design behaviour~ These models the way of design by reviewing what has
may be tested by experiment or against occurred in practice within his personal
practical observation and, with suitable experience. To broaden the basis of knowledge
validation, may provide the foundation for he studies the work of other people.
practical advice. Indeed, most of the methodical Eventually he constructs a picture in his mind
procedures which have been put forward for of the pattern of behaviour during design.
helping design rely upon some analysis of Other people do the same kind of thing.
this kind. Eventually it becomes possible to generalize
It must not be assumed that all is known this kind of experience and to pass it from one
about the elements of design just. because person to another. This communication will
they may be expressed in the form of diagrams. almost inevitably use models of one kind or
These diagrams represent a relatively simple another. The models will not only represent
kind of model of the design process. The one person's view of things but will provide
detailed course of design in a particular apportunilies for working out examples in a
situation may be much more complex, even for rather general way. Chapters 17 to 21 deal
a small product. For very sophisticated and with models of various kinds used in design,
novel designs the process may be difficult to either with specific examples of models in
THE DESIGN METHOD 9
practice, or with more general models. These it tends to be adopted without questioning.
chapters, in fact, show the use of models in Value engineering has shown the significance
the problem-solving activity of design, in of questioning, as did work study before it.
design for function and in the formulation of a The perpetual questioning arising from a
product. scientific approach to this area of design
After the practice of function design carries with it the promise of further returns.
through the use of models, a practice which is These returns may not come directly from the
concerned with development of design up to the practised procedure but from the invention and
point that steps may be taken towards its development of new mechanical or electrical
realization in some material or mechanical techniques to help the designer. But t h e
form, comes that region of design which is designer needs to be concerned not only with
more familiar to many working designers, design techniques, not only with mechanical
i. e. design for realization. or electrical assistance, but also with the
Design for realization is concerned with potentialities of the new materials, which may
all those aspects of practical design which substantially change the basis of realization.
include the selection of the best kind of All practical design, whether at the
mechanical features to provide for the carrying system, the function, or the realization level,
out of the required function; the selection of is likely to be the activity of some organization.
the materials which give the best performance The way in which design interacts with
within the limits of the situation; the choice of management is the concern of Chapters 27 to
the most suitable methods of manufacture and 31. Here is discussed the way in which design
the reviewing of the design within the terms of is the practical expression of that part of
manufacture; the communication of these management policy concerned with the
decisions to the people or machines which are preparation for the exploitation of opportunities
to carry out the detailed fabrication of the held to be favourable to the company concerned.
components and parts and subsequently to the In order that the design carried out in the
pcople who have the task of putting the parts organization fulfils the general objects of that
together in a working system. F o r man y organization, the policy dealing with design
practising designers this is the most important has to be clearly thought out and effectively
part of design. Such is a matter of opinion. transmitted. The opportunities for the company
Certainly it is only part of the total picture: lie not only in the market as such but in the
function and system designers obviously see minds of its designers. How should the inter-
the situation differently. But in the last stages, change of thought and decision about such
in order to be useful, all design has to pass opportunities be best arranged? Designers do
through realization design. It is a necessary not achieve their best work as a general rule
part of design. under dictation and rigid discipline. How is the
Even at the level of realization design most productive use to be made of designers?
the general principles of design thinking hold. The need for innovative design to be an
Here the opportunities for methodical procedure expression of the total attitude of an industrial
and for creativity reveal themselves, often company is important. Design is not solely the
delivering substantial rewards. It is in this function of the designers. Other skills and
area that the recent emphasis on value other departments can help and assist in the
engineering and analysis has made successes. development of a generalinnovative atmosphere.
This has been discussed particularly by MILES Most important as a contributor to this is the
(1961) and in a series of articles in Product attitude of top management. Unfortunate indeed
Engineering (1965). is the designer who faces resistance at all
Because much of the work in realization levels. Unfortunate, too, is the company which
design is well understood and well established has this kind of creeping paralysis, since in the
10 THE DESIGN METHOD
long run it will die from it, as have many well- evaluating and improving design activity and
known British companies in recent years. effort.
Given the correct attitude for the reception The Scientific Study of Design
of innovative design, in support of a policy
This book, as a whole, is concemed with
clearly set out to take advantage of the
the theme of the design method. lt begins with
opportunities which exist and which will come a statement of the underlying thesis and goes
into being, it is necessary to provide for the on to deal with the principles as known and
free and unimpeded flow of information and with the important practical procedures.
ideas throughout the organization. Designers Chapters 32 to 35 are concerned with the
must have the data that they need available in development of the empirical study of design,
the correct form at the time required. They with speculations about the connection between
must have clearly defined boundaries of various types of design, and with proposals
decision which competently reveal their for research into design, both in respect of
responsibilities within their own company and design activity as such, and to aid the
towards the consumer. They need the opportunity development of new mechanical and electrical
to discuss and clarify whenever it is aids. Chapter 32 is a comprehensive review of
reasonable. The essence of these requirements
the recent development in thinking about
is an attention to the communication system of
design.
the organization. So important is this need that
in cases of new projects. the company Conclusion
organizational structure has to be modified in You are asked to read on and take what
order to make certain of maximum ease of advantage you choose of this attempt to put
communication. design into a new perspective, both intellectual
This section of the book concludes with and practical - an attempt to show the scope
Chapter 31, which deals with techniques of and challenge in the heart of engineering.
Chapter 2
THE DESIGN METHOD IN PRACTICE
R.J. McCrory
lntroduction an end result which serves a valuable purpose.

Just as in Great Britain, the designer in It is the segment of engineering which devises
the USA has been in danger of becoming the and develops new things, in contrast with
forgotten practitioner of technology. Because other segments which emphasize the solving
he must, almost by definition, be a generalist, of problems or the generation of engineering
the designer has been discounted in favour of information. The responsibility of the design
the specialists in more tangible areas. But in engineer is to use the maximum powers of
the USA this tendency to undervalue the creativity, judgment, technical perception,
designer seems now to be reversing in both economic awareness, and analytical logic to
industrial and acedernie circles. In industry, devise uniquely useful systems, devices, and
the designer systems manager is being re- processes. His function is usually not to
discovered as the man who can grasp and originate the basic scientific building blocks,
accomplish a comprehensive programme with but rather to utilize them so that the result
all of its involved and nebulous problems. is a useful creation.
Universities are realizing that design is, if Design must adhere to a plan which has
anything, more intellectually and academically objectives involving cost, performance, amount
demanding then many of the more specialized of effort required for attainment, probability of
a.reas of study. success, and even aesthetics. The fact that
The challenge imposed by this trend is design must traverse a closely evaluated path,
one of both quantity and quality. Obviously, starting from a well-considered if not urgent
more designers competent to fulfil the demands need statement to a functioning achievement,
of systems management are needed and, requires that it follow a methodology. By
hopefully, outstanding students can be methodology is not meant the tricks of the
attracted to design presented as a scientific trade such as drafting competence, or
endeavour which, in fact, it is. Both industry analytical ability, or, for that matter, a flair
and the universities are responsible for the for brainstorming. Methodology in design is
quality of design - industry for demanding of rather the framework for the design process
the designer performance of a high scientific within which a sequence of action steps can
standard, and the universities for conducting be based and from which check-points to
further research and offering challenging and evaluate progress can be established.
useful courses in design. An important influence in the renewed
attention to design is a general agreement on
The Scope of Design the part of Americans en gaged in the study,
Design is considered as the process of teaching, and practice of design that a unified
selectively applying the total spectrum of methodology of design does, in fact, exist.
science and technology to the attainment of Although different authors have presented
11
various detailed descriptions of the process, connecting steps and auxiliary stages, but
these descriptions are essentially all similar these are only ancillary to the basic methodology
to the so-called 'design method' summarized and can vary depending upon individual
here which, in turn; is an adaptation of the situations.
scientific method of MCCRORY (1963).
State of the Art and Recognition of Need
Structure of the Oesign Method The starting point of the design method is
Although the design method is similar to more comprehensive than that of the scientific
the scientific method, it has not been as method. Unlike fundamental scientific research,
carefully defined nor historically as well design is motivated by need rather than by
established. Nevertheless, the design method curiosity. Therefore, in addition to requiring
is as inherent to the design process as the knowledge of the state of the technical art,
scientific method is to scientific exploration. the design method requires recognition of a
need which warrants an investment of effort
Sc'ent1fic 1np..~t and funds.
(1) 8aSIC and
appl1ed Recognition of need can be considered the
research marketing input to the design method. Whether
(11) Broad
SC1ent1fic govemment or industrial needs are being
concepts
considered, the designer must realize that
much of the input required to define the need
is not technical, but rather socio-economic-
geopolitical. Therefore, the designer must
appreciate those key non-technical factors
Fatl~res
which are significant in defining whether the
Techn1cal 1
I
Rev1sed Powers
Marke!
results of his design work will fulfil a basic
acceptance 1 concept of des1gn
I
acceptance social, economic, or security need. With this
I
'--- appreciation, the designer is better qualified
to extrapolate current requirements and
creatively anticipate tomorrow's needs.
Powers of Given a need-oriented assignment, the
development
designer can encounter a serious point of
personal vulnerability. The purpose is not, as
he might prefer, to provide a result which is
~--------------~4
technically self-gratifying and elegant, but
Figure 2.1. Graphical representation of from which the only pay-off will be a technical
design method paper to his peers. The purpose is to produce
something which is truly useful in satisfying
Designers will do well to recognize its the defined need. Because designers are
structure so that the design method can be required to satisfy profit or security motives,
used consciously to clarify some of the costly the definition of need is critical to the design
'mysteries' of design. method and each succeeding stage must be
The design method, graphically described planned and judged on the basis of the need.
in Figure 2.1, is a closed loop, with Along with the recognition and definition
experiences gained at various stages during of the need, the design method requires an
the execution and completion of the process appraisal of the pertinent state. of the art.
providing the basis for subsequent steps. The State of the art includes materials capabilities,
method is compounded by multiple inter- phenomena understanding, and previous design
THE DESIGN METHOD 13
experience. However1 as important as previous If an idea does not satisfy a need 1 a

experience is 1 if designers are too limited in design concept as defined by the design
their conception of it 1 design progress can be method does not exist regardless of how
reduced to a series of small improvements. clever or novel the idea might be. Nor does a
The design method requires that the designer concept exist when a design which would
tap into the total spectrum of technology with satisfy a need requires a capability beyond the
the objective of obtaining the greatest design state of the art. The principal gain to be
advance consistent with the state of the art1 derived from the work done is feedback to the
wherever the art may exist. Experience which research laboratories.
is available in technical or product fields The designer can fulfil his synthesis
foreign to that of the designer can often function by an orderly procedure (McCRORY 1
s u g g es t the most advantageaus design WILKINSON and FRINK 1 1963). He first
approaches. analyses the need in considerable depth 1
But perhaps even more significant to perhaps allowing the need analysis to suggest
design advance is the input which can be a design concept. He then s p a t i a 11 y
obtained from research. As materials are visualizes systems which are advantageous
devised and phenomena quantified 1 new raw combina~ons. Or he might utilize a further
inputs to the design method are made available. technique which is not as broadly recognized:
New scientific concepts open up fresh areas this is to explore analytically the area of
for design exploitation. The design method 1 design interest 1 manipulating generalized
therefore 1 necessitates keeping open a direct mathematical expressions with the hope that
link to the resources of scientific and unique design approaches will be deri ved
ep.gineering research. The degree to which which would not be apparent from only spatial
the designer can intercept the latest scientific analysis.
information can determine the extent to which Attainment of the design concept stage of
he can make significant design advances. the design method means that a design approach
The Design Concept has been deri ved with the potential of satisfying
A design concept is created when through
1 the need as well as the potential of being
the designer's powers of synthesis 1a attained within the state of the art. Many
recognized need and technical capability as ideas may be rejected by the designer before
represented by the state of the art are matched. he arrives at one which qualifies as a design
When the designer can arrange technical art concept. On the other hand 1 he may finally
into useful combinations which form a system have available more than one design concept
satisfying a need he has a design concept.
1
showing attractive potential. At the design
Matehing can originate from either the concept stage 1 the concept need not be
need or the art. Given a defined need the 1
described completely. Rather 1 it may be
designer can search the art for the inputs expressed in terms of functional requirements
which can be synthesized to satisfy the need. or 'black boxes'. The key criterion is that
Conversely there are many concepts which
1
the concept has sufficient potential to justify
are originated largely on the basis of known further effort in designing the individual
art, and the concept sta,ge is attained by elements of the system.
finding a need which can be fulfilled. The
latter approach to design conception is the Design Feasibility
principal justification for the massive The design concept stage having been
engineering research being conducted in attained 1 the next step is to establish design
energy conversion 1 materials 1 and other generic feasibility. Feasibility is established by
fields of technology. determining whether all of the necessary
functions of the system can be worked out Production and Marketing

and whether, when the design is in detail When the designer is convinced that a
form, it still is attractive in tenns of the need feasible design is in band, he is ready to move
and the probability of successful attainment. on to the next stage in the design method.
To go from the design concept stage to the This is the development of a design which can
feasibility stage means to convert the design be produced successfully and marketed. In
as described in its functional fonn to specific actual practice, developmenttasks are extremely
elements. The steps (CRESS and CHEANEY, demanding in tenns of engineering skills and
1961) which may be used in this conversion they usually involve the major expenditure
process include: of funds and time.
Within the framework of the design method,
(1) Definition of the concept in terms of its the development step is still very much a
optimum combination of functions. design function (MCCRORY, 1%5). The
designer remains responsible for perfecting
(2) Expression of detail design requirements
the design in tenns of perform an c e,
in tenns of functional and/or performanc~
reliability, and cost. Although specialists in
specifications.
value analysis, tooling, reliability, and
(3) Design of specific elements to meet marketing may more prominently enter the
specifications (to be clone in accordance with picture, they cannot recover success if the
design method using specifications as need requirements of the earlier stages of the
statements). design method were not validly satisfied.
( 4) Trade-off analysis comparing design
Presuming skilful engineering and marketing,
alternatives and, if required, revision of the development steps, although costly, are
not highly risky. The mistakes which lead to
specifications.
disastraus failures are more likely to occur
(5) Critical experimentation to test specific at the stages when decisions regarding need,
questionable aspects of the design concept. concept selection, and feasibility are made.
(6) Operation of experimental prototype to The loop of the design method closes
confinn adequate functioning of total system when the design is judged a technical and
or sub-systems. marketing success. This experience in market
acceptance extends the understanding of need
Frequently the design concept fails to reach and invariably leads to the identification of
the stage of feasibility because the technical new areas of need. The technical successes
problems cannot be solved successfully or and failures expand and temper the state of
because the concept does not fulfil its technical art and are inserted into other
apparent potential of being attractive in tenns design programmes.
of the need. The probability of this happening
should be reduced if designers conscientiously Variance in the Design Method
go carefully through the prior steps of the An idealized description of the design
design method. When failure does occur, it is process has been presented. If individual
necessary to retum to the concept stage and design programmes were discrete entities,
revise the concept in the light of experience the design method as described here in skeletal
gained. However, if failure is complete, the fonn would be accurate. In actual fact, design
most that can be salvaged is the failure efforts seem to blend into one another to the
experience which can be interjected into the extent that a complete progression of the
state of the art as a guide to subsequent design method is difficult to recognize. As
design programmes. actually practised, design progress involves
a pattern of superimposed programmes, each Technical Planning A Means of Defining

subject to the requirements of the design Design Opportun ity
method. This pattern may include: lf design is to be technical effort intended
to serve a purpose such as devising a profitable
(1) Design programmes in which the outcome product, then a priori judgment must be made
is the selection of a means of approaching a regarding the problems worth working on and
broad national or industry need. This type of the opportunities which offer the most promise.
programme is being applied to the transportation As defined by the design method, need
problems of the American east coast megapolis. definition is the critical first stage of a
The application of the design method here successful design programme. Be c aus e
will involve overall systems concepts whose technology has become so costly and failures
feasibility will be established by computer so devastating, considerable effort can be
analysis prior to operation of prototype justified in selecting the design objective
hardware. The outcome will be need Statements in which an investment will be made and the
for other design programmes. plan that will be followed. One approach is
termed 'technical planning'.
(2) Numerous design programmes which
Technical planning is a process of
originate ft:om the same need recognition. Some deriving possible courses of design activity
of these programmes will be parallel programmes by recognizing the technical and marketing
conducted by different design groups searching
environment toward which the design will be
for the same end result. Theoretically,
directed and then systematically evaluating
application of the design method should result
alternatives. This type of endeavour should
in the same outcome for each programme. But
be a continuing programme conducted co-
the personalities and backgrounds of the
operatively by the design group and other
individual design groups are so important to the
decision-making circles within a company.
functioning of the design method that results
However, technical and marketing inputs must
which are different in both approach and
quality are inevitable. be brought together within a framework of
corporate objectives originating from top
(3) New programmes which are initiated management levels. Thus, really effective
before an original design programme is technical planning involves corporate policy
completed and which have as their objectives makers, sales and marketing personnel,
the improvement upon the results of the first research staff members, and designers.
programme. The demands of progress often Technical planning cannot succeed without
cannot wait for the sequential completion of unique contributions from the designer -
related design programmes. contributions of which most designers are not
aware or which they do not know how to make.
( 4) Auxiliary programmes on sub-systems
and components which are part of an original Figure 2.2 is a PERT* type diagram of a
overall programme. The design concept having rather comprehensive technical planning
most potential may require an element which programme for a company which has the
is identified as beyond the state of the design corporate objective of being on the product
art. 8ut the attractiveness of the concept may forefront in its field of marketing endeavour.
warrant the risk of initiating a separate design This field is presumed to be one which is
programme to expand the state of the art. The technically demanding (such as prime movers,
designing of individual functional members electronics, or petrochemicals) although
becomes a number of sub-programmes calling * PERT = 'Programme Evaluation Review Tech-
for use of the design method. niques' developed for the US Navy.
technical planning of a somewhat more limited these inputs provide economic criteria with
scope is being used effectively in marketing respect to the technical alternatives. The
areas which, on the surface, would seem more research oriented inputs from the right indicate
mundane (such as home appliances and heat what can be clone on the basis of the technical
exchangers). state of the art.
Figure 2.2 shows a series of action steps The designer proceeds through a series
or tasks in a technical planning programme. of tasks starting with task (1) - ( 4) du ring
which he devises one or more 'ideal product
concepts'. These ideal concepts are the best
that can be imagined in terms of product
attractiveness. Restraints re gar di n g
attainability or internal workings of the ideal
concept should not be imposed at this point,
lmagtne id•at
product concepts although the designer cannot, of course, be
2 Evaluate strat~y for
applytng altern•hve
S~•ct technology
~rttn.nt to tdNI
given license to violate basic scientific laws.
tdeal concepts CQ"'ICI'pls
Usually the ideal concept is described in
For.cut M:tvances
tn technology terms of product characteristics (the efficiency,
Conctuct market AN~Iysts ldenhfy gaps ., technotogy
size, weight, cost, configuration, durability,
llmtltng rultzatton of
and evaluahon of
Konomtc requtl'e'f1Wnts tdeal conc~ts etc.) which would be iliought ideal for the
DevtSI' real product
concepts ~ upon type of product being considered. If the
present t•chnotogy
designer does not make facetious specifications
Progra"..."_ hme
sc•t• for (for instance efficiency of unity, no weight,
5eiKied strategy
Dehn• r•5eNch that no cost, indefinite life) he can in expressing
can basiully l'f"'hance
tull.l"• altalnment an ideal concept describe a product which
-~~"!
o.t•,... engtMt"nng to would have optimum attractiveness but be a
support pres.m:
IT\I.rk.. posthon
realistic challenge to innovative design.
Oeftne spectftc "..w
produc:t or proces.n The designer can also describe in similar
f
lo be work.ci to-w&rds
terms the best commercial achievement in this

7 product area. Then the gap between the
Support .nginNring New product destgn
description of the best existing product and
specifications of the ideal concept represents
Figure 2.2. Task diagram for technical the challenge to innovation and product
planning programme improvement. If this gap is small, either
(i) the ideal concept does not suitably represent
The starting point is the stated objective: product opportunity, or (ii) opportunity does
i.e. this company intends to attain and maintain not, in fact, exist. If it can be shown that
a pre-eminent position in the vehicle prime opportunity does not exist, it may be necessary
mover field and therefore wishes to plan a to recommend to corporate management that
technical investment that will achieve this objectives be revised.
objective. Having described the ideal concept and
The task tree of Figure 2.2 has three having decided where opportunities seem to
principal branches: the left branch is principally lie, the designer proceeds to task (4) - (5).
marketing; the right is research. The core or Now he must devise real design concepts.
central branch, however, is the work of For this he needs realistic marketing targets
designers. The marketing tasks provide inputs from task (2) - ( 4) and a thorough reading of
which say what can be sold in the market the technology from task Cn - (4). He must
place and how it should be sold; in addition, then formulate real product concepts using
the synthesis techniques of the design method. is indicated by the index number assigned to
It is important for the designer to devise each characteristic. In this hypothetical case
several alternative concepts which can be the evaluator has established that the most
subjected to subsequent evaluation. Some of important characteristic of the product is
these concepts may represent relatively small small size; in fact, the value index numbers
advances but they may also involve relatively show that he regarded this to be four times
low risk. Other alternatives may offer the more important than having optimum reliability,
potential of major steps toward the ideal and eight times more important than having a
concept but involve greater risk. Such continuously varying control system.
alternatives are required if the technical Two new product concepts A and B are
planning is to result in the identification of being compared against a commercially
the preferred business opportunity.
The next task for the designer and his ChoractHistic Volu. Sysl•m P,..s.nl
aw::lilabl• Charoct•ri5ttc pomt scor~s
cohorts in technical planning is to evaluate

Throshold ()ptimun mthx
r-r~ srst•m • 10 ,. '" -,. "' . . 0
design feasibility and to select the concept

Xin.3
Size
Pin~
4-0 •
i2 2i
7 6
24
,'-,.' ,'\]
which will offer the best business opportunity.

//_ '//1
In order to undertake task (5) - (8), the Cool
3 0
_!_
15
_!_
II
9
27 ~
X.C/ton Pl:/ton
designer and market analyst must work very
closely to estimate the saleability, market
strength, and return on investment potentials
of the alternative concepts. The designer
must also substantiate the feasibility of the
X Ion
X"/.
W.tght
P ton
R•liability
P"/.
Control
1-5
lO
10
15
5
6
9
T
7
10~ 6
T
4
5
--~
·"""'~
Optimum s.core-
PtHenl product -
Concept8
c:::::J
C2ZZI
•
~
possible concepts to the extent that he has modulat.on 0-5
10 2 Conc•pt A IS::S3:SI
Continuou•ly 5 4 1
reasonable confidence that further investment Off-On YM'I&ble
10·0 12 &7-5 63
is technically justified.
There must, of course, be a considerable Figure 2.3. Product effectiveness scoring
amount of subjective judgment in selecting matrix
preferred business opportunities. But the value
of comparisons can be greatly enhanced by available system and the optimum characteristics
proceeding through a deliberate rating of the of the ideal concept. The effecti veness of each
potential of each possible concept with respect one is judged by assigning a number from one
to the product characteristics used to describe to ten for each characteristic. An effectiveness
the ideal concept. number of one means that the product performs
The comparison procedure can take the at the threshold Ievel, whereas ten means the
form of scoring for each characteristic and performance is optimum or ideal. In this type
arranging the scores in the form of a matrix. of display it has been found convenient to
The matrix would be so designed that the note the threshold and the optimum values
overall or lumped performance of each candidate for each characteristic in the block with
concept would be readily calculable, and also the characteristic name. The product of the
so that the respecti ve advantages and dis- effectiveness number and the value index is
advantages of the candidate systems can be the score for each system with respect to each
perceived. Figure 2.3 is a sample of the kind characteristic. The assigned effectiveness
of display which has been particularly number and consequent score are shown in each
effective. characteristic block for the two new concepts.
In this figure, various characteristics of Note that the value index quantities have been
a hypothetical product are shown ranked by so chosen that the lumped score for the optimum
order of importance. The degree of importance overall system will be 100 points. In this case,
both the new concepts outpointed the commercial development because of its potential benefit
system, with concept A having the best overall to both the education of designers and the
advantage. execution of more demanding design work. In
Considerable guidance would be obtained education, the treatment of design as a
by the designer from a study of the scores and discipline makes it more attractive to students
the bar graph display at the right of Figure 2.3. and provides the teaching faculty with tangible
For example, it is clear that the greatest subject material. In industry, the design
opportunity for technical advancement over the methods being developed and gradually adopted
available commercial product lies in conceiving provide the tools with which talented designers
a system of reduced size since the commercial can extend their efforts into new areas and
system falls well short of optimum in this which will give designers more confidence in
respect. Both of the newly conceived systems the validity of their work. Although new
have some measure of advantage here, although developments in design methodology are still
it is not particularly dramatic. On the other far from fully adopted or always effective,
hand, both new concepts evidently fall short there 4oes seem to be increased interest
of the commercial system, and well short of stimulated by demonstrations that creative
optimum, in the important category of cost. design can be approached by orderly and even
With respect to the other characteristics, the predictable methods.
new concepts are highly effective but derive However, regardless of the sophistication
little advantage thereby because of the relative of methods which are devised to enhance the
unimportance of these characteristics. Thus, effectiveness of the design function, they must
the conceptual objective becomes quite clear: be recognized for what they are: only tools to
the designer must envision a system which will be used by the designer. The key to brilliant
retain the size advantages already in hand design remains the designer hirnself with all
and which will cost less. He has some his intangible design capabilities such as
conceptual leeway in striving to do this since intuition, judgment, determination, courage,
he may be able to devise a way to trade off spatial vision, and imagination. Some engineers
control modulation for cost. have these qualities and can therefore become
At the completion of task (5) - (8), there natural designers; others do not and probably
should be the intelligence necessary to make cannot be trained to serve other than ancillary
decisions regarding new product design roles to design. There can be real concern
activities and substantial information available that natural designers are not being recognized
to the designer to guide him to success. From and developed by the 'apprenticeship' approach
this comprehensive technical p 1 an n in g which trained many of the present vanishing
programme would also be derived plans for breed of designers. In fact, with the premium
providing engineering resources needed to placed on highly science-oriented graduate
support the company's present market position level education, there has been criticism in the
and for selection of research projects to USA that design potential is being snuffed out.
strengthen the long-range corporate position It has been Battelle's experience, however,
in its market area. that recent graduates of America's engineering
colleges are unusually capable of doing the
most demanding design work, and have a flair
Conclusion for the original and a capacity for analytical
As indicated by the preceding paragraphs, rigour that is refreshing. With this new
there is in the USA a distinct trend toward the generation of designers, the effort being
use of method and discipline in design. This is exerted in the derivation of advanced design
considered by the author to be a favourable methods is not being wasted.
Chapter 3
DEFINITIONS AND METHODOLOGIES
W. E. Eder
lntroduction obvious: firstly of overlooking possible

Design in its fullest meaning is the top connections, analogies, etc., and secondly of
role of the engineer (HRONES, 1960). Before overlooking or ignoring the fellow human being.
any work of human skill can be produced it
must be imagined. In essence, it is this human Definitions
power of imagining something that did not This book deals with the design method
exist before that is termed 'design' (HARVEY, in engineering. Engineering, design, and a host
1950). This chapter is therefore concemed of other expressions, must therefore be defined
with an outline of the knowledge, skill and in order to clarify subsequent discussions.
imagination required by the designing engineer; The following two definitions will be adopted:
the progress of a new piece of equipment (in
the abstract) is traced from the original idea (1) Engineering design is the use of
up to the point of production; the probable scientific principles, technical information and
approaches adopted by designers in solving imagination in the definition of a mechanical
engineering problems are indicated, and the structure, machine or system to perform
prespecified functions with the maximum.
use of various 'tricks of the trade' to help in
economy and efficiency. The designer' s
this process; finally where these tricks are
responsibility covers the whole process from
useful in the contexl of a small selection of
conception to the issue of detailed instructions
technologies is shown.
for production and his interest continues
The basis of much human activity lies in
throughout the designed life of the product in
a full and broad education, consisting of a
service (FEIL DEN, 1963).
(2) The engineer is in fact the means by
(X ).: .~ ~ ~"'
which people are able to enjoy the fruits of
Science Technology
science or invention, whether in building new
projects, or in maintaining and keeping up to
date what is already in existence. Each
generation must leam that technical knowledge
without a sense of mission and responsibility
Art Craft is wasted.
~ The wealth of a nation, and all that
implies, depends in fact upon the efficient
Figure 3.1. Ideal of education organization of its resources both natural and
industrial as well as human. In this
balanced view of the sciences, technologies, organization the engineer bears the chief
the arts, and the crafts (Figure 3.1). Two responsibility. Any large engineering project
dangers of a lacking balance in education are must depend for its planning and successful
2 19
completion on the closest possible co-operation Manufacturing Technology

and integration of many specialists (DUKE OF All engineering artefacts must be
EDINBURGH, 1961). manufactured, and invariably by persons not
connected with design work. This means that
In order to fulfil his d uties adequately, the the engineer must have a reasonable knowledge
designing engineer needs certain types of of present-day manufacturing technology
knowledge and skill, which may be summarized (although the designer can usually call on the
under the headings: theory, manufacturing p rod u c t i o n engineer for guidance and
technology, working constraints and design
information). This must comprise a knowledge
method (see Figure 3.2). All engineers must
of methods of forming and of joining metals,
have a working knowledge of these subjects
plastics, ceramics, glasses, semiconductors,
(although the subjects are by no means in
etc. and the properties of these materials du ring
watertight compartments). The composition of
manufacture and service use; the methods of
their individual fund of knowledge and skill
manufacture available or in development, both
depends largely on the nature of their work,
universally and within the engineer's own
.-- organization; and their economic use.
CM')" :
/ ........ ..........
/ ~
' / "- '\ Desogn
.
/
I "- /\ _ ~ yreahzatoon Design Method
.~;~
This is intended to imply that general
approach to a problern which is likely to lead
to a successful solution. It must be backed by
a certain amount of imagination and intuition,
Design method \ /Working constraints
" , ___..... /
' ~ / and a more or less systemalle investigation of
' ......... .....__/ / the problern including the use of helpful
techniques or 'tricks of the trade'. The general
Figure 3.2. Designer's knowledge pattem in engineering design consists of
preparation (gathering information, etc. ),
whether it be production, management, design incubation, ver i f i c a t i o n and finally
or even research. It also depends on the communication. Recent investigations show a
industry and the type and size of organization marked similarity in design method between
in which they work. the pure artistic (such as sculpture), the
functional artistic (furniture, industrial design,
Theory etc.), and the engineering fields. The differences
In this context, theory consists of the are found mainly in the use of theory, the
fund of knowledge of mathematics and its application of manufacturing technology and the
application to kinematics, dynamics and other working constraints which render design
statics of mechanisms, fluids (compressible progressively more difficult from pure arts to
and non-compressible), heat, light, electricity, engineering.
strengths of materials, statistics, etc. It must
in c 1 u d e a thorough knowledge of the Working Constraints
assumptions on which the t h e o r e t i c a 1 Although these could well include theory
investigations are based (usually stark and manufacturing technology they are derived
simplifications in order to make the phenomena from economics (cost and value), aesthetics,
amenable to mathematical treatment) and the production organization, available space
therefore also their limitations and deviations and time, and the type of goods in demand
from the actual phenomena described by their (capital, consumer durable, con su m er
mathematical formulations. expendable, etc.).
Type of Work including the human being as a funclional

A design engineer may wtlrk as an part of the artefact.
individual, as the leader of a design team, or A crankshaft, a resistor, or a filter packing
as a member of the team (together with other could be classed as a component (the resistor,
design engineers), and this will have some as an electrical component, is probably an
influence on the composilion of the above assembly of a nurober of mechanical components
factors, in parlicular the design method. Their including an insulator, wire and terminals).
balance will also be affected by the type of An assembly would be a heat exchanger, a
work on which the design engineer is employed, slider crank mechanism, or a valve voltmeter.
whether it is components, assemblies, system A hydraulic pump, or an electric motor could
elements, or complete systems. For the purpose well be a system element, working as part of a
of this chapter these artefacts may be defined machine-tool (a system ·closed by a feedback
as follows (these definitions cannot possibly loop including a human operator)or a mechanical
be rigid; there is a great amount of overlap analogue computer (including various mechanical
between these fields and vast differences feedback loops). Transport within a country
between industries). Components are individual can conveniently be regarded as a system,
pieces of material, usually with the properties with an individual motor car or a piece of
of 'size', 'mass' and 'strength' only. motorway as an element in it,and a mathemalical
Assernblies are colleclions of components, descriplion requiring the use of stalislics to
with properlies derived from the components, give it general validity.
but with additional properlies stemming from
the arrangement of components, such as 'speed',
The Design Process
'power', 'ratio'. System elements are usually
Following the definition of engineering
assernblies of components, but they are destined
design given above, a series of steps may be
to function as a part of a system, not as an
stated that constitute the design process. In
independent piece of equipment. Their properties
any given design sequence for a particular
are usually described by 'transfer funclions'.
piece of equipment some of these steps may be
Systems are collections of strch elements,
interconnected to achieve the desired properties, omitted or transposed in order, depending on
but working in an environment that can cause the general circumstances and the type of
disturbances (noise) in the system by goods to be designed, but this statement should
cover most eventualilies.
influencing elements or their Connections.
Systems may be classed as flow systems (the
flowing medium may be power, informalion, Sponsor's Requirements
chemicals, etc.) with one set of inputs and The design process invariably starts from
one set of outputs, the transfer taking place in a statement in broad terms of the sponsor's
a set of common elements; or as associative requirements. This may come directly from a
systems, where two or more separate sets of customer, as in the case of most capital
elements working on separate inputs and equipment; from directors or the s a 1 es
producing separate outputs are required for department in the consumer durable field; from
the total funclion of the equipment (for instance an inventor who recognizes an unfulfilled need
a motor car with its power transmission, its (either outside or within the organizalion,
electrical system, its braking system, etc.). this may be a designer, a production engineer,
The concept of the system implies a more a machine operator, an inspector, or even an
mathemalical approach and a consideration of innocent bystander); or from a customer
a full loop of responses but it has been f e e d back of deficiencies in, or new
broadened to cover most engineering artefacts, requirements arising from, existing equipment.
Analysis of Customer's Requirements Problem-solving

A full analysis of the customer's The next step in the design sequence, is
requirements is necessary to determine the a cyclic process consisting of an analysis of
true needs*, the desired function, and the the immediate problem, the suggestion of
conditions of environment, time, operation possible solutions (by searching for existing
modes, etc., under which the required equipment solutions or by true innovation), the evaluation
must function. This will also suggest where of these solutions (by experience, prior art,
further information is required, either from the mathematics or experiment), modification
sponsor, from existing literature, or from where necessary, and selection of the most
research ('pure' or 'applied' experimental work). promising of them. This process must conclude
with a complete specification of the hardware,
Sifting usually in the form of drawings with due
All this information must be taken through consideration for its spatial arrangement, the
a sifting process, to determine the essential proposed economic methods of manufacture,
functions, the disturbing or restricting factors, and where applicable also the aesthetics of
and the less important factors that need to be the solution.
kept in mind although they do not determine
Production Organization
the design itself.
From here on the need for a production
organization becomes evident. This must
Feasibility Study
include management functions, a marketing
A feasibility study is frequently useful
branch and an economic guidance division,
at this stage. Feasibility can never be proved,
as well as the actual production, inspection,
but violations of the basic laws of physics
storage and testing facility. Usually it will
can prove that a solution to the problern as it
take the form of a commercial-cum-industrial
stands is not feasible. The designer is
concemed here with feasibility - testing the undertaking, but other forms are possible
sponsor's problern - not with the testing of especially in the civil service, military and
any suggested solution to it. The result of this educational fields. The production organization
study may well be a set of evaluation criteria, is used by the designer to obtain the specified
against which the performance of a suggested hardware, initially as a prototype for the whole
solu tion can be tested. It may also result in or a small part of the system, in order to test
(prove its function) and develop it (improve its
necessary alterations to the sponsor's
specification to render his problern rnore function atleast up to the required performance).
amenable to a successful solution. The production models must also run through
a series of tests, which are usually used for
Sponsor's Approval quality control purposes.
The sequence up to this point should Marketing
result in a restatement of the sponsor's The marketing branch now takes over, the
requirements, in new wording, considering all designer only receiving notice of customer
possible influences, and showing how well
complaints and service faults, which help to
and completely his requirements have been
increase his experience.
understood. The sponsor's approval should be
obtained for this statement, and for any All this may sound rather idealistic, but
subsequent changes that may prove necessary. most of these steps can be found in any
• This is not necessarUy what the sponsor says he engineering design sequence. The steps will
requires, but the reason why he says that he requires not be of equal length or of equal difficulty:
this. some may be taken over by other functions
within the industrial organization, some will process on a number of engineering problems;
be performed by indi viduals, and others by but progress towards this experience can be
teams. Frequently a design sequence will not fostered by guidance through heuristics (aids
be a simple run-through, it will exhibit to problem-so! ving). This latter is the true
backward overlap, namely a return to previous purpose of design methodologies in the teaching
stages as new information becomes available. of engineering. Two recent conferences have
A 'forward overlap' should in general be discussed various aspects of this: the
avoided: it usually involves jumping to proceedings edited by jONES and THORNLEY
conclusions, which may prove expensive. (1963) and by BOOKER (1964). The imagination
needed to bring the design process to a
successful and economic conclusion can be
Design Methodology - General Remarks fostered by appropriate personal contact (the
In recent years some attempts have been master-apprentice relationship). As yet there
made to recognize and rationalize the design are no hard-and-fast guiding principles about
approach, and set up a universal and systematic
'best ways', but investigations on creativity
design method. All the methodologies so far
are in progress in the USA and Great Britain
proposed fit into the design sequence outlined
among others. One result was a symposium held
above, but each uses different techniques to
at Birmingham in July, 1964.
reach one or more solutions to the sponsor's
Experience of design realization is defined
problem. Some methodologies cover the whole
here as the experience needed to specify
of the design sequence, others concentrate on
hardware in such a way that it can be produced
important parts of it and may be fitted into
economically in the existing (or slightly
other methodologies to improve their probability
extended) production organization. This can
of aiding the solution of engineering problems.
only be obtained by work within an industrial
The author has made a study of method-
organization, by making proposals for hardware
ologies for the purpose of a coursein Mechanical
and discussing these with the production and
Engineering Design (EDER, 1964; EDER and
economic engineering staff. It requires an up-
GosLING, 1965). The summaries presented
to-date knowledge of production technology
here are his opinions on the uses of the various
techniques and methodologies, with all due and of the properties of existing materials,
respects and thanks to the originators. and a flexible mind capable of leaming from
Although the borders are by no means errors (including those committed by other
precise, the six methodologies discussed in people).
the following section are (1) experience, Formal design methodologies (3) - (6)
(2) modification and running redesign, act as a framework for guiding the designer's
(3) check-lists, ( 4) design trees, (5) the thoughts. They are no substitute for creative
fully systematic method, and (6) the system thinking, but they can help to spark off the
search methods. intuitive processes. They can also free the
Some authorities claim that design can mind of the designer to some extent by
only be effectively performed 'by experience'. relieving him of the constant worry of
This begs the question of which type of remembering all that has occurred before,
experience is necessary, since there seem to especially the reasons for certain decisions.
be two basic types, the experience of the In addition, it becomes easier to follow two or
design approach and the experience of design more possible solutions until the superiority
realization. The fund of knowledge required of one becomes obvious when they are compared
for each type is outlined in Figure 3.2. by objective criteria of assessment. These
Experience of design approach can only methodologies also appear to help the designer
be obtained by working through a design by allowing him to find more solutions to
simple sub-problems, and then eliminate and functional piece of equipment, but to one
unsuitable combinations, rather than think containing many redundant parts that were
all the time about the main problern to the included to treat a symptom when the real
detriment of detail. cause was overlooked.
Check-lists
Design Methodologies
It is possible toset up alist of influencing
Experience
factors for each step of the design process
This involves 'throwing the future designer
outlined earlier. These lists could conceivably
in at the deep end'. The budding designer
cover the requirements of all industries, and
either leams his own methodology, or sinks
the individual designer can select the ones
in the process. A good master (one who is not relevant to his problem. Such an approach helps
only a good designer, but also a good practical the designer to remernher and complete every
psychologist) can assist in acquiring this stage of thedesign process, watch his progress,
methodology. It is probably true that eventually and put his thoughts and decisions on paperf
every designer who works alone will fall into in the opening stages in words, later as
this c a t e go ry, he will use the other sketches and layout drawings supported by
methodologies (or only their heuristic words and calculations. Great care is required
techniques) subconsciously. The mainstay of to make the reasons for any decision clear,
this design methodology is 'cut-and-try', not only to the designer hirnself (for reference
empiricism, trial and error. It stays in the at a later date) but also to any other person
mind: only the occasional sketches, the who may have cause to refer to them, such as
layouts, the calculations, etc., show that development engineers.
there is work in progress. The reasons behind A !arge number of such check-lists are
a design decision rarely reach paper, and are presented by MA TOUSEK (1963) and a few
therefore di(ficult to follow; this may be a others are published in Eder and Gosling.
severe disadvantage in cases of absence of Matousek and AL GER and HA YS (1964) give
the original designer, or of the discovery of a similar techniques for evaluating possible
major error in the principles of the design. solu tions, based on a suitable check-list.
Each solution is graded under the heading of
Modification a number of factors on the check-list into, say,
Improvements in an existing piece of five categories of quality or compliance:
apparatus are frequently performed in this way. number one as the lowest, five as the best
It involves studying reports on the performance possible. Each factor is given a weighting
of the existing apparatus, particularly of its to show its relative importance among all
short-comings, and altering the design to other factors (e.g. whole numbers 1 - 10).
avoid these faults. If this is done with an The grading number G is multiplied by the
inadequate study of the original design, such weighting number W, and the resulting products
modifications can aggravate other faults. are added for each solution. The solution with
Long-term improvements in mass-produced the highest rating may look the most promising,
products are usually attributed to this but the next one or two should not be ignored.
methodology, but innovations (radical departure An example of such an evaluation is shown in
from existing principles of operation) rarely Table 3.1
result from its use. An attitude of 'design-and-
modify' can prove very expensive in the early Design Trees
development of a new product, especially if A design problern may be broken down into
the design has not been fully considered. smaller problems, each of which may again be
Such running redesign rarely leads to a pleasing broken down, until finally each sub-problern
THE DE~GN METHOD 25
Table 3.1
General purpose Solution A Solution B Solution C Solution D

domestic suction Cyllnder with Globe model 'Slipper' model
clesning machine Hand-held unit paper dust-bag I with cyclone with handle
Factor Weight Grade GxW Grade GxW Grade GxW Grade GxW
Handling (fioor) 10 3 30 4 40 4 40 5 so
Emptying 8 2 16 4 32 3 24 3 24
Handling (movement) 5 5 25 5 25 4 20 4 20
I
Storage slze 4 5 20 4 16 3 12 4 16
Versatillty 5 1 5 5 25 4 ' 20 4 20
:
Suction power 4 2 8 4 16 5 !
20 5 20
Partide slze range 1 2 2 4 4 4 ! 4 4 I 3
Fitting accessories 2 2 4 5 10 5 10 4 8
Handling (accessories) 8 1 8 5 40 5 40 4 36
Appearence 2 3 6 4 8 4 8 3 6
Sum 245 124 216 198 203

max.
i
has a simple solution. Combination of such Only few situations are amenable to this
solutions should yield a nurober of solutions treatment, because the sub-problems are
to the design problem. This form of approach difficult to foresee unless the problern has
is illustrated in Figure 3•3 and is amenable to well-defined rules like a game.
search techniques using an electronic digital The alternative suggested by MARPLES
(1961) is to start from a statement of the main
problem, stiggest principles along which this
problern may be solved, and find the sub-
problems that must be solved before a solution
to the main problern is possible. This involves
a cyclic process of analysis of the problem,
theorizing solutions, delineating these
solutions, and modifying them (which again
involves analysis, theorizing, delineating,
etc.); this constitutes the ATDM cycle proposed
Figure 3.3. Game tree by WALLACE (1952). The design sequence
could weil be plotted as in Figure 3.4 where a
computer, as suggested by NEWELL and vertical line denotes a problem, a slanting
SIMON (1964). Such a programme would be a line denotes a solution. Any problern or solution
vast undertaking yielding many solutions which may be picked out by means of the series of
arenot feasible, but some heuristic letters and numbers required to reach it from
instructions may be entered to cut down the the origin (such as a2blb), and this permits
amount of work, and therefore time, spent. easy cross-referencing of notes, Iayouts, etc.
Ideally, a short statement of each problern as possible. The methodology is most useful
and solution should also appear on this for team work on large problern complexes,
gradually growing design tree, with estimates although many of the techniques may be used
of the difficulty of solving a problem, and by individuals to great advantage, especially
availability of solutions. A good knowledge in connection with the formal procedures
of precedent solutions (BOOKER, 196~) to suggested by MA TCHETT (1963) and
similar problems or sub-problems is very illustrated later in this book. It is also usefully
supplemented by heuristic questions and hints
listed by GREGORY (1963). lt is claimed that
the use of this method can lead to design
innovation on scanty evidence andin situations
where experience is lacking, such as in new
technologies, by bringing the design process
consciously and (almost) completely outside
the mind.
Termination of solution sequence Cno sub-problems) In the analysis (or preparation) stage the
designer (or the team) sets up a Iist of factors:
Figure 3.4. Design tree statements that have relevance to the problern
in hand or to its solution. This should include
helpful, especially if these solutions appeared considerations of the basic functions to be
in a different technology. Of course, all fulfilled, the capability and limitations of the
precedent solutions to the main problem, production organization, the customer, the user,
for example the competitor's model, should the environment and laws and standards; the
also appear on the design tree. A solution to life of the artefact (from the sponsor's problem,
the main problern exists when all problems through design and manufacture, via transport,
thrown up by any one alternative solution maintenance, storage, installation and use,
have been solved (see hold lines in the to the scrap heap); the probable criterion of
diagram), and these solutions do not throw up 'life ended' (end of useful life by failure in
any further problems. Work on any one branch service, by completion of the required duty,
of the design tree may, of course, be interrupted or by obsolescence), etc. It is important to
if other solutions are preferred. This technique ensure a complete ban on criticism, to include
can result in novel solutions if it is all ideas even if they sound impracticable, and
intelligently used by a lone designer, by a to allow duplication. Useful aids to this step
designer in charge of a team of specialists, are 'brainstorming' which includes looking
or by a small team of designers. In the last deeper than the external features and for
case the documentation must be good to avoid common features, and the use of the work study
unnecessary duplication of work. questions (what, why, when, where, who or
which, and how; asking after the needs,
Fully Systematic Method justification, time, location, available . or
As already stated, a certain amount of employed resources, and method). Each factor
backward overlap during the design process is should be numbered in order of appearance
inevitable. This methodology, reported by and should consist of a single statement.
J ones and Thornley attempts to cut overlap Composite statements must be split up before
to a bare minimum by separating each stage of the next step.
the design process and using various data To reduce this information to workable
processing techniques as described below to proportions, these factors must be classified
provide as complete a coverage of each stage into a set of categories. Any factor that
contains a suggestion for the solution of a be listed separately, and checked to see if
sub-problern (a partial solution) should start there are any obvious omissions.
a new category. The primary needs of the At this point it may be useful to place the
sponsor (if these are not solved, the fulfilment categories in an order of importance, using a
of other needs is pointless) should, of course, weighting chart (see Figure 3.6). If the category
suggest the first category, or categories. in the horizontal line is considered more
Sponsor•s problern General purpose domestic sucl1on c ine
I I! ![[ IV oesign
No Factor content Power Dust
Housmg Nozzles Ac ~IIC Fll~
dnve extract,on
1 Easy to move around X 0
2 Stable X 0
3 Positivedust coUect10n X 0
4 Use domest JC power X 0
5 Gel 1nto small gaps X 0
6 Darnage del•cate f abnc ? X 0
7 Dusl catch1rg 1n m/c X 0
8 Suck up !arge b•ts X

9 Reach ce•l•ng
10 M/c easy to clean X
11 Electncal safety X 0
12 Clean wood mould1ngs X 0

~
~·
Figure 3.5. Factor classification chart
Category I u III IV V VI j Vll VIII

~
Hous•ng I ~ X \' X X
I X 4
Nozzles !1 I~ X X I X X 5
Power drive 111 X i X ~ X x. I X i X X 1
I
Dust extract•on IV • X
i X ~ X i X X 3
Aceesseries V I :~t X ! 7
Tube extension VI i I ""'-.J 8
Human VII X X X X ~ X 2
Aesthelic VIII X X I I~ 6
Sum 3 3 0 2 6 7 l 2 5
X = Honzontal category more 1mportant than vertical category
Figure 3.6. Category weighting chart
Any suggestions of solutions should at this important than the one in the vertical column,
stage be filed for future reference in a design place a cross in that space, and a reference
Eile. Each factor should be placed in one dot in the 'converse' space. The category with
category only (if necessary, change the ti tle the lowest sum (category III with no crosses)
and scope of the categories). A form as shown is the most important, the placing of categories
in Figure 3.5 may be used. Having found with equal sums is decided by the position
suitable categories, the factors in each should of the crosses (e.g. VII more important than IV).
Each category will inevitably influence meter to be 1· 6 m from ground level), and
some others, and this should be investigated include permissible limits of performance.
with an interaction matrix (Figure .3. 7). In Complete separation of problern and solution
each case the nature or lack of the interaction can only be achieved if the p-specs contain no
reference to shape, materials or design. The
Catt>gory
completed list of p-specs is a processed form
of the sponsor's needs, including all other
Nozzt~
data that may be required for solving problems,
Pow.r drlvP
and criteria for assessing the solution. The
Du~t Pxtraction
sponsor should approve (or modify) this
Ace f' SSO r IPS
specification before the next stage of the
proceedings, to ensure that the right problern
is being solved and an acceptable interpretation
of the sponsor's requirements has been reached.
Figure 3.7. Interactionmatrix Conception (incubation) of solutions
starts by solving each p-spec independently,
between two categories should be questioned in words and sketches finding as many
(another use for the work study routine), and solutions as possible and listing them in a
the results noted. This process can re·veal 'morphological chart' after NORRIS (1963)
further factors that were overlooked in previous (Figure 3.9). This is aided by such devices as
steps, especially if the lists of factors in each listed by Gregory (1963) and JONES (1963):
category are studied carefully. This matrix playing with words and concepts, studying
may be transformed into a straightened net precedent solutions, looking up the design
by the process shown in Figure 3.8 (each line file (started during factor classification), etc.
denotes an interaction). Weak interactions The designer should note whether any p-spec
between groups of categories may suggest or partial solution requires r es e a r c h,
splitting the problern into two or more parts. consultation of a specialist, a component test
programme, or particular attention in any other
w ay. Partial solutions should contain
statements of limits within which the overall
solution would be acceptable.
The partial solutions to each p-spec must
be tested for compatibility with one another,
(the interaction matrix may be used, asking
the work study questions). By deciding on an
order of importance, plotting on a tree such
( a)
as Figure 3.3 with the most important p-spec
first and the branches 'growing' only where
partial solutions are found to be compatible,
Figure 3.8. Interaction nets and from this producing a set of dimensional
layout drawings, a nurober of alternative
The interactions noted above must now be solutions to the main problern should be found.
transformed into performance specifications, The best overall solution to the sponsor's
at least one such p-spec for each interaction. problern must be determined by evaluation
The p-spec must be worded to define (verification) of the solutions found by the
performance, not a design requirement (e.g. above procedure. This should be done by using
meter to be visible from operator seat, not the evaluation criteria included in the p-specs;
by simulation with analogues (mathematical, obtainable 'off the shelf', or that technology
models, mock-ups, layout drawings, experiments, is (or will be before the element is actually
Computers, etc.); by using component matrices needed) far enough advanced that they can be
after QUIRK (1961); by submitting the solution made. System search concems itself with
to the judgment of an independent authority obtaining the required system properlies by
or team; or by trying to modify, simplify, or connecting the available elements in a
transfer functions, eliminate parts, include suitable way, and thereby bridging the gap
other parts, combine or standardize parts or between the inputs and the outputs.
functions, etc. Initially this is done on paper using an
The chosen design solution can now be abstract 'model', a throughput flow diagram
communicated to the production organization, (TFD), where each block has a function but
lnttuactm g
No. P- spec Solut1ons Remarks
caf.gori•s
Housing to p•rmit Atr
No d-requtrH
1 1- VII NSY movem•nt
action Wheels Skid cushton devetopm•nt
and stability
Housing of Linewtth Mak• out of No
2 1- Vlll pleesing shape, to sound-deac Break sound-dead
r~sonance acllon
r•duc• n oise matenal mat•rial
Drive to function Sealed Oil- Dry- Take thrust
3 1- 111 in all positions ball retatntng lubrtcant ln both
of housing bNrings beanngs beanngs dtrecltons
Nozzles easy to Cone and Cylinder
4 II- VI fit,with air-light Bayonet socket
joint JOintseal
JOirrts (seil-hold) • clip
Nozzl•s eff•ctiv• Fixed Variabl•
5 II- VII forall conditions, geometry geometry
easy to use nozzles nozzles
Rigid Flexible b- narrow
Brw>h ·
6 II- Vll II nozzle nozzl• nozzl• nozzle only
mat•rial material
Drive must not Dust cont. Dust cont. b- care tn
7 111- IV obstruct dust before alter passag.,s,no
l!lttraction I an fan dead spaces
Access.ories to Joints as
No
8 V-VI fit onto tub" action for p- s.pec
l!xtension 4
M/c to permit Dust cont. Dust cont.
NSY l!mptying outside
-
inS~de
...__
9 "llll ._)UJI_
~
Figure 3.9. Solution chart
together with a written report containing a little or no relation to a separate piece of

summary of all assumptions and their effects hardware (a system element). The blocks of
on the design; a description of function under the TFD can then be c o m bin e d, split,
all anticipated conditions; model instructions redistributed and augmented by 'matching
for assembly, control and m a in t e n an c e; elements' (abstract or real blocks required to
expected behaviour; a model test programme; match or transform the output of one element
and a critique of the design which should be to the input of the following element), until
kept open to include all customer complaints each new block has an equivalent available
as they come in. element of hardware; this is the block schematic
System Search diagram {BSD). The resulting system must be
The basic assumption here is that the checked by calculation and experiment, and
elements are available: that they are either modified if necessary to yield the required
perfonnance. It may now be possible to if an available system appears to fulfil the

reshuffle the elements or the inputs and bulk of the function, the problern may be
outputs to optimize it with respect to reduced to finding the bridging system at
performance or cost (capital, running or overall each end, as in Figure 3.11.
cost). Experimental checks at this stage are
done on a 'breadboard' arrangement (in the flat)
without regard to the future spatial arrangement
of the elements relative to one another. Only
when a satisfactory functioning system has
been set up does the designer consider spatial
arrangements, the optimum means of physical
interconnection, and the aesthetics and human
operator (ergonomic) aspects of the new
equipment. This is done either on paper (a
layout drawing) or by means of solid models.
The methods used in the early stages, Figure 3.10. System search tree
up to and including the BSD, have been
abstracted into a systems engineering approach
by GOODE and MACHOL (1957), HALL
(1962) and GosLING (1962). Briefly, three
approaches are possible:
(1) lf a similar system already exists

(with the same inputs and outputs, or with
Figure 3.11. Double bridging
analogous inputs and outputs), a BSD can be
derived, a TFD abstracted, and a new BSD
can be set up to yield the new system using Output
different elements.
(2) lf an almost similar system exists
(with the same or analogous inputs or outputs,
but with differing values for the other) the
TFD may be subjected to 'perturbation' before
the new BSD is derived. For this purpose,
the properlies of a block in the TFD are Figure 3.12. Feedback
(arbitrarily) changed and the effects of this
change are propagated downstream towards the It is important to note that most systems
output (or upstream, or even both) to determine are not 'open' as Figures 3.10 and 3.11 imply,
the possible change; the change is then but that they usually contain feedback: a
optimized until the required properlies are portion of the output of the system or a small
obtained. sub-system is suitably transformed and added
(3) If no similar system exists, then into (fed back to) the input side (Figure 3.12).
cascading a number of sub-systems or unit This can occur deliberately (to make the system
elements may yield the required system. Work dynamically stable, or to make it oscillate in
proceeds either from one end, or from both, a controlled fashion), or by accident as stray
listing all possible steps in the form of a 'tree' interference due to electromagnetic coupling,
(Figure 3.10) until one or more bridges that sonic vibrations, trapped matter (in boundary
appear possible can be found. Alternatively, layers or in dead spaces behind baffles) etc.
THE DESffiN METHOD 31
The loop may even be 'closed' by a human possibilities, capacities, and pattems of
being, watehing the properties of an output failure. All design evolves and revolves around
(displayed by meters or signals) and creating the human being, as creator, user, and master
a deliberate input by force or position of a of engineering artefacts. Therefore the designer
lever, handwheel, pedal, pushbutton, etc. must constantly be aware of the needs and
This obviously is one of the main features properlies of his human environment, whilst
of the interaction between man and machine. applying technology, methodology andeconomics
The system forms part of the environment for towards fulfilling those needs.
the human operator, and the human operator The interrelationships between specifi'c
is an environment for the system. Each technologies and varieties of design are
contributes properties in terms of practical considered in Chapter 33.
PART II
THE HUMAN PERSPECTIVE

Chapter 4
THE HUMAN PERSPECTIVE, THE DESIGN SITUATION

AND ITS OPPORTUNITIES
S. A. Gregory
lntroduction must be defined. Such definition is by no

To design is to plan for the fulfilment means easy nor, for that matter, is it likely to
of human satisfaction. In Chapter 5, Mayall be produclive except in time of siege or
develops the idea of the spectrum of product - national catastrophe.
person relationships, of the variety of ways in Some foods are preferred above others
which a product may affect the individual who because of their taste, or their consistency, or
uses it or consumes it. His argument is their after-effects. Some, indeed, find favour
particularly expressed in terms of the because of their colour or shape, their social
relationship between capital equipment, such prestige, the advertising which has promoted
as machine-tools, and the operators. them, or because they fulfil some deep-down
The consumer is more than an operator of urge or some traditional formula. Where men
machines, although, in modern society, an are no Ionger at starvation level, and
increasing number of the satisfactions wbich sometimes even if they are below it, they
are at his disposal come from mechanical express choices which cannot be traced
equipment in and araund the home. directly back to simple physiological need.
Mayall sees the first approach to design In the examination of the needs for
as likely to be in terms of the immediate shelter, for example, definition is found to be
'technical' solution. This holds true for either even more difficult attd imprecise. Shelter is
the simplest consumer products or sophisticated related to food and clothing, to environment,
capital goods. For capital goods the technical particularly the weather and its changes, to
function of the machine is important. For the behaviour of animalil and people in the
consumer goods, on the other hand, concem vicinity, and to the materials available for
i& with the function of the consumer. building. Apart from such material features,
Certain fundamental needs of the consumer the psychGlogical demands of security artd the
may be expressed, and these are listed by level of aspiration must be consideted,
Stobart in Chapter 6 as food, shelter, heat, tagether with the complex of tradition and
light, and pleasure or recreation. MAsLOW, fashion; the social background and the skills
(1954) approaches the matter in a different way. and techniques available must also be
Some of these needs may be put in quantitative examined. Such ate the requirements fot any
terms. A certain basic requirement of food detailed study of needs and wa11t11.
exists for humans to grow, to work, and to live
without obvious discomfort in specific
atmospheric conditions. This is a requirement The Marketing Approach
of physiological function and is related to The study of opportunities for design,
age, to work, and to physical environment. which means opportunities for business action
To define the basic food need the other factors or for the exercise of public welfare agencies,
35
36 THE DE~GN METHOD
is essentially that of marketing, as it is particular occasion a product profile may be

expressed today. The marketing approach, as set up similar to the proposal of HARRIS
discussed by DRUCKER (1964) and others, (1961) and WARD (1965).
is the orientation of organizational resources People who have a close contact with the
to the satisfaction of the consumer. In this needs of the market see clearly that market
the first task is to establish the market factors are design factors: the points which
opportunities which exist for the organization. need to be made clear to a potential customer,
The study of opportunities does not in order that he may acquire the product with
initially demand the full range of skills now the greatest potentiali ty for satisfaction, are
made available by the human sciences. What those points which have been carefully dwelt
is usually needed is an examination of upon by the designer. They are not, of course,
commercial data on sizes of population, the only points which a designer has to worry
population growth rates, national incomes, the about. He has to consider t e c h n i c a 1
breakdown of consumer expenditure, and so on. possibilities, the resources available to his
For basic consumer goods, raw materials, and organization, and so on. Somewhere too
capital equipment i t is usually possible to consideration has to be given to the future,
find emde overall figures by one means or both of the market and of the organization,
another. and this has to be built into current products.
Methods of arriving at data of this kind The strategic approach is dealt with by
are discussed in some detail by STACEY and WILLIAMS and FINLA YSON (1963) and by
WILSON (1%1); some other references with DE VRIES (1964).
relevance to industrial products are provided
by Turner in Chapter 27. The theory of needs The Rise of the 'Soft' Seiences
research is dealt with in depth by HALL Although much may be clone by fairly
(1962). In a chapter of his book he examines simple procedures to improve marketing,
the mathematical relationships which are eventually a limit is reached when it becomes
generally relevant, even if the data are absent. necessary to employ some more sophisticated
The provision of data regarding the present approaches. Usually some application of
circumstances, in situations where specific economics is the first step. The behaviour of
information on the particular market is lacking, the market and of the organization are defined
has been discussed by SALOMON and BROWN in economic terms, and the product is specified
(1964). The difficulties, both technical and and analysed in economic terms. Similarly the
personal, of long-range forecasting, have been sum of satisfactions is set out in economic
picturesquely described by MUMFORD (1961). terms.
These activities may be seen as part of a Analysis of practical situations shows that
larger operation, of marketing as a whole. a simple economic model is not satisfactory
Most of the thinking about marketing has been for a company, as in the recent study by
in terms of consumer products. The general WILLIAMS and SCOTT (1965). Although there
problems and requirements of marketing have has not yet been an adequate study of design
recently been covered by RODGER (1965) and practice, it may be reasonably supposed that
the requirements of the introduction of a new design decisions are the outcome of some
product have been reviewed by COOKLIN operation more complex than technical and
(1964). For engineers and people concerned economic considerations. The customer and
with industrial products the field has been consumer requires much more from his purchase
clarified by the new book The Marketing of than adequate technical performance at a low
lndustrial Products edited by WILSON. For a price. It is to the study of this point that more
THE HUMAN PERSPECTIVE 37
work is being devoted by experts in various tend to concentrate upon a single approach,
branches of the human sciences. an approach which depends upon their primary
What does the consumer want? This itself training or inclination. Between the most
is a problern of considerable technical extreme of these approaches, the strictly
difficulty. What makes the consumer take his technical and the finely aesthetic, there tends
decision? From simple description of need to be a gap. Most of the aspects of Mayall's
more information must be dug. The psychologist spectrum can be seen and procedures set up
is required to reveal motivation, even the for their ascertainment, but the aesthetic and
social anthropologist to lay bare the role of symbolic largely evade satisfactory quantitative
the product. treatment.
Nor is this a problern just for consumer This is not a matter which is limited by
goods. The pioneering work of Peplow simple items of consumer demand. Large
(Chapter 9) shows how even with industrial complexes of structure such as town centres,
equipment there may be substantial new liousing estates, and major industrial
psychological factors at work to influence sites lay claim to attention. What can be
acceptance. His contribution links up found out from the students of aesthetics to
remarkably with the earlier work of ROGERS help bridge the gap which lies between the
(1962) which deals with the diffusion of new purely technical solution of the problems of
techniques, largely those used in agriculture. function and the ill-defined attempts at the
It is by no means without significance that, attainment of style and form?
as yet, there is no uni versally satisfactory or Philosophers, such as CARRITT (who
accepted theory of economic development. gives a compact list of references), have
Growth may be described and projected but little to offer other than a critical approach,
not readily explained. and would direct attention to physiology,
But it must not be taken that shortage of psychology, anthropology, sociology. For many
this kind of fundamental theory leaves the years, artists have known of particular
designer without resources and possibilities geometrical shapes with the ability to provide
of action. It is a good rule to provide in a a minimum of satisfaction, or at least an
design those functions which are essential absence of tension (see Newman in Chapter 13).
and to do this at the lowest cost. It is a further Such is the relationship given by the 'golden
good rule to seek out information from those section'. Artists and others have developed
able to influence decisions. But even with structures of theory around the subject and
these rules success is by no means assured. many of the shapes recorded in nature by
With novel technical products, such as THOMPSON (1917) find favour. More to the
electronic equipment, the information that may point has been psychological investigation
be obtained is likely to be as good as that and much of this is conveniently brought
given 'by putting probes through the side of a together by V ALENTINE (1962). But this kind
cocoa-tin to find out what is inside'. Contact of psychological work is only able, through
may never be made. experiment, to deal with relatively simple
phenomena. Those aspects of perception now
being studied by the ergonomists to help in
Aesthetics and Symbol ic Value their appreciation of man- machine interaction,
Each specialization in design has its own concem the delivery of data to the human
magic and meaning. Instead of treating the brain, a complicated piece of equipment which
range of skills as a basis for dealing with the is able to influence the choice of things
continuum involved in each problem, people perceived and the way in which they may be
interpreted. GARNER and colleagues (1963, Conclusion

1964) are approaching aesthetics through A person's first task as a designer is to
experiments based on information theory. find opportunity where need may be satisfied
The practical consequences in art are by the use of his particular skills. This need
well brought out by GoMBRICH (1960) who will be based largely upon what he recognizes
brings to his discussion c o n s i der ab 1 e as economic and social considerations, and
sophistication regarding the views of his decisions in design, beyond technical
psychology. The attempt to bring together feasibility, will be determined by them. But
modern views of psychology and the nature of he has to recognize the important part played
architecture has been made by NORBERG· by perception and the way in which its function
ScHULZ (1963). This is a topic in which and extent varies according to the product.
discussion may run far from practical He has to put into his design just the right
considerations. Architects, of all designers, amount of 'perception engineering'. He must,
seem to run the most risk of an inadequate in the words of Mary Follett (as reproduced by
definition of objective. PERLMUTTER (1965) METCALF, 1941) find and obey the logic
shows the problern in other ways. of the situation.
Chapter 5
DESIGN AND HUMAN SATISFACTION

W. H. Mayall
'An engineer ls one who dlrects the great sources of

power in nature for the use and convenlence of men.'
Thomas Tredgold
lntroduction has, by and large, created a seller's market.

By whichever route most engineers reach Especially with engineer-based consumer
the point of designing products for human use products such as automobiles, refrigerators
and convenience, they acquire no more under- and so on, people have been only too anxious
standing of people and society than their own to acquire them; and with little or no prior
personal experiences and powers of observation experience have not been able to discriminate
permit. With respect to engineering designers, between those which are suitable and those
as indeed to other types of designer, both of which may be unsatisfactory. Wider ownership
these abilities may be limited. For the of such products, increasing ownership
engineering designer the limitation may be experience, and the work of a variety of
especially severe. Technological development consumer-conscious organizations are changing
has forced him to concentrate upon increasingly this situation. Without formal instruction, a
complex technical problems. Consequently condition which might well change in
his professional education is heavily biased educational pattems of the future, the buying
towards techniques. Indeed, for many engineers public is slowly becoming more selective.
the design of engineering products is largely In the so-called affluent societies the
interpreted as a process of producing technical purchaser's ability to discriminate is clearly
solutions. This task is not, of course, to be backed by a financial ability to do so. lt can
underestimated. It would be entirely wrong to be argued that affluence encourages an appeal
suggest that engineers should gain a greater to comparatively slight or superficial needs.
understanding of people and society at the Thus a car may be bought for its colour rather
expense of acquiring technical expertise. than its technical specification, for its quality
Such expertise is needed to an increasing as a status symbol rather than its performance
extent. But if it is to be properly utilized characteristics. Indeed, the problern of a highly
then a fuller knowledge of the users of more affluent society may be the creation of needs
adv9Ilced products and how these products which engineers, in particular, may regard as
will affect society becomes more necessary. entirely ephemeral. But however slight these
There are several reasons why this need created wants may appear in the design of
must be recognized. First, the comparatively consumer products for such a society, they
rapid development of so many types of product cannot be overlooked. Even a slight
since the beginning of the nineteenth century understanding of social history demonstrates
39
that the affluent have always exercised their the personalities of those he is intending to
foibles in their choice of products; a very serve. The financial loss to the Ford company
sound reason for being affluent. was in the order of .f120 million. Yet the
Although the financial power to discriminate engineering was probably perfectly satisfactory
may be smaller in less affluent societies their in terms of the current state of the art.
ability to discriminate is growing, partly due to Product investment costs are likely to
trading competition and partly because financial continue to rise so long as products continue
limitations demand selection of the best to become more complex in technological
available products. Such discriminatory ability terms. Thus those who may be engaged in
in these societies may relate more to the some of the most specialized technical tests
purchase of capital plant and equipment. Such might reflect that the cost of their own services
products may be seen not simply as means to must be regarded, in part, as a reason for_ paying
but emblems of social development and thus more attention to those who will use their
carry psychological overtones which a products.However refined their accomplishments
perceptive manufacturer should recognize. and however large the costs they incur in
While increasing powers of discrimination product development their work could fail for
at most purchasing levels is perhaps the prime purely 'human' reasons.
reason for making product design more than a A third, and for some the principal, reason
matter of solving technical problems, a close for recognizing human needs and attitudes is
second is the far greater investment required that a time might be rapidly approaching when,
to design, develop, and tool-up for modern unless a greater sense of social responsibility
engineering products. For a new car the is undertaken by manufacturing organizations,
investment can run into several million pounds. with active participation in social planning,
A very small electronic unit recently examined, society may suffer rather than benefit from
cost in the region of !60,000 to bring to a technological development. The effect of the
preliminary production condition. Obviously automobile is usually cited in this context,
investments of this order are unwisely marle whether in terms of safety - the writer recently
if a thorough study of those for whom the observed the effects of the stylized prow of
product is intended is not undertaken. And one vehicle upon an elderly lady - or in
although it is sometimes believed that the relation to the points raised by the Buchanan
creation of wants by mass-advertising may report, for example. In one sense this report
carry a pr-oduct through, as it were, it is can be regarded as an attempt to correct
doubtful if such 'created wants' will prevail foreseeable, indeed foreseen conditions. But
unless they are related with current or correctives may take so long to apply that,
projected social conditions and aspirations. bearing in mind changes due to other types of
Perhaps the most quoted example in this development, they could well be frustrated.
respect is the ill-fated Ford Edsel car. A Luddite solution to this type of difficulty
Wrongly believed to have failed because it inevitably springs to mind, but this is neither
was marle by a company with an excessi ve desirable nor practical. Altematively to impose
zeal for market research, the Edsel failed some kind of dominating central control for
because, although a market had been discerned what some people regard as a technological
(the rising executive class) the Ford company free-for-all would be unpalatable. The main
did not recognize that this class could not, on task must surely be to recognize that products
the whole, see the Edsel as an emblem of help to form social patterns; an aim must be to
executive status. The Edsel affair makes the predict how these patterns will change with
point that it is necessary for a manufacturer the introduction of new types of product. Thus
to recognize his own personality as well as the engineer must look beyond his own product
and the people who will use it to other products reduction of the degree of human skill required
and their relationships. In consequence, and to manipulate a product may well cause
as a creator, he must necessarily associate resentment rather than receive acclaim,
with other creators - architects, industrial especially if this skill has been hard-won and
designers, artists - if human satisfaction confers distinction upon those who possess it.
expressed in terms of social responsibility Such attitudes are not arguments for ignoring
is to be fulfilled. ergonomic factors but simply illustrations that
designing for people involves more than current
An Outl ine Approach to Human Needs interpretation of ergonomics tends to recognize
in the Design Process and, more fundamentally, evidence that people
The growing interest in and partial cannot be encapsulated by formulae. It will
appreciation of systematic thinking in the always be necessary to treat each product
design process must be developed with a situation on i ts merits.
recognition of human needs. But it would be A second reason why questions are more
naive to suppose that systematic techniques important than answers lies in the nature of
can be devised which, in themselves, will the design process. A fundamental characteristic
produce what might be called human answers. of this process is that of making adjustments
There is the dangeraus chance that this kind or compromises in order to achieve a so-called
of thinking may develop very largely because optimum overall solution. In the design of
designers can easily become formulae-bound. most products the 'optimum' is not necessarily
The formulae need not necessarily be a unique or ideal solution. It is determined by
mathematical in character. Styling formulae, management policy, by available resources,
in the sense of picking up and perpetuating a and by the target dates, all of which vary from
particular shape or configuration, are all too time to time. Thus answers to human aspects
apparent in the more aesthetic areas of design listed for one set of conditions may not be
for products. In dealing with human require- relevant in another set of conditions. Some
ments it may be inore profitable to consider engineering designers may be familiar with
how the designer can systematically ask the those who attempt to press a particular
right questions rather than how he can by mathematical formula into service because it
system obtain the right answers. appears to approximate to some mechanical or
There are two reasons why this course is structural design condition. They know that
likely to be more profitable. First, a most misleading information can frequently result
important point about people is that they are from this tendency. Similar effects may occur
human. They do not behave in the neat if data on human behaviour 81'e applied without
predictable manner of mechanisms. Their questioning the overall person-product re-
reactions to and ways of using a product in lationship.
one set of circumstances may be entirely Just as it would be naive to suppose that
different from those in another set of systematic techniques can in themselves
circumstances. They differ in dimension, in p rod u c e 'human' answers, it would be
physical and mental abilities, in culture, and presumptuous to believe that a kind of human
in aims and aspirations. They can confound questionnaire can be produced on virtually
those who, with the best intentions, contrive a priori grounds. Such a questionnaire must
to produce the 'logical' answer to their eventually come from continuing sturlies made
perceived needs. Thus a comparati vely by designers, human factor specialists, social
transitory style, say in the structure of a chair, scientists, and objecti ve market researchers.
may well influence them more than that the Many aspects of person-product relationships
chair is satisfactory in ergonomic terms. A remain unexplored while others may seem to
have been veiled with a mysticism which one he will, or should, be concerned with how
cannot but feel to be sustained purely for the operational information is presented to the
personal profit of a variety of 'experts'. But human being, how this information is
while waiting upon more research into person- interpreted, and how control information is fed
product relationships it would be reasonable back to the machine. This aspect of product
to suggest a mode of thinking which, for design cannot be considered without taking
engineers in particular, might help to extend physiological characteristics into account.
their attitudes and provide a basis for asking It seems reasonable, in consequence, to locate
questions. The mode of thinking suggested this as the next human aspect along the
here is certainly not put forward as the only spectrum. But psychological considerations
one which can help to encourage a wider extend beyond those c o n c e r n e d wi th
appreciation of human needs. In fact it tends manipulatory or control activities. They
to overlook one important aspect of product broaden out to what are usually defined as
design which will be considered later. But at aesthetic considerations. This is particularly
least it may serve as a starting point for the case with the majority of production
further investigations. machinery in so far as visual and tactile
faculties are concerned. For example, reactions
The Spectrum of Person- Product Relationships to form and colour on production machinery
For the engineer the core of the design will usually be based, either consciously or
task is always likely to be the achievement of unconsciously, upon whether they assist
a technical solution. Taking the colour spectrum control requirements. Even where an ergonomic
as an analogy, the so-called technical factors approach to the use of form and colour may be
involved in design could be regarded as a less important, it is reasonable to assert that
broad band at one end. These factors, basic aesthetic reactions can be classed as
performance characteristics, efficiency, life, psychological. As long as there is an inadequate
suitability to ambient conditions, and so on, understanding of the constituent elements of
might be achieved in theory without reference an aesthetic reaction, it would be unwise to
to human characteristics. On occasions one do more than tentatively suggest how such a
suspects that this happens. However, the reaction may be composed. A study of the
choice of technical solutions must de:pend aesthetic critiques tends to show that certain
upon whether they permit the product to be concepts, particularly the desire for unity and
transported, installed, operated and maintained. orderliness of form, can be related with some
Therefore recognition of human characteristics of the concepts propounded by the Gestalt
in terms of size and physical ability will psychologists. These concepts would appear
usually be first for consideration. Satisfaction to have universal relevance and, indeed, they
of atithropometric and physiological require- can be utilized in determining machine-control
ments could be regarded as forming the next layouts. Mainly for purposes of identification
band of colour along the spectrum, linked to these concepts might be described as 'basic
t e c h n i c a 1 aspects by equi valence in aesthetic factors'. However, they are certainly
measurement, i.e. in terms of length, mass, overlain by others which are less understood.
time and energy. It may be convenient to identify these other
Further along the spectrum the next aspects by the word 'style'.
distinguishable band could identify the While a product's style is strongly
psychological relationship between human and conditioned by the materials and processes
machine. For the engineer this relationship employed in its construction, the effect of
will have the greatest emphasis where i t the product in relation to its intended
relates to operational aspects. In this regard environment cannot be overlooked. If the
aesthetic desire for unity is required in the of the product itself can vary considerably
product itself then it is reasonable to expect between different people and different social
that it will be sought in relation to current groups.
standards in architecture and in the form of There are those for whom the symbolisms
other products. It is important to recognize generated by many types of product are to be
here that environmental influences upon either derided or ignored. The fate of the
product style may be conceptual rather than Edsel car should be a salutary reminder that
real. For example, one would not expect to derision is an extremely risky pastime. Others
design a nineteenth century appearance into a may attempt to make ethical issues out of
machine destined for a twentieth century such influences, suggesting that 'education'
factory. The machine, regarded as modern in would encourage people to see through what
concept, will usually be more acceptable if it they may like to describe as sham values.
bears a recognizable relationship wi th other But however these values are described they
modern products. Architecture has always had have existed, in their kind, throughout the
a strong influence upon other products. history of human society. Much could be
Conceptual attitudes to products bring one learned from a closer historical study of these
to the end of the spectrum. Here a band can be aspects, for, up to the moment, very little
identified which, while linked to so-called objective analysis has been undertaken in
basic aesthetic concepts through environmental relation to product 'images'.
considerations, begins to relate with a variety
of symbolic interpretations which people tend
to place upon products. lt is in this area that Possibilities in Adoption of the
more needs to be known than is at the moment. Spectrum Analogy
For, though they come to the end of the By listing human requirements in a
spectrum of human satisfaction, symbolic connected order stretching from meeting
interpretations can play a very large part in ergonomic requirements, through basic aesthetic
determining product sales. Their influence is needs, to the recognition of environmental
certainly more noticeable in the character of influences, with symbolic meanings finally,
consumer products but is not to be ignored in a form of importance rating is automatically
the field of capital plant and equipment. expressed. Thus in general, ergonomic needs
For example, some time ago a manufacturer should be regarded as having greater
redesigned a machine to save time on finishing importance than style. The analogy stretched
cast housings. He calculated that it was to include technical factors, helps designers
eheaper to surround these housings with a sheet to produce, if only very roughly, a graphic
metal casing which would be eheaper to produce expression of the relative importance they are
and finish. But while a number of purchasers giving to the various design factors listed
readily accepted the change, since the along the spectrum. In all design situations
manufacturer was able to cut the price, one one would obviously expect technical
potential purchaser rejected the product requirements to have the greatest importance
because he could not see that the machine was and a hypothetical 'importance rating' graph
made of the 'good solid cast iron' he had would always slope down towards the symbol
always associated with machinery. Indeed end of the spectrum. But it is not too difficult
he preferred a more expensive machine without to visualize different rates of slope for different
difference in performance characteristics. types of product. For example, factors making
Conservatism in product acceptance and up the style waveband are of more importance
belief in social status acquired both in owning in consumer product design than in capital
a particular type of product or by the nature goods design.
Finally, if it is accepted that a logical extreme case, an ergonomically and

connection can be made between the various aesthetically satisfying flick-knife is no
factors making up the spectrum, then the contribution to social betterment.
analogy may help to serve as a preliminary lf they are to be seriously concerned
model upon which more detailed studies of with human satisfaction then engineers and
human needs and attitudes can be made. While designers in general need more than systematic
hoping that these more detailed studies will techniques, analogies, or other devices, to
be made, one may suggest that the analogy help them extend their attitudes. Certainly,
could form a preliminary guide for the designer, for great areas of the world, possession of the
both in forming questions on human aspects benefits brought about by science and
and in seeking and assessing available engineering are, and will be for some time,
information in relation to his design project. a major aim. But it may be wrong to suppose
that, because it has been so to the moment,
technology will always be beneficial. lt
Conclusion might be considered that a stage is being
The analogy suffers from one drawback in reached where consciousness o.f human values,
that it does not take account of the product's in the very broadest sense, must play a more
effect upon society. The concern is with dominant role in the way that scientific and
the total effect of the product. To take an technical efforts are directed.
Chapter 6
INVENTION, DESIGN AND MARKET RESEARCH

A. F. Stobart
lntroduction development before going out and studying the

Every man-made object was originally an market itself. The success of the market
invention in the mind of one or more people, development will depend as much on the
followed by a design for its manufacture and, design of the goods themselves and of the
in due course, by the development of a market service by which they are supplied as the
for the product. Inventions can occur without price at which they are offered. To quote
design and marketing; inventions and designs Sir Artbur Bryant: 'The quality design of
can take place without being marketed; but no goods exported, and the integrity and despatch
marketing can be done unless invention and with which British traders met their customers'
design have previously been carried out. demand, gave prodigious dividends.'
This refers to 150 years ago. Can the
same be said today?
Market Research and Design
In modern civilization it is quite possible Design for Markets
that a study of a particular set of human Leaving aside for the moment original
conditions, often known as the process of invention, which is usually not connected
market research, can throw up the need for with any immediate prospect of sale but is the
certain inventions and designs to take place, result of necessity, curiosity, or urgent
to fill gaps in the requirements, or the luxuries, personal need, design can be said to be aimed
of the civilization concemed. The court at fulfilling a few basic human needs. These
jeweller to a potentate of ancient time would are: food, shelter, heat, light, transport, and
take good care that bis products were designed pleasure or recreation.
to be suitable for bis master's taste or those One of the most fundamental features of
of bis female entourage. Equally since there modern civilization is its transport system,
is fashion in engineering, as in jewellery, for men, ideas, and materials. It is proposed
the modern engineer frequently has, at least to discuss design in connection with transport.
outwardly, to design bis works to suit the In the words of Kipling: 'Transportation is
taste of a particular dient or bis e ngineering civilization, and only when perfect freedom of
entourage. In either case the result can be a transport and all that it implies, is maintained
very gaudy display of little intrinsic merit. in the world, can the world be said to be truly
Market research for design, then, can be civilized.'
divided into two parts: (a) to discover what a The British people, possibly more than
market thinks it needs; (b) to discover what a any other race in human history, have been the
market would ask for if it really knew its needs. repeated instigators of new methods of
The second type of market research transport. In the early years of the medieval
involves undertaking a considerable pnrt of the kings there was freedom of transport between
design work for the market technical one town in England and the next by whatever
45
means were availahle, while Europe maintained meeting that need, and then adapting these
its petty independent states, levying tolls on natural ideas to use. The development of
all travellers. England developed heavy aircraft came originally from the study of the
transport, first hy the exploitation of existing flight of birds. The development of structures
natural waterways. The development of came from the growing of trees and then from
railways, as a kind of overland canal, came using parts of the tree as the basis for man-
later. Today, the oldest form of organized marle buildings. In some old structures there
transport, by road, is being taken in hand is, in fact, little pruning of the tree trunk
after a lapse of nearly 2,000 years since the hefore it is fitted into the building. Given
Romans first laid down the backhone of the the tree-like shape as important, sometimes
present road system. In addition Britain has other materials were formed into the same
three different kinds of transport of more pattern. The round stone columns of ancient
modern character: the transport of power hy Greek and Roman temples obviously owe
high-voltage electricity; the transport of gases, their ancestry to timher columns used for
liquids, and occasionally solids, hy pipeline; earlier buildings.
and the wide diversity of air transport. But the activities of mankind in invention,
How then does design influence transport? design, and marketing, or merchanting, as it
And how, indeed, does transport influence was called, were very limited until transport
design? How do both have a bearing on the was developed.
exploration of markets?
No market can be explored without Transport of ldeas
transport, whether it is transport of knowledge No invention or design springs into life
or physical heings. Primitive man, faced with complete, as Athene from the head of Zeus.
the problern of exploring another part of the A long, often slow, process of meditation,
forest, which was difficult to reach through the trial and error, and calculation comes first.
thick undergrowth, conceived the idea of Mankind has done so much of this groundwork
sitting astride a log in a local river and in the last fifty years, on so many subjects,
floating downstream until an attractive beach that much of the initial thinking on any new
opened up the possihility of a good set of development has probably been done, if one
living quarters. But logs were notoriously can find it. What are the harriers to the
unstable. So, either hy some process of movement of ideas? There are many, among the
extremely ingenious deduction, or more probably commonest of which are:
hy the advent of a partially rotten and hence
hollow log, primitive man achieved the idea (1) Language - The USA speaks American,
of the dug-out canoe, the ancestor of all not English.
hoats. (2) Prejudice - 'Can any good thing come
Herein lie the first two lessons of design: out of Nazareth?'
the study of the application for which a design (3) Lack of fundamental knowledge, orof its
is required, and the awareness that an already application - Iron is a chemical as well as a
availahle commodity can, with modification, metal. A pipe is a fundamental structural
be used to fill the need. A great deal of design shape, more economical for some applications
for various markets is the adaptation of than the sections commonly applied. Are either
existing ideas, or the extension of existing of these facts taken fully into consideration
designs and inventions to some apparently as often as they might be?
unrelated field. ( 4) Inadequate classification and poor
There are many examples of mankind information retrieval - The author once
feeling a need, studying the natural method of developed, in theory, a device for the
purification of chemieals by sublimation. A evaluation required involves the total cost of

prototype was built which worked well, but the purchase or rent of the idea, against the
further search uncovered two facts. Machines total profit accruing from its use, compared
of this type were in commercial production in with the same figures calculated for one's own
East Germany, and the principle and outlines developments. The profits available from a
of the process had been worked out by the quick entry into the market may more than
company's chief ehernist in 1925. A very outweigh the extra cost of buying a process.
glaring example of inadequate searching and To develop one's own process involves
poor intelligence. waiting and an expenditure which is as yet
unknown, but it could produce greater profit
The true inventor has none of these hindrances. in the long run than a purchased process.
True, he may have prejudices, but they are Market research is vitally necessary at
his own. He is often not accepted by society this stage of development, to assess the
(the market) because of this, although his courses of action which are available for
work is useful and may ultimately find wide the projected design, the influence of the
application. A true inventor rarely does market timing on the success or otherwise of the
research. The designer takes the fruits of the venture, and the level of profit which can be
inventor and converts them to articles which expected from the various courses of action.
conform to the prejudices and fashions of the At least one firm of consulting engineers has
market. This is the site of the first transport gone into this field very thoroughly. Computer
problern in ideas. How to get them from the programmes are set up based on the detailed
long-haired to the short-haired sections of economics of all the known possi ble
the organization? alternatives in a plant design. The programme
Sometimes the market may dictate a is run until two or three of the most profitable
requirement for which there is no complete combinations of processes and equipment are
answer and at this stage the inventor may be thrown up, in the context of given market
deliberately brought into the picture. This is conditions. These combinations then form the
more commonly known in industry as 'calling basis for a tender to a customer.
in the research department'. Market research Design is, however, of little value, unless
has a big part to play in feedback to the something is made and/or sold.
designer of the present and future requirements
of customers. This means that the market Transport of Materials
researcher must understand the designer, Trade is the transport of goods from
another problern in the transport of ideas. The seller to buyer, and the transport of money or
designer in turn must be capable of sorting other goods in return. Before the goods can be
out the problems which can be solved from made materials must be transported to the
current knowledge, and those which require factory to make them. The buyer is ultimately
the expansion of knowledge. The designer has, in charge of this function, but it is the custom
himself, to do market research. He must, as a for the seller to seek out the buyer, rather than
potential buyer, do research in the highly the other way round. Market research comes
specialized market for ideas. into play to locate the buyer, study his needs
A cost-time factor comes in at this point. and habits, and plan how to provide satisfaction
Many ideas are sold, or leased, such as for them. The barriers to the passage of goods
licences for patent exploitation. It may be from seller to buyer can thus be uncovered.
worthwhile to pay for an idea from someone Apart from the most obvious, and partly
else if one cannot be sure of devising a man-made, barriers of freight and duty, there
eheaper alternative in the time available. The are two main obstacles to overcome. These are
48 THE DESIGN ME THOD
price and value. Price reflects the efficiency profitable outlets in world markets. Sometimes,
of the producing firm in keeping costs to a low indeed, there may be a market for the idea
enough level to allow a profitable sale to be alone, in the form of a licence to manufacture.
made at an attractive price. Value is the cost An interesting result of the reduction of
of use to the buyer, and is, or should be, the weight of a product, with an increase in 'brain
result of successful Co-operation between the content', is that the price per unit of weight
market research and design teams. Market should go up. This, in turn, will reduce the
research can pinpoint what aspects would be percentage increase in price which has to be
of most value to the user: design teams have charged to cover freight. Thus a more distant
to try to incorporate them in the finished market may be tackled with this new commodity
article without overstepping the allowable than is possible with a commodity with a low
production cost. price-to-weight ratio.
Another aspect of value is that most of
the input should come from the brain and as Conclusion
little as possible from materials. Great Britain Invention pays little heed to market
is an island, dependent for much of its food research, although it may be initially fostered
and raw materials upon imports. Her only or encouraged thereby. But design, which is
major exports, apart from a little coal and iron, the basis of all industrial effort, should -
derive from the ingenuity of the inventors, indeed, must - be guided in detail by market
designers, and craftsmen who live in the research to achieve the objectives of all
island. Market research has a vital role to successful business, maximum profits with
play in guiding the brain power into the most maximum customer satisfaction.
Chapter 7
ERGONOMICS AND DESIGN
B. Shackel
lntroduction and plant, and these decisions obviously

The contribution initially invited from affect greatly the efficiency and welfare of the
me was psychology and design, which should humans to be involved. Errors can be costly
be a broad enough brief to satisfy any author. in both human and economic terms, but not all
However, the theme of this book is design such errors are necessary, Iet alone inevitable.
method, and to concentrate on applied Research has already yielded much relevant
psychology alone seems unjustifiably limited knowledge; but in many situations the engineer
in this design method context. Neither the or manager may still have to fall back on 'rule
user nor the desi gner of a product or a of thumb' or guesswork. To replace such a
production machine Iimits his actions or potentially costly basis for decisions by a
responses to the psychological plane. The systemalle scientific approach is a prime aim
human factors side of this field must be of ergonomics. Moreover, the engineer or
concerned also with anatomical, anthropometric manager hirnself often can and should use the
and physiological aspects, as weil as the relevant research findings, provided he is
sociological and other areas examined in other competent and able to interpret them correctly;
chapters. Therefore, to embrace these related it is not essential, though often helpful, to
biological sciences, I propose to discuss have the consultant advice of an ergonomics
ergonomics and design. specialist in order to adopt this scientific
Machine designers are mostly, and quite
approach to the human factors aspect of modern
rightly, concerned to improve the mechanical,
technology*.
electrical and other performance aspects of
their machines; they sometimes forget that
what matters most is the efficiency and What is Ergonomic:s?
performance of the total system, of which their Ergonomics is defined as the study of the
machine is a part. Often the machine is one relation between man and his occupation,
element in the man-machine system, and what equipment and environment, and particularly
may be more important is how weil the machine the application of anatomical, physiological
works in conjunction with the operator who has and psychological knowledge to the problems
to use it day after day in his routine work. Now, arising therefrom. This definition, and the
there is nothing especially novel about field work, clearly can embrace both research
emphasizing this need for a good fit between and practical application.
man and machine; but the difference with Research is essential to increase the
ergonomics is that it provides the methods for the knowledge available on the multitude of
a systematic scientific approach to the problem. ways in which men behave, how they are
Engineers, managers and administrators
have to make the final decisions on the design, * References for this Chapter appear in the special
acceptance and establishment of equipment section on p. 342.
49
so THE DE~GN METHOD
similar to and different from engineering under the Industrial Health Research Board
components, and how they respond to and are between wars. World War II led to greater
influenced by their task and environment. emphasis on not merely matehing men to
Aspects of practical application are machines by selection and training, but also,
discussed in the rest of this chapter, but the much more than previously, to the designing of
equal importance of research must not be equipment so that its operation was within the
forgotten because without it there is no new capacities of most normal people. This fitting
knowledge to be applied as new practical the job to the man increased considerably
problems arise. In the study and treatment of the collaboration of engineers in certain
practical situations, Ergonomics (or Human fields with the biological s c i e n t ist s.
Engineering, or Human Factors - broadly This collaboration, beginning primarily with
equivalent names used in the USA) places military problems, because it is there
major emphasis upon efficiency in the operation particularly that operators are pushed to their
of the equipment as measured by the speed and limits, continued after the war and led to the
accuracy of the human performance. Allied formation in 1949 of the Ergonomics Research
with efficiency are the safety and comfort of Society. Attention has since returned to the
the operator. Because the aim is to optimize industrial field, and the joint approach has
the man-machine and man-environment increasingly shown its value as an ally,
combinations by altering the machine and the according to circumstances, of design
environment, this aspect has also been termed engineering, work study, industrial medicine
'fitting the work to the man'. Equally important, and personnel management.
although often regarded as a separate subject The field has already grown so far that it
from ergonomics, are the personnel factors is impossible to count, for certain, the exact
such as selection, training, and adaptation to
total of groups at work in the world. In Great
environmental and working conditions. These Britain there are some fifteen academic or
are studied both as part of ergonomics and as applied research groups, and at least four
separate topics under the headings of Work
units sponsored by the Services for military
Physiology and Occupational Psychology. problems. For training there are now courses
From this knowledge people can be helperl to
available of at least one year's duration for
alter themselves, within limits, to improve the
postgraduate students at the Ergonomics
man-machine partnership; this personnel
Laboratory, Cranfield College of Aeronautics
aspect has also been termed 'fitting the man
and at the Department of Ergonomics and
to the work'. Loughborough College of
Cybernetics,
'Fitting the work to the man' and 'fitting Technology; undergraduate courses have
the man to the work' are obviously also major recently begun at the latter. At the former,
aims of managers and engineers, but ergonomics ergonomics lectures are also a routine part
has a real contribution to make because of the of the regular engineering courses.
specialized knowledge available about human
characteristics and performance from the On the industrial side, the British Iron
relevant biological sciences. and Steel Research Association, British
European Airways and the Central Electricity
How did Ergonomics Develop and Where Generating Board have small sections at work.
is it Practised? Smith's Aviation Instruments, Design Magazine,
What is now called ergonomics had its Michael Farr (Design Integration) and ICI each
beginning in Great Britain in the scientific have an ergonomics consultant on retainer.
study of human problems in ordnance factories The British Transport Commission, the British
during World War I. This kind of work continued Motor Corporation and British Railways
(Southem Region)have commissioned significant Wh ich are the Relevant Biological Seiences?
work. Each of the following major companies, The biological sciences, from which
British Aircraft Corporation, Pilkington specialized knowledge is applied, are functional
Brothers, Richard Thomas and Baldwins, and anatomy, anthropometry, work physiology and
Lyons, has one qualified staff member working applied or experimental psychology.
on ergonomics matters. The staff at Philips Functional anatomy is concerned with the
have worked mainly on shop floor problems but body framework, posture, and the use of muscles;
are now developing design work as well, and it can, therefore, provide knowledge of the
EMI Electronics has, it is believed, the largest best ways in which force can be applied or
and oldest (11 years) group, with a permanent objects lifted and also about the limits of
qualified staff of four, working almost entirely joint movement. This knowledge is, of course,
on design problems. very important in the design of controls such
In the past, Tube Investments, AEI as levers and foot pedals.
(Manchester), Vickers Research, GEC Atomic Anthropometry yields data on body sizes
Power Division, and the Boot, Shoe and Allied and dimensions, both for male and female,
Trades Research Association have employed and thus can help with the optimum design of
ergonomics staff and significant work has the height and size of working surfaces, the
been clone. In the present, acti ve interest in position of hand and foot controls, the shape
the field is being shown at least by the and height of seating, and so on. It is very
following: Machine Tool Industries Research important but often forgotten, that when
Association, Motor Industries Research designing equipment for females, female
Association, English Electric Atomic Power dimensions should be used. For example, the
Division, Pilkington's Glass, Heinz and United usual height of an upright chair is 18 in.,
Steel. However, although there now is this which has been shown to be too high for about
growing interest and activity over quite a wide 60 per cent of both the male and the female
field, it must be admitted that the total of real population (FLOYD and ROBERTS, 1958).
work and qualified specialists in industry is Work physiology covers the whole field
still very small indeed in Great Britain. of bodily activity, particularly with reference
In Europe, the type of work and the rate of to many different types of environment. For
development is similar, but on the whole it is instance, reasonable measurements for physical
more scattered and a little less advanced. work can be made in terms of calorie
In the USA, particularly under the impetus of consumption, and on the practical side, it is
military funds, development has been much possible from this knowledge to calculate the
more rapid. At least 110 industrial companies organization of work and rest pauses to
have organized human factors programmes with minimize fatigue, etc. LEHMANN (1958) has
an average group staff of ten, and the Services- shown how 'hidden' rest pauses may often
sponsored units and the academic research have to be taken during heavy work and how it
groups are similarly more numerous. It is is more efficient to organize official pauses
difficult to learn what may be happening in the realistically.
USSR, but from a Iiterature survey, RONCO Psychology involves all areas of behaviour
and SAWYER (1962) concluded that 'although and performance, particularly the mental and
only a few articles were found that have to do emotional aspects. Two broad sections must be
with the direct application of knowledge of distinguished, personnel psychology and
man's psycho-physiological capacities to the engineering psychology. Personnel psychology
design of machines, .... sufficient volume of relates for instance to selection, training,
pertinent material is being generated to merit moti vation and job satisfaction, whereas
systematic attention by human engineers'. engineering psychology is concerned with
3
perception, decision-making, and particularly for having any men in the system, and then to
such aspects as how an operator recei ves and examine and decide which sub-tasks within
processes 'information'. From this one can the whole system should be assigned to human
estimate, for instance, the rate at which the elements and which to machine elements.
operator can be gi ven information by a system When considering in turn whether the use of
and act upon it, what delays and errors may man or machine is more appropriate for a sub-
result, and how best to design the input and unit, such factors as cost, weight, size,
output equipment leading to and from him. reliability, safety and efficiency, must be
CONRAD (1960), in describing some of his assessed and compared for each sub-task
research related to post office telephone and separately, and then an optimum balance must
letter-sorting problems, illustrates well this be decided for the combination of sub-units
part of applied psychology. into the complete system.
It should also be emphasized how workers Secondly, for each machine or for each
in ergonomics are aware of the future problems part of a system where the human element is
likely to arise from advances in technology. used, the interaction between the man and the
As long ago as 1955, MACKWORTH discussed equipment must be optimized. To do this, an
the relevance of research then to the problems approach is used which is slightly different
expected when automation came; and with from that of the engineer, who rightly must
automation still hardly arrived WELFORD start his thinking from the machine and
(1960) could survey the aspects of ergonomics concentrate upon that. The ergonomics approach
likely to bear on the design of automatic at this man-machine level, is to examine the
equipment and the human problems likely to task and the operational sequence that the man
arise from automation. will have to do and to consider the man's
abilities and limitations and how they will
influence the total task performance(Table 7.1).
What is the Approach of Ergonomics
to Practical Problems?
With the man as the centre (Figure 7.1) it is
The primary emphasis is upon a scientific possible to work outwards from him, thus
approach, i.e. upon objective study and the
gathering ofreliable facts from which deductions
and recommendations can be made and proved
statistically to have an adequate measure of
validity. Opinionati ve advice and solution are
avoided unless specifically requested in
preference to no advice at all, and only then
with great emphasis upon the unproven and
speculative nature of such recommendations.
In tackling specific problems to fulfil
the general aim of designing well-integrated
man-machine combinations, the ergonomics
approach has three definite sections, some or Figure 7.1. The field of study of
all of which are implemented, according to ergonomics divided into four parts
the type and complexity of task, machine or
system being considered. firstly examining his interaction with the
The first process, more significant with machine or task (Figure 7.2); this involves
major systems than with individual machines, studying the operational sequence required of
is to define the system goal and the reasons the man in terms of information going into him
Table 7.1. The Sector .4pproach m F.rgonomics
I Man
Consideration of:
Sex Physique Training

Age Intelligence Motivation
Size Experience
Definition of operational modes required in final situation,e. g.
Searching Monitaring
Tracking Decision-making
Thus consideration of abilities and limitations of human Operator for all

aspects of the task.
ll Man-Machine Interaction
Influence on Operator and his decisions of:
Displays - sensory input to operator

Controls - motor output from Operator
Panel layouts
Display-control comp,atibility
m Man-Workspace Interaction
Influence on position, posture and reach of:
Machine size
Chairs, desks, etc.
Adjacent machines, structures and material, etc.
IV Man-Environment Interaction
Influence upon behaviour and performance of:
Physical aspects
Chemical aspects
Biological aspects
Psychological aspects
Physical - light and colour, noise, heat, ventilation, gravity,

movement, electromagnetic and nuclear radiation.
Chemical - gas or liquid, composition, pressure, smell.
Biological - microbes, insects, animals.
Psychological - workteam, command structure, pay and welfare, shift

conditions, discomfort or risk, socio-psychological
aspects of the particular factory, neighbourhood,
town and type of industry concemed.
for decision, his actions to signal his and (iii) the general environment. Only in this
decisions, and the compatibility of layouts of way can it be certain that all possible
panels and equipment with the way he does his influences upon the man and his work are
work. Next is studied his interaction with the taken into account.
immediate workspace around him (Figure 7.3);
Werkspace
the size and position of the chairs, desks, Machone s1ze.
machine console, etc., influence the operator's chairs. desks,t>lc
position, posture and reach, and thus his

comfort and efficiency. Finally comes the
general environment (Figure 7.4) in which he
and the machine are at work, including such
Man
Dec•s•ons
Figure 7.3. Man-workspace interaction
Thirdly, even if redesigning an existing

working situation, and especially if involved in
the design of a completely new machine system,
the proposed and agreed final design should be
evaluated by mock-ups and trials to test the
validity of decisions on human factor aspects,
Mach1ne in exactly the same way that models and
Panel Iayouts
o.splay-control compatibility
trials are used to check important engineering
sections of any system. Such evaluation trials
should particularly use samples of the expected
final operators.
Env1ronment
Phys•cal-l•ght, heat. vent•lat1on, no1se, t>lc.
Psycholog•cal- workmates, management, t>lc
Figure 7.2. Man-machine interaction
problems as heating, lighting, ventilation, and

the social structure of the group, team or
factory of men in which he is; many different
fields of science are irtvolved in studying the
various aspects of this section. Figure 7.4. Man-environment interaction
To ensure an adequate study and reliable
recommendations it is important to consider in The importance of this threefold approach
turn each one of these three sections of lies in three aspects: ( a) it emphasizes the
interaction working outwards from the man: importance of defining the purpose of the
(i) the machine, (ii) the immediate workspace, design process by concentrating first on a
consideration of the design problems at a But as man in the future is progressi vely used
systems analysis level; (b) it ensures less for bis muscle power and more and more
comprehensive gathering of all relevant data for his ability to process information and make
concernin g the particular problern situation, decisions, so he is in control of more assets
and greatly diminishes the risk of wasted time and more output, and the cost of the mismatch
and inadequate solution by premature grows in proportion. As with military problems
concentration on the apparent but not real in the past particularly, so now with modern
cause of a problem; (c) it emphasizes that industrial equipment, the demand upon the
working situations are dynamic: therefore, to operator grows as the complexity grows. The
be successful, anyattempt to solve the problems more the operator is stressed, the greater is
of these situations must itself be dynamic, the need to ensure a good match between man
must study the man-machine work as a series and machine to minimize the risk of error and
of actions and interactions, and must maximize accuracy and output. This is the
fundamental reason why the specialist
visualize it all as a three-dimensio nal eine
knowledge from ergonomics is needed in modern
film rather than as a flat drawing and static
industry and particularly during the process
blueprint.
of design.
Why is Ergonomics Needed?
This question would not be asked if the Where i s Ergonomics Relevant
in Modern lndustry?
subject of ergonomics were not mainly human.
On the shop floor there are many situations
But because everybody is human, each person
where scientifically proven methods of
tends to think that he automatically knows most
selecting and training operators would yield
of what needs to be known about man when
matehing up a working situation to him. worthwhile economic improvement. Organ-
Whereas in fact the amount of specialist izationally, it would seem that these situations
knowledge now available about man is such can best be dealt with by personnel and work
that there is probably no one man, even a study departments making use of technical
specialist himself, who can learn it and know data and consultant advice from ergonomics
it all. If a problern arises which is primarily where relevant. In many companies it may not
in, say, chemistry or quality control, a person be good economics to employ full-time
quickly acknowledges if it is outside the ergonomics specialist for this work alone.
limits of his own general knowledge and calls When new machines and task situations
upon the ehernist or quality control specialist are beingdesigned or considered for installation,
for assistance. But if he has a problern to deal ergonomics becomes more particularly relevant,
with involving humans, he still tends to use and the cost of obtaining such knowledge,
bis own subjective opinion or go to consult the especially if it is sought early enough in the
subjective opinions of other men instead of design stage, becomes relati vely insignificant
summoning a specialist, perhaps because the in proportion to the typical capital cost
existence of specialists is not yet widely involved and the potential savings to be
known. expected.
Accidents, errors, poor quality and low The same arguments apply with much
output are the usual symptoms of a problern greater force still as complexity increases and
which amongst other things, may need ergonomic the design of larger production units and
attention. When a man-machine mismatch in systems is considered. A continuous flow
industry today results in such symptoms, the production line is a typical example of a
cost of not finding a eure is often considerable modern industrial situation requumg some
but may not be so high as to compel action. thorough system design. The bigger such a
56 THE DESWN METHOD
line is the greater is the importance of adequate manufacturers of electrical generators, tyres,
feedback and feedforward of information from submarine cables and water tanks. Again, on
one stage to another to maintain everything the question of ergonomics applied to
within limits; but very frequently one finds production methods, manufacturers with a high
that still only the most primitive means of ratio of labour to other costs (such as in the
communication are available between operators, boot and shoe industry), with special
who are now usually at much greater distances environmental problems (such as in steel or
from each other. Only thorough and extensive tyre making), or with important safety
study of what information each operator needs considerations (such as passenger transport
from the machine and each other, and of the services), can expect significant improvements
way that the system should be designed to in efficiency and safety by applying ergonomics
present this information most simply and knowledge.
easily, can safeguard against lengthy Given a prima facie field of application,
commissioning time and the risk of costly the scale of ergonomic activity required and the
shut-downs. The emphasis of the ergonomic manner of obtaining it will then depend on the
approach on dynamic rather than static size and tumover of the company. In general,
investigation and planning could be very it would seem that only large companies, say
fruitful here. of more than one or two thousand employees,
or those wi th a large turnover, say of more
than one or two million pounds, are likely to
What Would be Correct Management Policy? feel justified in employing a full-time specialist
Only broad generalizations are possible, on their staff. Most managements would be
because so much depends upon the nature of a better served by seeking consultant advice
company's product and organization. The when appropriate. This, however, points to the
manufacturers of complex capital goods are need for one important action. When a decision
more likely to find ergonomics help of practical has been taken to seek ergonomic advice only
and economic value than are manufacturers of when required, it is essential to establish,
simple products. On the other hand, any within the normal organization of the company,
manufacturer contemplating major capital a clear definition of who is responsible for
expenditure on new plant or new production watehing over the general running of the
facilities, particularly envisaging a major company to detect when and where ergonomic
increase in the complexity of his manufacturing assistance would be economically advisable.
processes, might well save much trouble later Some one or more persans should be gi ven the
by getting ergonomic assistance early enough. necessary training and charged with the
In general, the best criterion is probably responsibility of watehing for operationally
the financial size of the company compounded significant ergonomic problems and, when they
with the amount of man-machine conjoint arise, with the duty of seeking the specialist
working time involved, either (i) in the advice required. In various industries and
consumer's use of the product, or (ii) on the companies this function is carriedoutsometimes
production floor, according to which aspect of by the medical officer, sometimes by the work
ergonomics application is being considered. study department, sometimes by the production
For instance, on the question of ergonomics manager, and sometimes by the chief engineer
applied to the design of the company 's products, or his staff.
manufacturers of machine-tools, motor cars, Therefore, in relation to ergonomics, a
furniture and hausehold appliances have much management decision is required, in a company
greater need for ergonomics knowledge than of any size, upon the following three points:
(1) What areas of ourfactory and/or products because man is so adaptable that the designer
might benefit from the application of ergonomic is often able to hide the shortcomings of his
knowledge? machine design behind the skill of the operator,
(2) Does our requirement justify a full-time who is rightly proud of that fact. But the more
specialist or a consultant called in as skill required and the less well-trained the
necessary? operator, the more easily he may break down
(3) If a consultant is to be used at any under stress and make an expensive error.
time, what department or person shall be This is not his fault, and the task of
charged with the duty ofwatching for significant ergonomics along with the designer is to
ergonomic problems and calling in a consultant simplify the machine design enough and train
in good time? the operator enough to remove the risk of
error but leave the pride of skill.
Conclusion
There are many situations and equipments
where ergonomics could greatly improve the Acknowledgment
operational usage and minimize the waste and I wish to thank the Directors of EMI
'down-time' which are so often dismissed at Electronics Ltd. for permission to prepare
present as operator error or ignorance. It is and publish this chapter.
Chapter 8
SOCIOLOGY AND DESIGN
A. M. Penny
lntroduction these different problern areas would fall within

Sociology and design can cover two the competency of sociologists proper, social
separate kinds of social situation: the situation psychologists, anthropologists both cultural
of designing and the people concemed with it and social, and of economists. Forthis chapter
and planning for it; and the situations for I am taking sociology to include all the fields
which designs are carried out - what might of these people except the economists.
be called the user requirements area. I shall Since, in addition, designing is an
try to show how different relationships work by individual business to a large degree in
referring later to an example of the briefing solving problems and in the matter of design
of an architect of a large teaching hospital, judgment, I think that an important group of
in which I had a hand. This will show some of questions is that of the relationship between
the kinds of problern which occur, though there social and individual aspects of designing.
will be some, such as the way fashions are
initiated and catch on, which I shall not The 'Two-culture' Problem
touch on. As Sir Charles (now Lord) Snow described
Perhaps the basic type of problern of this problem, it was the failure of scientists
sociology in relationship to design, or rather and non-scientists (by which he referred mainly
of sociologists to designers, is the type which to people concerned with literature) to
would now be called 'two-culture'. This kind communicate, and with not wanting to do so.
of problern applies also to the relationships In the design field also this kind of non-
of designers with their clients and with other communication is frequent. For example, if the
people of non-design backgrounds working architect of a science laboratory is given
alongside them in planning teams. This has briefing instructions to which he works, it may
the rather odd consequence that the relationship happen that he cannot check that he has
of sociologists (and other scientists) to understood the needs of the laboratory users as
designers does itself present a second Ievel the drawings proceed. If this occurs, as it has
of sociological problems. been known to do in some university
There is one further consequence of the departments, the plans will not meet the
form which sociological problems of design requirements. If, on the other hand, there is a
take: it is that designers and their non-design regular dialogue between the architect and
colleagues not only relate to an organization his dient, the final result will be a collaborati ve
in which they work. They also relate to each effort: the client will work out his needs
other and to each other's ideas. The systematically in a way the architect can use
sociologists study the functional situations and probably with the advice of the architect.
for which designs are to be prepared. In terms The architect's design proposal can then be
of the way disciplines are usually thought of, checked regularly with the user so that mistakes
59
60 THE DE~GN METHOD
can be eliminated before too much work has The Case of Science and Design
been carried out on mistaken assumptions. To the question: 'Should science and
In this case there is clear understanding how design be mixed?' it is possible to give
the dient and the designer play their separate answers which are for, against, or uncommitted.
roles and how these roles are interdependent. I am only dealing with the answers for, because
These two elements, the separation of the arguments about the other answers do not
roles and the way they interact, can be analysed come into the scope of this chapter. The case
in more detail in terms of their nature and in favour of mixing science and design
possible solutions to the problems they raise. presupposes the possibility of doing so. I
think that it is possible but I do not think that
I shall discuss first of all the case of
the proponents have made explicit how this
architects and scientists working together,
should be clone. For example, LORD
partly because I have had experience of this
LLEWEL YN DAVIES, in his 1965 Design
kind of relationship, and partly because it is
Oration to the Society of Industrial Artists and
becoming more and more important.
Designers, said:
Many difficulties of the two-culture sort
arise just from lack of understanding of their 'Today in any field of design the range of
nature. They are called difficulties of technical knowledge which the designer must
communication but the first important have at his command is tremendous. Knowledge
distinction is not made as to where the is also continually expanding and changing, so
difficulties of communication lie. Are they in it is impossible to teach technical know-how
the way the communication is made? Or do they effectively in a school of design. Anything
arise from what is being communicated? The that is taught may be out of date by the time
distinction is between the techniques of the student is in a position to use it. Instead
signaHing (even if it is just putting a message teaching must be directed towards the
into clear language) and what it is that is in scientific basis on which technology is
the signal, its content. founded .... '
lt is my contention that the problems are
those of content rather than of techniques, It would be very unfair to take the
though lucidity of communication is essential. statement of the last sentence as it stands,
lt is a further part of my thesis that the kind without referring to the teaching methods at
of understanding which can be reached between the Bartlett School of Architecture which have
people from different backgrounds is of the changed the situation there in so short a time.
same kind as one which has been well As I understand it, the students learn about
understood for half a century. I am referring the natural and social sciences because the
to the kind of understanding which an planning of buildings at the present time has
anthropologist reaches in a society foreign to to be based upon reliable information, and the
his own by coming to understand the poin.ts of methods of the different sciences provide the
view of the people he is studying. This is an best chance of getting this reliabili ty. This
essential part of the fieldwork method which approach seems to me to be right to a radical
was first developed by MALINOWSKI (1922); degree. But I do not think the statement
the process is the commonsense one (also used 'technology is founded on a scientific basis'
by Sherlock Holmes) of the anthropologist either shows how to solve the two-culture
putting hirnself in the position of his problern of science and design, nor does it
respondents, as far as he can. lt is called show the full extent to which results may be
'observer-participation '. carried on solution.
My Argument The first two paraphrases are different in kind

I think it is easy to show that science and from the third because of the differences which
design (and technology, for the same reasons) can occur in the application of rational
are different from each other as acti vities. criticism and scientific criticism. Criticism of
Science is concerned with finding out the a scientific theory includes rational tests to
truth, with discovering regularities in nature see whether it is internally consistent, whether
and explaining them. Designing (and technology) one part follows validly from another and so on,
is to do with making things which will serve but also includes empirical testing in an
a certain purpose for their users. attempt to falsify the theory. Rational criticism
The question a scientist asks about his can be applied to situations and cases which
discoveries and theories is: 'Is it true or do not have the characteristics of a scientific
false?' The kind of test he uses is that of theory, for example the truth of a historical
falsification. The question a designer asks statement, or (relevantly to the present
about his invention is: 'Does it work?' The discussion) the rationality of a planning or
tests he applies are to confirm that it will work. design decision.
The problems of the scientist are those of Students' training based on the third
describing, explaining and discovery. Those paraphrase would be far broader than that
of the designer are of inventing artefacts based on the first two, because rational
which are fit for their purpose, whether this is criticism can be applied to discussion of
functional or aesthetic. architectural and design topics which could
These contrasts leadback to the statement: not in any sense be called scientific. An issue
'Technology is founded on a scientific basis.' of this kind might be whether a particular
Two questions can be asked: 'How can theory of design had a rational relevance to a
technology be based upon something which is particular design situation, or only an aesthetic
different from it in problems, values, kinds of one. The conclusions in either case would be
solution and methods of testing?' and- 'In what quite different. (It is a matter of common
sense (from at least three) is technology experience that some design students argue
founded upon a scientific basis?' I shall from an aesthetic statement as though it were
assume a positive answer to the first question rational or factual.)
to provide the basis for the second. The three This argument, I think, shows that design
possible senses which I can see for the and science can be different kinds of activity
statement are the following: yet still employ rational criticism. The more
local case of the kind of support scientific
(1) Technology uses knowledge turned up studies of a sociological nature can give in
by scientific methods, and to make reliable design is now to be dealt with.
planning forecasts gathers information by
research methods.
Forecasts
(2) The second meaning includes the first
When scientific investigations are carried
with the addition that technology includes
out to provide 'hard' information in support of
management by scientific business methods.
policy decisions, such as those taken in
(3) Open rational criticism which forms planning a building, the people would probably
the basis of scientific method can also be used think of the information in terms of research
on the rational levels of design and technology findings or scientific facts, or of
and has been the reason for modern technology recommendations based upon them. I think this
being what it is. is a mistake as I shall try to show by an
example. This concerns the amount of storage to say that the sun will rise tomorrow at such
space needed in a teaching hospital. and such a time precisely, by deduction from
I had at one point to find some information Newton's Laws or their successors.
abou t just this problem. I had to provide the I think it is important that designers,
planning team of the hospital for which I was planners, and scientists make use of forecasts
working with this information in terms of so or predictions, because it is with the testing
many square feet per bed. The simple method I of forecasts that scientists can help, thereby
used to start with was to find out the areas validating information on which planning and
scheduled for other comparable hospitals. design decisions are based. I think that it is
I found the areas for three others. The figure necessary to point out that some forecasts
for each was about fifteen square feet per bed. marle by designers are in the nature of untested
It seemed to be reasonable that hospitals hunches. A forecast, therefore, may be tested
were more or less alike in what they stored or untested. Forecasting is still a human
and in the area this took up in relation to their activity used by designers, planners and
size. It seemed that this figure of fifteen scientists, a common highway which can be
square feet per bed would be big enough. used for different ends and with different
It is already clear, I think, that I was not criteria of dependability. It makes a second way
dealing with random facts but with regularities - in which design and science can share
the more or less constant ratios of areas to something without any claim being marle that
numbers of beds. If there had been no they are the same in kind.
regularity the figures would have been useless
as such. I was making a forecast based on the
assumption that these regularities would apply The Two-culture Problem as a
to another hospital. Sociological lssue
I wished to strengthen the basis of my My argument is that there are two stages
forecast, and I was lucky to find a hospital in this sociological problem, using my broad
store which had been purpose-built in the last meaning of the term to include cultural and
few years. The ratio of area to number of beds anthropological problems. Firstly, scientific
was significantly smaller in this hospital. and designing thinking are cultural in the
In this case the kind of deduction I could anthropological sense in that they have a
formulate was this: tradition and certain conventions and methods
Three hospitals have scheduled their which are handed on from one person to another.
storage accommodation as fifteen square feet My second point is that the kind of logical
per bed. One other hospital has a ration less criticism I have been using is part of the
than these. The hospital I was working for was tradition of science, including sociology.
organized on broadly similar lines to the My argument has been about different ways of
others, so that its needs were likely to be thinking and the way people who normally do
similar. The deduction that I drew was that one or the other kind interact. This falls
fifteen square feet per bed was likely to be quite clearly into the definition of sociology -
more than enough for present or foreseeable the rational study of social relationships.
needs.
The forecast based upon the regularities
alone was of a different kind from the second. Example - a Hospital Briefing
It was of the kind: 'The sun rises every day, When it is decided to put up a new building,
it is likely to do so tomorrow.' The second such as a hospital, to meet a need which is
forecast was of the kind that makes it possible either not met or not met well enough, the
following stages occur, from the realization of research is better than treatment based on
the need to the occupation of the new building: Superstition.
(3) The intention comes into relationship
(1) Realization of need and decision to with the si tuation stated previously, leading
meet it by the responsible organization to certain implications if the intention is to
(committee, board of governors, local council). be realized.
(2) Setting up of the executive team to find ( 4) The decisions are rational to the
out the user needs of the building and to deal extent that they have been exposed to rational
with the people who will design and put up criticism. If the need for a new hospital could
the building. be shown to be non-existent, then pushing
(3) Collection of information about user ahead with it in spite of this would not be a
methods. rational act.
(4) Formulation of the 'brief' as a whole, (5) There also appears the important
probably with the collaboration of the architect. formulation of BARTLEY (1962) that there
(5) Various stages of designing to the brief. cannot be any rational justification of a case;
(6) Going to tender. a rational argument is one that can be, and
(7) Construction. has been, exposed to criticism.
In this range of activities and events there is Setting-up of the F:xecutive Team
a whole range of social problems. I shall deal In order that the decision of the initiating
with (1) to (3), since I know them best. body to build can be realized a whole series of
decisions have to be taken. Decisions have
Realization of Need and Decision to Meet lt both a content and a ratification and both of
With a teaching hospital in present-day these need people. The ratification can probably
conditions the need is usually expressed in be provided by the governing board of the
the following situational terms: hospital, but it will need a planning team to
formulate the content of decisions, to collect
(1) There is an increase in population, so information for them and to deal with the
that the present hospital is overcrowded. different interests involved. The following
(2) There is need for more space for factors have shown themselves to be important
research and the present facilities do not give in the appointment of planning teams:
any more scope for rebuilding.
(3) Departments have become scattered (1) They should be mixed, having medical,
away from the main hospital. nursing, and administrative members, who
should be able to understand the points of view
On this account it is decided to rebuild the of future users of the projected hospital, so
hospital already in existence. This simple that they can offer sensible criticism of the
example shows the following features: opinions of need.
(2) The architects and engineers should be
(1) There is an intention to provide brought in at an early stage so that they can
treatment and research and evidence of a learn what is going on and help in the
present shortage of proper facilities. formulation of the briefing instructions.
(2) This intention stems back to more than (3) The planning team should have a close
one set of values. One holds that it is good to internal relationship, involving both seeing
treat people who are ill. Another holds that each other's point of view and having the
the modern way of treatment based upon chance of discussing and criticizing those of
anyone else. This shows both two-culture The Overall Pattern

situations and the open rational criticism I hope that some of the sociological
already mentioned. problems inherent in the design of hospitals
( 4) The planning team has a special formal have shown themselves. These can be listed as:
relationship to its superior body. There is the
(1) There is the question of values and
relationship of a team which finds and
intentions in relation to a known situation
formulates the content of the decisions and a
leading to certain implications. This is a
superior board which ratifies them.
formulation of POPPER, in his Poverty of
(5) In the planning team itself there are
Historicism.
formal and informal roles so that informal
(2) There are prob 1 e m s of formal
discussions become formalized into minutes
organizations and their relationships to each
and recommendations.
other, political, social, and in terms of ideas.
(6) A good planning team can resist
(3) There are problems of the two-culture
pressures from outside and act as a buffer
type caused by people from different
between the potential users and the architect.
backgrounds working together.
(7) Eventually the planning team will
( 4) There are issues in which forecasts
produce the instructions from which the
are arrived at on the basis of non-existent
architect can work.
information, or information which is scanty or
lacking comprehensive test.
(5) Problems become formulated in a
Obtaining the Information multiple way, resulting from the successful
This is the stage in which the activities solving of communication difficulties between
of a hospital are assessed and a projection the disciplines of the different members.
made for the new one. The obtaining of (6) There are problems of traditions in
information means the interviewing of present medicine and science which lead to the
users who carry on their respecti ve acti vi ties adoption of particular forms of building, or to
in their present departments. The information their rejection in favour of some alternative.
will be in the form of forecasts as in the case (7) There are questions of interaction
quoted earlier. between one phase of preparing to build and
The information obtained is used to the next, in terms of how it should be clone and
estimate the terms of future needs and the managed, who should carry it out, and the
policies selected to meet them: how many passing of clear instructions, all based as far
rooms will be needed with their size in each as possible on rational decisions.
department; what the overall siting relationships (8) There are problems of people from
should be. Arguments will be given for all the different professions trying not to do each
decisions and their rationality will depend other's jobs but understanding them sufficiently
upon their standing up to discussion. to fulfil a complementary role.
Chapter 9
DESIGN ACCEPTANCE
M. E. Peplow
lntroduction staff. The department serves, in a particular

Whereas in the past one has been region, a large number of dispersed
concerned with whether design effort has been specialized production units and aims to:
aimed at the most fruitful targets, today (a) examine plant processes, for the purpose
there is a public awareness that a valuable of providing new knowledge upon which to
national resource is not infrequently directed base new or improved designs; (b) design and
at wrong targets. This subject is often develop process equipment and techniques;
discussed but one usually lacks a perspective ( c) sol ve certain operating problems; ( d) provide
in terms of the frequency of various types a specialized test service.
of failure and the results of trials of methods Approximately 90 jobs have been
of selection of design projects. completed in the creati ve categories ( a) to ( c)
The present chapter therefore attempts and a similar number in the test category.
the limited task of considering the reasons
There are also 90 current jobs, over half of
for success or failure of approximately 90
which have been acti vely pursued for more
jobs; the jobs being aimed at the design of
than a year. The jobs were started by
process equipment and techniques or the
'customer' request; by R & D department
solution of allied problems. Using the elements
initiative; or by suggestions from various
of behavioural psychology as an approach to
HQ officers. The outcome of the 94 creative
the customer's system, methods are then
jobs, in terms of implementation, is as follows:
suggested for increasing the probability of
design acceptance. 48 successful - accepted by customers.
Ideas for conducting controlled experiments 8 successful - equipment for R & D
with such methods are now required and may department's own use.
well follow the pioneering efforts of
8 negative - i.e. a current theory or
CHURCHMAN (1961) and HALL (1965). Any design concept disproved.
results thus obtained might then be co-
12 partly successful - i.e. partly failed
ordinated by a theory of adoption and diffusion or a slow adoption.
of innovations, such as that recently 18 failed.
propounded by ROGERS (1962).
This failure rate lines up with that reported
Outcome of Completed Equipment by the 1960 Federetion of British Industries
and Process Design Jobs Survey but is much less than private report
A survey was recently marle of the work gives for consumer goods in the USA (Booz,
of a research, design and development (R & D) ALLEN and HAMILTON). lt was found that a
department that was formed six years ago failure rate, for process development and new
with five professionals and now has 45 products, of one-sixth was 'remarkably con-
professionals and the same number of support stant throughout the whole of industry'.
65
The present failures lie more with the In respect of the latter, some failures could
basic jobs started by R & D initiative. Advice have been avoided if other engineers and
was of course sought from senior people scientists had been consulted on the possible
whose judgment was based on observation boundaries to the problern or used to ensure
and past experience of customer operations. that a traditional approach was not used
However, not only do judged needs sometimes inappropriately for a new type of problem.
differ from actual needs, but the latter can Designs were mainly inadequate because
change with time. On the positive side, it has they provided insufficient improvement on
been found that 10 per cent of all the jobs existing designs and inexperience contributed
have gi ven useful by-products. Thirty of the to this. Occasionally inefficiency. unreliability,
successfully implemented jobs were customer's complexity or oversize were factors. The
direct requests. The remaining 18 were unscientific approach entailed:
characterized by:
Lack of initial applied research
5 sol ved a customer's recognized problem.
(unverified or unanalysed ideas and under-
1 helped a customer's own development
estimate or disregard of basic parameters).
project.
Lack of accelerated life tests.
3 gave a good cost or time saving (but
Basic knowledge lagging too far behind
not all such jobs were implemented).
the technology.
6 were implemented at a future date,
Lack of perseverance with a particular
when a severe operational problern occurred.
concept.
3 were the subject of continuous sales
pressure. Although many design problems will appear to
A similar analysis of the 30 part or complete be solvable by means of an existing approach
failures with respect to implementation gives: and no-one wishes to spend time on unnecessary
research, it is suggested that the decision to
7 were technical failures.
follow the trial and error trail should only be
8 were economic failures.
made after unhurried deliberation.
15 failed due to personality factors.
Discussions with other engineers and
scientists have indicated other causes of
Design Failures on Technical Grounds
technical failures. A list is given in Appendix
The technical reasons for the seven failures
A at the end of this chapter.
were:
Inadequate ini tial specification. Design Failures on Economic Grounds
Inadequate design. The main cause of failure here was a
Unscientific approach. change in the commercial environment during
Insufficient study of existing designs. the period of research, design and development;
Problem unsolved within set man-time particularly if the job had been a prolonged
resources. one. Changes in the customer's Operations,
maintenance or material supplies and the
The consequences of an inadequate
advent of competitive designs are also
specification, or one not agreed wi th
involved. Forward looking economic feasibility
representati ve customers, were:
studies and continuous contact with customers
Inflexible (or even unsuitable) design. might keep the amount of abortive work to a
Overcomplicated design. minimum. It is particularly important to give
Design failed to cope with a critical the customer an idea of the cost of the intended
factor. equipment at the earliest date.
Sometimes a competitor's design was Customer's interests change during the

found to be better or cheaper, particularly if a design period.
development had been started without a Customer prefers to await a universal
'brainstorming' or other search for a variety of solution to his problems.
initial design concepts. Even in the absence Reversal of a particular policy with a
of a competitor, the development or subsequent change of management.
production cost of a device could be out of Suspicion of a process only verified on
balance with the value of it to customers. model scale.
lmprovements in estimating, based on planned
Organizational structure incompatible
development, therefore seem necessary.
with innovation (CARTER and WILLIAMS,
Where a design is to be put into production 1959; BURNS and STALKER, 1961).
by the customer, it may be essential to study Duplication arising from personal
the latter's internal organization before competition within an organization.
deciding which of the possible designs can Design limited by customer's attempts
be economically handled. The alternative can to reduce the cost.
be a considerable waste of effort on a
Designer fails to obtain or keep the
specific design.
customer's support for his ideas.
Some other causes of economic failure
are gi ven in Appendix A. Designer fails to persuade the customer
that risk involved in design acceptance is
justified by the potential benefits.
Design Failures due to Personality Problems Customer unwilling to change his
Half of the failures lay in this area. The procedures or retrain his operators.
main cause of failure was apparently due to Prestige of people who must advise or
the customer being too busy to implement a agree to a design is involved. (Sometimes
plant or other improvement, even though he the 'not invented here' attitude.)
accepted its value. R.ecognizing that everyone
has his priorities, consultations are needed
before ernbarlang on proposed research ot Tlle Designer-Customer Relationship
design work. However, it may sometimes be The last four items of the abeve Iist are
possible that the p r i o r i t i es of those examples of a failure in the designer-customer
responsible for design implementation are out relationship or failure to keep in mind that, for
of step with higher management policy, due to design as for beauty, 'the value lies in the eye
misinterpretation or the fact that a rewards of the beholder'. To probe this, one can
system cannot line up perfectly with a policy consider the psychologists' concept of basic
system. For example, there may be a policy needs which humans strive to satisfy, directly
of keeping production going and improving the or indirectly (MASLOW, 1962). The basic
process or product; whereas career rewards needs are usually listed as:
may appear less linked to improvements, even
when these can be detected in trends or Physiological - air, food, etc.
framed by targets or competition. There rttay Security - orderly, predictable world;
even be a misconceived fear that an improve- freedom from worry.
ment venture which fails will interfere with Emotional security - belongingness.
career progression. Other persortali ty Self respect.
difficulties found in the way of design Others' respect- status.
acceptance were: Self expression - self actualization.
68 THE DE~GN METHOD
The last two needs are difficult to satisfy fully (3) Is he conservative or radical?
and so are the ones to which a new design can (4) Does he fear failure to cope with
be addressed. It should be noted that self change?
expression includes gatmng a personal
This method of tracking down customer's
understanding of the mental, emotional and
needs is probably unnecessary to those with
concrete constructs of others as well as
'social intelligence' or an intuitive knowledge
personal creation of these. Financial rewards
of the likely behaviour of others. On the other
can, of course, lead to the purchase of goods
hand, some of us need, in addition, to know
and services which help to satisfy some of
from the psychologist the type of indirect
the basic human needs. And job easement
question which will yield the clues to a man's
can provide the time for obtaining the
basic needs and interests. An alternative
satisfaction of needs.
approach is to show, in a conspicuous manner,
Analysing further, one notices that an
that a new design can contribute, significantly
individual may seek self expression, etc. in
and in a reasonable time, not only to financial
one or more of about eight main fields of
success and job-easement but also to status
interest (ALLPORT, VERNON and LINDZEY,
and self expression in the fields of practicali ty,
1931). Relevant to design acceptance are the
social usefulness, ingenuity and harmony: at
following:
the same time too conservative or radical a
Economics - production of goods and
tone should be avoided. The lower the level
services. of a customer, the shorter is the time he is
Political - power, responsibility. likely to accept as reasonable and lhe more
Social - helping others. predictable must be the consequences of any
Theoretical - rational understanding. change. Thus slowly maturing investments
Concrete construction. should only be recommended to a high level.
Aesthetic. Although the need for belongingness may
The individual personality may of course see encourage lay customers to follow a fashion,
the field in a conservati ve or a radical light. in respect of technical people it usually arises
as a need for maintaining a particular group
The Customer culture; leading to the granting of mutual
In the customer's eyes, a new design is favours as well as an attitude to innovation
acceptable if, in relation to the effort required in line with group norms.
of him, it fills one of his needs previously The 'not invented here' attitude, which
unfilled or incompletely satisfied. However, often takes the form of finding a possible
he may not realize that there is an aspect of snag in an affered design, can be understood
the design which could meet such a need. The in the above terms. Acceptance of a design
designer or his agent must therefore get to may be perceived as a deterioration in relative
know the customer's needs so that, through status, outbalancing the gains which
a detailed knowledge of the design, the most acceptance could bring. It may also interfere
relevant aspect can be identified and drawn with doing a job 'in one's own way'. Equally
to the customer's notice. The questions then understandable is the difficulty of gaining an
become: acceptance of a rational proof that the
(1) Does he seek one or more of status, customer's contrary proposition not only has
self expression, job-easement, financial reward? snags but is bound to fail.
(2) Has he a strong interest in the A brief indication of the relation of
economics, poli tics, social, constructional, customer to new design adoption, in terms of
theoretical or aesthetic field? Rogers' ideas, is given in Appendix B.
The Customer Without Needs - Whereas a Although the R & D personnel perceive their
man dissatisfied with his plant is likely to role as equal service to all the other branches,
welcome help in his creati ve endeavours to the two branches making by far the most
improve it, the satisfied man is usually a demands on R & D are the two that returned
poor customer. As one of the greatest men of the biggest proportion of their questionnaires.
our profession, Edison, wrote: 'show me a Particular features of communications with
thoroughly satisfied man and I will show you these two branches are: (a) one branch is
a failure' . Since progress entails risk-taking, geographically much nearer than all the others;
one might also say: 'show me an unventuresome (b) R & D people are frequently at the factories
man and I will show you a failure' . (the other 'branch ') since the latter provide
facilities for R & D fieldwork on local and
Customer Survey general problems.
Where reports of new process equipment, The reactions to the reports were:
techniques and principles are circulated to
(1) Reports from R & D encouraged.
potential users, it is possible to invite
(2) A few felt that presentation could be
comments and ask questions such as:
improved by emphasis on illustrations, summary
(1) Is the subject of the report relevant data sheets and examples of practical
to your problems? applications.
(2) Is the presentation suitable? (3) Other than where the report met a
(3) Has sufficient work been clone? specific request, the belief was expressed
(4) Are the results of use now or possibly that it was helpful in small ways and could be
in the future? of benefit to others, rather than a stated
Even though it was realized that some of the intention to act.
answers received would be guarded opinions, These comments show a sympathy to R & D
each of 45 potential customers were sent a but underline the need for adequate customer
consultation before investing much effort,
f<'~""' -~~'"' •oO •• ,0 0HOOO ""' and the need to present results in terms that
are both easily understandable by customers
f PIconstruction
ant desogn and
(80)
.l
3 Operatoons H Q (10)
and couched in terms of their interests.
The poor response to the questionnaire
was associated, in the case of one branch, with
the difficulty of dealing with an over-abundant
supply of information. Conversely, a special
effort to disseminate Ieaflets on new equipment
to the user Ievel was frustrated by filtration,
at a higher Ievel, that was no dnubt intended
Figure 9.1. Numerical distribution of
for saving the reading time of others. The
questionnaires and replies
exercise of local imagination in the light of
questionnaire in respect of one of several detailed local knowledge was thus made
recent reports. The simplified chart in Figure impossible. Although one is drawn to
9.1 shows the numerical distribution of personal communication, which is more
questionnaires (fraction denominator) and effective since it is difficult to ignore as well
replies (fraction numerator). The figures in as allowing illustration and questions of
parentheses give approximate distances from elucidation, the time involved for both parlies
R & D, in miles. can be very great.
The Designer here is to insist on man-day estimates against

One can appreciate the designerJs need to each item on a list of the various sequential
wrap hirnself around his desi gn problems and activities in design work. Delays in the early
his possible annoyance if requested to distract stages will then not be able to eat into time
hirnself and talk to production people and properly required for the later stages.
potential customers. For some designers, Another cause of weak final stages, or
there is also a need to avoid the risk of a the Iack of an adequate 'hand-overJ of data
'loss of faceJ, occasioned when an idea and background, is due to the pressure of new,
eventually proves to be misconceived. There urgent jobs. Department planning should
is sometimes the fear that the proprietorship therefore try to keep a uniform rate of job
of the scheme may become doubtful, after closures and balance this against a rate of job
early disclosure. starts that more than covers urgent work.
Against this is the need to establish, There must obviously be a sufficient number
under changing circumstances, the production, of people in each of the necessary professions
economic or time feasibility, and to build up to allow this rate of job closures.
an understanding of the customer's personal
and technical needs. R & D management are
Conditions Favourable for New Designs
thus placed in the position of establishing a
system which measures success in terms of Although growth and keenly competitive
design acceptance and rewards on this basis. industries are more likely to encourage or
General management may find it necessary to accept the trial of new equipment, much can
ensure that co-ordination, between the various be done to favour new methods in all firms.
functions of research, design, construction The author is associated with an industry
and user, occurs at the problem-solving rather where the number of links between functions
than solution-accepting Ievel and starts at the within a firm and between supplying and
first inception of any new scheme. purchasing firms is continually increasing,
Multi-department project teams with a with the aim of encouraging improvements
responsibility for particular types of develop- and new developments. Within the firm, for
ment or improvement work is one method of example, standing working parties comprised
organizing this type of co-ordination, so long of Operations, design and development
as executive authority is given to implement engineers examine plant design defects and
or demoastrate schemes. Experience has shown Operating difficulties. This acti vity can Iead
that members of a project team that produces to a team spirit and a common language which
data for one of its member departments can assists in the acceptance of innovation outside
feel that they are only presented with parts of the working party remit.
the problern and that decisions are biased. Collaboration between a customer and
Under this arrangement, it is difficult to his suppliers may appear difficult, when the
sustain everyone's best endeavours. latter are in competition, but areas can be
found where the cost of potentially useful
research is too great for an individual firm or
Design Planning where the creation of new data will not inhibit
Design acceptance can be jeopardized by a variety of competitive designs.
skimping in the final stages. This is sometimes A technique for forcing new designs is
traceable to a designer becoming concerned reported in the 1961 ZUCKERMAN report to the
that if he requests much further time, to meet a Minister of Science. This is exemplified by
late difficulty, then this may be judged as a the Development Group of the Department of
sign of his incompetence. Perhaps one answer EducationJs Architects and Building Branch,
which is designed to bring scientific and comparison was therefore made of the 28 jobs
technical knowledge to bear directly on the completed between May 1964 to April 1965,
fonnation and efficient execution of building against the 66 jobs completed in the period up
policy and is an integral part of the Branch to April 1964. The implementation results
responsible for approving building projects. are shown in Table 9.1.
The Group, for example, is given the It will be seen that although the failure rate
responsibility of building a required school, remains unchanged, there is an apparent swing
whilst at the same time attempting an from partial to full implementation.
improvement in building design, components
and methods by exploiting applied research Conclusions: Proposed Methods for
results that have become available and by lmproving Design Acceptance
scientifically analysing all the normally Based on the facts presented and the
accepted assumptions. Since financial attempted interpretations, it is suggested that
discipline is imposed, a system of cost experiments could be conducted to prove
planning is necessary and this provides a whether the adoption of any of the following
common language for the collaborating methods, at particular stages of the design
professional and lay parties. These comprise process, can significantly increase the
people representing the policy-makers, teachers acceptance of new designs of equipment and
(users), development architects, surveyors and techniques.
engineers as well as co-opted designers,
suppliers and builders. Before Design Work Commences
(1) To identify a customer's needs:
Attempts to lmprove Design Acceptance
The work that was started by the new (i) Obtain an adequate knowledge of
R & D department in 1959 began to mature his outlook, his perceived functions, and
in 1961. During the next year, some failure of the work problems upon which innovation
design acceptance was noted, particularly in may be perceived as intruding.
respect of jobs started by R & D initiative. (ii) Test out his need for self
Approaches to improve the situation for these expression by relating the proposed
Table 9.1. lmplementation Results
Period Successes* Part Successest Feilures

('1o) ('1o) (%)
Up to April 1964 55 27 18
May 1964 to April 1965 71 8 21
* Includes equipment or technique for R & D department's own use.

t Includes work to disprove a current theory or design concept.
jobs took the form of: (a) obtaining, where development to his own, possibly latent,
possible, a sponsor; (b) in the absence of a schemes or by attempting to share the
sponsor, conducting the fieldwork at factories creative work with him.
where the management was keen on innovation. (iii) Consider the nature and relative
The effect of such approaches should priorities of the demands on his time,
have begun to appear during 1964 and a other than for innovation.
(iv) Set up, where appropriate, a (iv) Prove that the design meets one
Multi-Party Development Project Team of his needs (e.g. successful adoption
or otherwise prevent the conflict which by one of his colleagues).
can arise between departmental aims (v) Show that the steps he will have
(e.g. bring the rewards system fully to take are convenient and practical.
into line with company policy).
(v) Agree an adequate initial Acknowledgment
specification of the problern or require- My thanks are due to Mr. R.H. Coates,
ments. South Western Regional Director, Central
Electricity Generating Board, for permission
(2) To assess feasibility: to use the various facts collected in the
(i) Carry out a thorough analysis of course of my official work. The opinions
the technical needs. expressed are, however, personal.
(ii) Estimate development and likely
implementation costs by constructing a
detailed plan of the work. APPENDIX A: Equipment and Process
(iii) Test out the likely costs on Design Fai lures
potential customers. Technical Causes
lnadequate ini tial specification.
Inadequate design.
During Design Work
Unscientific approach.
To increase quality and relevance: lnsufficient study of existing designs.
Problem unsolved within set man-time
(i) Avoid, wherever possible, the
trial and error approach. resources.
(ii) Generate a sufficient number of Incompatibility with existing plant
initial design concepts before a final system.
choice is made. Sub-designers not consulted when scheme
(iii) Consult other professionals, to later modified.
ensure possible critical factors are Design beyond average craftsman's
recognized and fully assessed. skill.
(iv) Check back frequently to Insufficient manufacturing control.
representative customers, to ensure Equipment beyond average operator's or
maintenance man's skill.
that the design work keeps in phase with
Customer lacks background for dealing
changes in their understanding and in
with application problems.
the commercial situation.
Economic Causes
After Design Campletion Change in the commercial environment.
Competitor's design best.
To present the design to a customer:
Development or subsequent production
(i) Ensure his interest is caught. cost too great.
(This may involve the cost of a personal lncompatibility with customer's internal
approach.) organization.
(ii) Use his language. Market too small.
(iii) Relate the design to his personal Each customer required special
needs. modifications.
High installation or maintenance costs. Sampie trialability .

Customer paid for a design which he Ability to give customer satisfaction,
might have needed quickly at a future date. within his social group's norms.
High relative economic or social
Personality Gauses advantage to the social group.
See p. 67 Compatibility with the social group's
pa!ilt experience.
APPENDIX B: The Design and The Customer The Customer must have
Rogers, in his valuable source book on Youth,
innovation research, shows that a number of Outward-lookingness,
concepts are relevant to the adoption of new Relevant specialization,
equipments and techniques. It appears worth- Opinion leadership,
while to attempt a validation of such statements Dissatisfactions.
regarding maximum adoption as:
The Design must have The Social Group must have
Simplicity, Modern values (rather than traditional

Ease of demonstration, ones).
PART 111
THE ELEMENTS OF DESIGN

Chapter 10
A MORE DETAILED VIEW OF DESIGN
S. A. Gregory
lntroduction outcome is uncertain or unpredictable in

This section of the book provides an practice. For such situations strategies have
analysis of the acti vity of the designer in his to be developed: GREGORY (1966) has
response to the environment and his own discussed the relationship between such
inner drives. He is the link between strategies and the total design situation in
opportunity and realization, between the dealing with the problems of new process
situation and the product, and by his action design.
may contribute to them and change them. He is In the model which Watts proposes he
bound by the logic of his own mode of working stresses the autonomic aspect of the designer.
and by the constraints of the design task. This means that the designer or the design
In Chapter 11 Watts sets out to provide a team has to be responsible for the decisions
model of the designer at work. This model, made. Watts makes the point that at some
which he terms an iconic model, is abstract stage it may be necessary to refer decisions to
and does not attempt to reproduce the features higher authority. This may be because of the
of a designer and his environment except in so constraints of the situation, or because of lack
far as they represent the key logical relation- of potentially ascertainable information, or
ships which the dynamic activity of design because uncertainty has to be absorbed. For
requires. his paradigm or model of autonomic acti vity he
Watts uses a model which appears at turns to general systems theory, which, as a
first sight to be somewhat abstruse and theory of general models, includes within its
couples it with the language of set theory and scope the theories of different aspects of
of general systems theory. His reasons for systems behaviour such as communication,
doing this are compelling. He uses set theory control, learning, adaption, etc.
in order to provide general rules and the lt is within this area of autonomic activity,
subsidiary rules which are enclosed within which includes the possibility of the
them: necessary relations or consequences exploration of the environment for alternatives,
thereby become exposed and obvious. This that the methodical procedures for design which
development based on logic is later supported owe some of their inspiration to work-study
by the practical evidence of the various should be placed.
methodical procedures in design, particularly WADE (1960), formerly at Gosta Green, but
those of ARCHER (1965) and of LA THAM, now with the Engineering Employers' West of
TA YLOR and TERRY (1965), the latter being England Association, seems to have pioneered
better known as P ABLA (prob lern analysis by the application of the work-study approach.
logical approach). The implications of set This has been followed by the well-known
theory hold not only for determinate situations development of design teaching at Bristol
but for situations in design in which the which is dealt with by Matebett and Briggs in
77
78 THE DE~GN METHOD
Chapter 21. The technique, in a form known as sometimes thought better to observe this zone
Critical Examination, has been developed of activity as one of search and analysis.
inside Imperial Chemical Industries Ltd., Watts indicates the importance of some
particularly in the Dyestuffs and Mond device for sorting relevant information from
Divisions. Aspects of this have been provided 'noise' and this holds true whether the
by BIRCHALL (1960), BINSTED (1960) and designer is submerged in information or has to
BAXTER (1961). From the United Kingdom search for it. This item in the perception
Atomic Energy Authority, WOLSTENHOLME machinery tends to sort or organize information
(1962) has made a general appraisal of work- into sets of some apparent relevancy, usually,
study in design. for practical purposes, sets with as little
Watts turns to the work of MESAROVIC interaction as possible. He makes clear the
(1964) within general systems theory and his fact that some attention has to be accorded to
development of the foundation, largely in the treatment of 'isolates', in the terminology
terms of set theory. This work of Mesarovic of LEVY (1938). Components need to be
has been conveniently summarized in a piece regarded in terms of their interactions in order
of doggerel by BOULDING (1964): to obtain a satisfactory summation.
The model of the activity of the designer
•According to Mesarovic put on the external surface of the vertical
A set of proper statements which cylinder is topologically the same as the
Has mastered, in well-ordered schools, activity recorded by AsiMOW (1962), provided
A set of transformation rules that the cylinder trace is unrolled. SCHER
Which rules in turn have rules to twist 'em (1965) has discussed a version of this process.
Deserves the name of general system. Watts hirnself sees Archer's latest network
scheme as the best available paradigm,
All systems, i t is now proposed including, as it does, working rules in
Are either open, or are closed. considerable detail. How far this may be
The closed have one-to-one relations maintained will probably depend upon the type
But don 't result in innovations. of work involved, upon whether the design is
The open are disturbed, adaptive design of a system, or whether it is radical or
Or Heisenberg-observer-capti ve.' novel design.
Watts considers the case of system design
The model designer is depicted as moving as such, i.e. the design of an artefact having
spirally upwards on the surface of a vertical interrelated components and a pattern of
cylinder, rising from the abstract to the more input-output behaviour involving motion,
concrete, and cycling successively through energy, material, or information. Concepts for
zones of analysis, synthesis, evaluation and design may be generated by varying the
decision. Watts emphasizes the fact that these component characteristics whilst maintaining
changes of activity cannot be separated the interconnection or morphology. This gives
completely and that much of the work involved rise to a sequence of steps termed the
is preconscious. morphological approach. This interpretation
In his view the designer is submerged in of the morphological approach to design is
information of the kind which McCrory rather more in line with the viewpoint of
discusses in Chapter 2. Not all designers NORRIS (1%~) than might be gathered from
undergo the same sort of experience, many the frequent references to ZWICKY (1948) who
suffering a shortage of information. This, in coined the term. Even here, as a possible
turn, leads to a search. For this reason it is technique in the design of systems, the
THE ELEMENTS OF DESIGN 79
morphological approach, as seen by Watts, Farradane is primarily concerned with the way
represents a member of the set of approaches in which concepts are handled by the human
in system design. The scope of the set of brain in order that mechanical assistance may
approaches is probably best given by be correctly developed. For Farradane the
GOSLIN G (1962). important tasks of information processing are
The work of Zwicky may be viewed in a those of collection, storage and retrieval. In
number of different ways. It may be seen as these cases the information is extemal to the
related to methodical exploration techniques, designer and has to be handled externally.
such as the tree method discussed by Eder in This information consists of concepts of
Chapter 3, or the matrix method. The original what may appear to be in the extemal world.
morphological method, without any system Although designers must themselves pursue the
attachments, provides also a foundation for information that they need, it is possible for
certain creativity techniques, as mentioned them to be helped by the specialists who
by Broadbent in Chapter 14. process information and who study the nature
The chapter by Watts represents the most of information.
substantial attempt to date to provide a Storage and retrieval depend upon some
plausible model of designer behaviour in formal kind of structure in the information. Already
terms. Its power may be gauged by the way in reference has been made to the need to deal
which it is able to bring together the principal with an excess of information by some
recognized methodical procedures for design filtering device, or, if search is needed when
which have been listed in this chapter. The information is scarce, to have some guide for
reader is encouraged, therefore, to make the selection.
attempt to familiarize hirnself with the argument Farradane sees the method of developing
which is put in terms of currently available possible relationships between concepts in
theory. For man-computer design, as discussed terms of practice which has been identified in
by CooNs (1963) some such formalization problem-solving. This is justified, in particular,
is useful. by the researches of PlAGET (1953) and
For an introduction to the way in which GUILFORD (1959). The latter, in his work on
models may be used for specific design creativity, has stressed the nature of operations,
problems the reader is directed to Chapter 17. products, and contents. The operations are
But before trying these it is recommended that those involved in problem-solving: perception,
reasonable efforts should be made to draw memory, convergent thinking, divergent thinking,
such useful consequences as there may be and judgment. These concem products which
from the general approaches. System design are essentially arrangements of concepts. It is
is particularly susceptible to treatment in upon this kind of basis that Farradane has
this way. developed a system which classifies in terms
of nine c a t e gor i es of relations. This
The Handling of Information classification has been checked against
In Chapter 12, Farradane deals with the 10,000 items.
topic of information for design. Design may be In its present state Farradane's system
seen as essentially an information processing in concerned with semantic information, i.e.
system. Indeed, acti vi ty studies and analysis written material, and does not deal with visual
of designer behaviour (e.g. the study of design material. It presents the normal range of
office behaviour by TURNER, 1964), stress the static relationships between such concepts.
large amount of time needed for the collection, Although this provides a substantial coverage
reading, discussion and testing of information. there is still the problern of preparing for the
future, for repeatedly different aspects of Newman points out that a network of the
information. There appears to be li ttle type that he describes may be taken as a
immediate prospect of having ini tial analysis computer programme. But his programme has
done by computer. In spite of the difficulty of more 'degrees of freedom' than a normal
analysis it should be possible to transfer the programme. It is possible to develop a set of
logic of the present system to the computer. nodes to deal with any logical situation. lt
It is hoped that some development of the is also possible to set up models which will
present logic may be discovered which will provide 'language-type' patterns along the
permit the mechanization of creative thought lines indicated by Noam Chomsky. (An
processes. ENGELBART (1962) and Ross and introduction to this topic might be gathered
RODRIGUEZ (1963) suggest some lines of from PuTN AM, 1964.) Chomsky sees the need
work in connection with Computers. for a complex transformation system to make
sense of a stimulus-response model as far as
A Speculative Model of Pattern Production
linguistic behaviour is concemed. Newman
lf methods by which design thinking might
gives as example the phrase 'old cow'. The
be transferred to the computer are to be
response to this stimulus will depend upon the
developed, possible ways in which a computer
immediate environment and previous experience
might carry through such Operations must be
of the listener, and possibly other things.
considered. At the same time light might be 'Meaning' according to Newman suggests
thrown on the way in which the brain works by a special structure in each cortex. This might
starting from consideration of possible modes be isolated, as a programme, by 'killing' all
of computer operation. the nodes around it thereby rendering it precise.
In Chapter B Newman puts forward some But in any living situation it is attached to
speculations which go along the road to the whole of the cortex and the whole of the
fulfilling part of these ambitions. In his thought cortex may be in action during the operation
about how a brain might work the central
of the programme concerned. Given such a
feature is the pattern. The pattem contains
model of meaning it is possible to understand
a hierarchy of bits and shapes. This may be
better the way in which humans recognize
well represented by a network of lines. The
words or visual patterns.
properlies of such a network of lines may be
The identification of patterns and the
discussed in terms of graph theory.
generation of patterns are implicit in design.
Such a network may also be seen as a
Information has to be accepted by the restricted
definition of procedure. In passing through
input system of the human being (45 bits per
such a network the action at any junction or
minute?) according to some pattem. The pieces
node may be influenced by what happened at
of information have to be arranged to provide
the previous j unction. A characteristic of
some hierarchical pattern. It might even be
nodes is their 'connectivity'. A network may be
that the autonomic model experiences relief,
seen as a pattem of connectivity. In a pattern even an aesthetic reaction, on achieving the
there may be many isol~:~ted procedures; there required pattem following the practical activity
may also be dense regions interconnected by
of design.
tenuous links. With simple networks it is
possible to follow through the alternative kinds
of behaviour which develop according to the On Practical Techniques of
value of the initial impulse fed into the network. Stimulating Creativity
With a complicated network or anastomosis, From the model of the autonomic brain,
however, it is difficult to predict what is which in turn can be seen as a possible
likely to happen. explanation of the specific design behaviour
of the iconic model of the designer, it is no The techniques which he proposes to

long step to the consideration of practical assist in the drawing out of existing creative
techniques of creativity. Here arises concern ability in c 1 u d e check-lists, interaction
with finding the kind of stimulus which will techniques, and free association methods.
produce, or assist in producing, a creati ve These tend to be verbal in character and some
response. This may be anything from a simple effort might be devoted to extending visual
word to a challenge at the TOYNBEE (1960) methods: already the designer is aware of the
level of history. stimulating effect of freehand sketching.
Each individual has his own mental Discussion of creativity inevitably raises
equipment and his own experience. Is it the question of its adequate definition in such
possible to provide a stimulus which will terms that it can be measured. This problern
directly produce a creative result? Is it has been treated in some detail in the book
possible to restructure experience or mode of edited by TA YLOR (1964), particularly by
thinking to yield a valuable result? Can the BROGDEN and SPRECHER (1964). Creativity
individual be put into some fresh relationship is bound up with value, and value changes with
with other individuals and/or machines to the circumstances. Creativity, just like design,
facilitate the necessary transformations? is involved in the product, in the process, and
in the situation. An attempt is made to link
In practical situations creative work is them in the GREGORY-BURDIS model (1965).
likely to involve something much more complex This should be seen against the Watts model
than simple logical transformations. The of design.
evidence suggests a considerable depth of
personal involvement. This goes beyond the
intellectual information processing model Value and Design
largely suggested by Watts, although he The practice of design turns upon some
recognized the considerable extent of system of values, whether intuitive or
preconscious work. In some way yet undefined conscious. A product is designed because it
in any model, the interaction of the life process is held to be of potential value to a consumer.
of a specific designer with his task must be The consumer may find one manufacturer's
accounted for. product more valuable to him than that of
another manufacturer. The product itself will
Broadbent, in Chapter 14, proposes a
have a different value to the designer.
simple model of creative behaviour, which has
relationships to other models developed in The consumer, particularly if he is a
the book. He thinks in terms of a Stimulus- government department, may make some choice
response situation in which the incoming which appears to be nothing but arbitrary, but
stimulus is compared with constantly changing henceforth the designer may have to take this
basic patterns. He believes that the imaginative arbitrary choice and use it as the standard
against which to evaluate the major and minor
designer has either intrinsic flexibility in his
factors in his design. The choice provides a
patterns or the ability to restructure them
pattem in relation to which alternative
readily. He emphasizes the fact that this
possibilities (information, shapes, components)
ability may lie in dealing with a specific
are selected or ranked.
kind of material, such as words or visual
patterns. It should be noted that conferences Does the autonomic designer operate in a
tend to attract verbalizers whereas many universe of absolute values, or does he have
designers are visualizers. SMITH (1964) has built-in values special to himself; are the
recently come to the defence of the Visualizers. values imposed for the occasion; does he
develop them for himself, or are they random? difficulties involved in practical use of either
How far, indeed, is he autonomic? games theory or statistical decision theory.
Pleydell-Pearce, in Chapter 15, tackles This is discussed at length by FISHBURN
the discussion of value from the standpoint of (1964). The essential difficulty lies in the
the philosopher. He puts the arguments in a treatment of probability for a single occasion.
very general way, hoping thereby that people Furthermore, at the outermost boundary of the
will not become emotionally involved. decision, the value is unlikely to be simply
He finds that values may be instrumental, determined. Methods of handling non-linear
arbitrary, or through exemplification. An object value expressions and subjective probabilities
may have instrumental value through its ability may be combined to give expected utilities or
to help some further aim. An arbitrary value expected relative values. Mathematical
may arise from some preconscious choice; expressions of this kind, although intuiti vely
bu t, in rational situations, arbitrary values acceptable, require quantitative values for
must also occur since there is a practical practical applications. Such quantitative values
Iimit to search for rational explanation. are most unlikely to be obtained in a decision-
Such a discussion of values is of particular
bounding situation.
significance to those designers who have to In industrial design organizations the
deal with intangible and non-quantifiable criterion of decision operated by the board of
values. Architects and planners find such directors is not readily discernible. lt is
problems and the predicament of the planner usually stated in oversimplified terms and the
has been discussed by LICHFIELD (1964). finer details people are expected somehow to
But it should be noted that in potentially assimilate by absorption from the atmosphere.
quantifiable systems the concept of optimum WILLIAMS and ScOTT (1965) have studied the
may be meaningless. background to decision on investment in the
case of fourteen firms. They find that the
goodness or otherwise of a decision cannot be
Decisions
judged by any of the simple criteria proposed
The philosophy of ethics is largely devoted
for investment analysis and that there must be
to exposing the arbitrary nature of theories of
some study of the context of the decision.
value. This is not to suggest, however, that
theories of value should be thrown away, but The decision, so to speak, only has meaning
that they should be treated with the relevant in a particular background. This agrees with
amount of respect. Rational decisions are not the heuristic appraisal method of GARGIULO
possible without value systems, whatever the and colleagues (1961). The problern is
values may be, instrumental or otherwise. analysed by SHUBIK (1964).
A decision, particularly a design decision It is possible to visualize the investment
when taken in modern society may affect many criteria of a company as the outer boundary
people, and may involve the expenditure within which subsidiary decisions have to be
of substantial resources. Decision implies- carried through. The directors' criteria should
responsibility for action and outcome. It is comprise the set of all decision criteria likely
preferable that the decision should be made on to be operated in the company. Actually the
bases which are communicable and acceptable. company is an open system and apportunilies
This implies some kind of rationale and for change may exhibit themselves to different
rationali ty. levels in the company. Technical opportunities
Decision theories have been developed are likely to present themselves without
with the hope that they might provide such a entering the practical range of the directors'
basis of convi'ction. There are serious decision criteria. In the last analysis the
THE ELEMENTS OF DE~GN 83
individual designer has to make his own different fields of study, conceming practical
decisions, even decisions whether to refer to investment behaviour. Investment decisions
higher level. Concem is, therefore, with the and design decisions are linked together and
designer's own motivation and his structure of with operational and research and development
expected relative value. decisions as shown in Figure 10.1. As yet
there has been no empirical observation of this
Social objectives and constraints overall theme, but in view of the interest of the
subject, such work should be strongly
Organizatton objectives and constraints encouraged.
'
I
Investment criteria
Decision and Models
Any practical action in design almost
inevitably involves the use of one or more
' models. These models may be most diverse in
I , ()perallonal Oe51gn Rand D Sal"s
llf1f1flf1
character and used either for solving a problern
or for communication. lf, for example, an
opportunity is perceived, it has to be described
by some model. If, again, there is a need to
I I l determine the distribution of stress within a
Extsttng and future opportuntttes component, this is calculated by the aid of a
mathematical model. In each case it must be
Figure 10.1. The open system of decisions decided how applicable the model is to the
given situation.
Any theory which is developed regarding
decision has to be related to practical Within the practice of design there is
behaviour. Recently there has been a welcome concem with the exploitation of three kinds
growth in the empirical study of decision in of model: the design method, the range of
design. At the same time there has been a detailed problem-solving modes, and the models
growth in interest, an interest arising from of specific technologies.
4
Chapter II
THE ELEMENTS OF DESIGN
Ronald D. Watts
lntroduction represented by a line rather than by an area to

An iconic model of a designer or design indicate they are autonomic to ~(STARR, 1963).
team ~. as suggested in MESAROVIC (1964), Design proceeds by repeated cycling at a given
is shown in Figure 11.1. ~ is in dynamic level, until ~ obtains sufficient confidence
relationship with an environment &, containing upon evaluation to advance to a new level.
the total s p e c tr um of scientific and In the main, progress is from a lower, more
technological knowledge (MCCRORY,1963). A abstract, level to a higher, more concrete one:
necessacy, but not sufficient condition for a this is represented by the helical path on ~.
design process to eventuate is the existence However, ~ learns during the process and
frequently reiterates at one or more levels.
A state function D of the design is associated
with the process path and can be externalized
Concrete as a set of statements at intersections of the
path and the decision line. Various states of
the design thus relate to the different levels.
Asimow has given a two-dimensional flow
Dec1S10n
diagram defining the processes at the different
levels and the states as outcomes: this diagram
·--------- is reproduced as Figure 11.2, and would be
obtained by 'unwrapping' the design path from
the cylinder of Figure 11.1. The design states
(giving a vertical structure to the process)
Abstract also proceed through analysis, synthesis of
design concepts, evaluation of feasibility,
later giving way to optimization, revision and
communication.
The process can be considered complete
when ~ releases into & a communication P,
Figure 11.1. Designer~ in environment 8 being a set of prescriptions for the embodiment
of the design. The end to which P is a means
of a valid human need evidenced by a is an artefact A: this possesses several
communication N. ~ performs the proce:;ses functional attributes, some of which fulfil the
of analysis, synthesis and evaluation leading need implied by N; others enhance the profits
to decision as shown exteriorized on the and reputation of ~ and his company, and yet
surface of ~ (JONES, 1963; AsiMOW, 1962; others may have effects which are far-reaching
AL GER and HA YS, 1964). Decisions are into the socio-economic environment.
85
86 THE DE~GN METHOD
Phase II Preiiminary destgn Phase ni Detailed design
SymbOls
Procns ~
'0
Figure 11.2. (From Asimow, by courtesy of Prentice-Hall)
The existence of N is thus symptomatic optimum design often arises through failure to
of a two-sided problern situation. On one side define the need problern situation adequately,
there are the factual causes giving rise to N or to evaluate well enough the implications of
which are the subjects S for definit i o n, the chosen solution.
analysis, and measurement; on the other there
are the implications of the proposed A which
are the objects 0 for evaluation and decision. Analysis of the Need
lntermediately are the requirements and The following comments are affered to
resources of !:::, and the company he represents: emphasize the necessity for an objective
these tagether with S and 0 constitute a set analysis of the need and for caution against
of circumstances. The purpose of !:::, is to accepting statements of need at their face
achieve optimum conciliation of S and 0 in value. Sometimes N may be interpreted as
this set of circumstances. Failure to achieve · evidence that a client is merely ignorant of an
available device: in this event the need is Evolution of the Design and Documentation
considered not valid. Frequently a customer In stating that the design states are
fails to appreciate or state his real describable by a set of statements, e.g.
requirements: N may refer to an artefact by
name instead of by the function or service it is Dm = {dld = statement of design at
the mth intersection}
to perform, e.g. a 'voltmeter' may be called for
when the need is to measure voltage. Taking the widest interpretation is to be gi ven to the
into account the needs problern situation it is term 'statement'. For example, the description
possible that a potentiometer would be required. may assume the forms of verbal specifications,
A less trivial instance would arise if N referred circuit diagrams, interaction diagrams, chemical
to a house: is the functional requirement then formulae, mathematical models, indeed, any of
(i) protection from the weather, or (ii) an the forms of written communication. The
instrument for living in as averred by complexion of D changes as it progresses from
Corbusier? Indeed it may be noted that (ii) does N (where its elements d are sparse, general
not exclude (i), and in a needs analysis, 6 may and predominantly qualitative) to P (where they
well log weather protection as a design are detailed, particular and richly quantitative).
requirement and proceed to investigate the Any new D, say Dn, must imply any previous
functions implied by (ii). On occasion, N is one, say Dm: hence Dn :::> Dm is a condition
presented as a comparatively tight specification for the design to exist; this entails there be
rich in quantitative terms: this perhaps calls no d E D which throws the design outside
for the greatest scepticism on the part of 6. criteria specified by 6. These criteria are
Why are tolerances. set to ± 1 and the panels
inclined to 45°? It is not unknown for some
o.~.~o,.
well-meaning but ill-informed intermediary to
close to costly limits the tolerances of fit of
relatively unimportant parts to 'be on the safe
I
side'; and if 6 does set the panels to 45°,
ostensibly for ease in reading meter indications
or manipulating controls, how is he to know
that existing lighting will not thereby be
(o) Failure (b) I Dmn
( cntena met
onl (c) Design exists
reflected into the eyes of the operators? Dm$Dn.

crlteria not met
It is also necessary to appreciate that the
needs situation changes with time: thus a
continuous measurement of the market may be
necessary in some instances if the innovations (d) Dn divided into several
of today are not to become the bric-a-brac of sub-sets, each containing
Dm hence each a solution.
tomorrow. In general, 6 must regard all the
'Best' depends on criterion
information recei ved from 8 as time-varying of optimality.
and 'noisy': this refers to information about the
needs situation as well as to scientific, Figure 11.3. Variants of D U .D
technological, economic and other data. 6 acts m n
as a perceptive filter, deciding what is and
what is not relevant to the particular written into D at the start: the function of the
circumstance. Considerable reiterative prohing evaluator in 6 is then to apply the criteria when
into, and feedback from, 8 may be necessary investigating relations on the union Dm U Dn.
before 6 decides to proceed. Figure 11.3 shows the possibilities. Attention
88 THE DE~GN METHOD
is called to the fact that the boundary of Dn Approaches to Design

may be kept as wide as 6 is able: this is the A. logical approach is by envisioning the
policy of 'minimum commitment'. As an implication of P and of the resulting artefact A
important instance in the evolution of design, from its completion to its ultimate disposal.
adoption of this policy allows the requirementsThe embodiment of the design according to
P ensures the fabrication of A. Questions
of a specification to be met whilst still giving
room for manoeuvring among innovative, therefore arise as to manufacturing resources:
worthwhile solutions. One problern in designinghuman (labour, etc.), physical (materials,
machines, energy), informational (prescriptions,
is not so much what information is to be brought
in from the outside world to be synthesized models, etc.), and economic. The fabrication
of A ensures that it comprises a set of
into a prescription, but rather, what information
exists which the designer can afford to ignore components which are logically assembled:
or reject as irrelevant to the particular this enforces the concept of system. A (closed)
circumstance. The more knowledgeable 6, and system is, by definition, a set of components
the richer the state of the relevant art, the or processes that can be bounded under the rule
that all relevant interactions or inter-
greater the likelihood of the flourishing of a set
of solutions from which the optimum can be dependencies must be enclosed (Mesarovic).
selected. Regarded as separate entities, each component
lt is evident that 6 is an information- has attributes, some of which make it of
processing agent seeking conciliatory relation-particular value to the circumstance (e.g.
ships over a large set of elements, or factors,rigidity, resisti vi ty, chemical inertness), and
and that these change as the design progresses.others of which are of nuisance value to the
Systematic book-keeping is necessary to circumstance (e.g. weight, heat dissipation,
provide: cost). The mode of assembly of the chosen
components is such as to exploit to advantage
(1) A means of keeping the goal in sight. the desirable attributes of the components
(2) A ready means of communication between whilst minimizing the undesirable ones. lt is
all involved, especially so that decisions may the attributes of the components which are
be made 'with one mind'. related by the mode of assembly. The
(3) A store of factual information. characteristics of A, both desirable and
(4) A store of propositional information. undesirable, are determined by (i) the character-
s
(5) A record of decisions which may be of istics of the components, and (ii) their mode T
value in the post-design stage. of interconnection.
(6) An archi val record of value to other (In terms of general systems theory, one
design situations. refers to the s
as the relational constituents
and T as the structure, topology, or morphology
Jones, LATHAM (1965) and Alger and Hays of the system. One cites a set of relations such
give information upon this topic; attention is as R on A such that R ={ T, s }.It is these
drawn to the last item. Choices often have to relations which are investigated by 6.)
be made in design against non-metrizahle Questions now arise as to the desired
qualities, for instance comfort, as criteria. characteristics ( outputs, functions, per-
If the decision, along with the processes formances) of A since these give meaning
giving rise to it, is recorded in one instance to the expression 'exploit to advantage' in the
this constitutes a case history for a second foregoing paragraph. Questions also arise as to
occasion: at very least the probability is undesirable characteristics. Both sets of
reduced of making the same mistake twice. questions can be answered in part from an
analysis of the needs situation. The totality An alternative approach is described by

of characteristics is obtained by considering Jones. Given a statement N, each member in !:::.
the future probable history. The artefact will compiles a random list of factors which he
then be a labelled component interacting with associates with N. The factors are serially
others in a series of environments, i.e. A is a listed (this formulates an initial set D). The
component of a series of systems: the processes list is communicated to all members who by
it undergoes are predominantly out of thecontrol constructive criticism, discussion and reference
of !:::.. If for exampie A is a measuring instru- to information, make additions. The factors are
ment, the probable processes in order are: in- then classified into categories which are
spection, test, calibration, storage, packaging, suggested by the factors themselves. One
transportation, installation, commissioning, category comprises ideas, solutions and
operational usage, maintenance, repair, removal propositions which are documented separately
from service, disposal. Questions arise as to from other categories comprising factual
the possibility of vibration in transport information and design requirements. (A
destroying the calibration, the range of propositional set is thus separated from a
environmental temperatures in various usage factual set of factors for evaluation.) I suggest
sites and so forth. The interacting environments that reiteration of procedures sooner or later
relevant to this example are again human, yields, for a given problern situation, the same
physical, in f o r m a t i o n a 1 and economic. or equivalent classifications and categories
Consideration of A in all its probable roles as as other approaches.
a system component leads to sufficiency of Either of these approaches indicates areas
specification. where further information is required.
Table 11.1. Classilication of Operational and Environmental Aspects
Usage Influence Existing resources
Occasion Environment Previous Designs

Durstion Safety Existing equipment
Frequency Pollei es Services available
Seqaence Testing and installation Experience
Operators Time-scale
Maintenance Finance
Personal acceptabillty Menufacture
Analysis on these lines has been described Acquisition of information has been dealt with
by Latham in P ABLA (problem analysis by by Jones, and conditions for accurately
logical approach). Table 11.1 is based on Chart establishing facts by LARRABEE (1945).
Cl of his paper, and classifies the operational ARCHER (1965) has devised a 'Systematic
and environmental aspects under the categories, Method for Designers' providing a sequential
usage, influences and existing resources. list of activities to be performed and events to
GOSLING (1962) also proceeds on the basis of be recorded. The programme is set out in
a prior concept of system. considerable detail, reiterative procedures
10
0
Table 11.2. Morphology of Instrumentation
Definition: An lnstnlment is a system which malntalns functional relationships between prescribed variables.
Definition: Instrumentation relates to the design. construction and use of instruments for measurement and control.
Characteristics Electrical Mechanical Mecharucal Fluid now Thermal Chemical Otbers
Desired outputs Voltage v Velocity i. Angularvelocity !:J; Pressure p Temperature f Across i

f.l
Identify Current i Force f Torque T Flow rate g Heat now Q ~ Through j
Quantify
Resolution 3 3
..;
:I:
Purposetui Inputs V l x f :::- T p Q tri
~ E ~ it t
ldentify
-
~
Quantify
tri
Resolution Cll
sz
UncontroUed Inputs Electric field Vibration Vibration Acoustic Conduction Corrosion
Amplitude MagneUe field Shock Shock Pressure Radiation Humidity ii:
Frequency tri
..;
:I:
Undesired outputs 0
~
tdentify
Tranafer properties Ga in Response Sensitiveness Range Error Reproduclblllty Damplng Bendwidtb

Short term
Long term
Performance Re!iab!Uty Maintainab!Uty Durablllty Safety
Processes Fabrication Calibration Storage, etc. TransportaUon Ins ta!la Uon Amortlzation Re pair Diapoaal
Usage Ergonomie•
---- ~ -- --- --~ ----------
being specified at the appropriate stages of the Informational (standard frequency signals).
design, and it is associated with a graphic Economic (amortization, maintenance,
network of the whole process. Table 11.2 is a depreciation).
check-chart shown for use during the design of Uncon troll ed
measuring and control instruments to ensure Physical (ambient temperature range, 5°-
sufficiency of information.
30 °C; Vibration, mechanical, acoustic;
dirty atmosphere; ambient ligh ting;human).
Engineering Specification
(4) Constraints on Outputs
Figure 11.4 shows an artefact (or system) Included in (1) and (2) above.
A with its environmental connections, and (5) Constraints on Inputs
suggests the formulation of an engineering Included in (3) above.
(6) Constraints on A
~SirPd
Size ~ x x y x z; weight ~ p cwt.
Purpo~d
1nputs
charact~ristics
(outputs)
Expected service, 20 years.
Undesired Location as in Drawing 537.
Cost } .noo per each sub-assembly.
Uncontrotted charactenstics
inputs (outputs)
Attention to ergonomics and aesthetics.
(7) Unitsof Measure
Constra,nts Conslramts on Constra~nts lncluded in above.
on 1nputs A on outputs
(8) Objective and Criterion
Umts of measurP of ,npuls,outputs and constra,nts; Demonstration of work of the department*.
assoc,ated mPasuru of values where possible
Ratio of non-operative periods to total
OvPrail objl'CI s and design criterion time.
In the example given, the location of the

Figure 11.4. Artefact in its environment
device is specified and hence it is possible to
specification to the following scheme. (A measure the environmental situation in respect
running example relevant to a frequency of vibration, atmospheric pollution, lighting,
standard is appended.) In the event of a device being intended for use
in a variety of locations, decisions would have
(1) Desired Characteristics (outputs) to be made as to the nature and range of
Continuous indication of passage of time: interfering imputs to be catered for.
stability, 10-a; resolution, 1 sec.
Frequency outputs: 100 kc/s, 10 kc/s, Generation of Design Concepts
1 kc/s to 1 c/s with random access; Design concepts can be generated using
each 4 V p-p; impedance 10 k ohm. the t
notion of system: one seeks to maintain a
Record of error with respect to a Bri tish relation (or set of relations) R by varying the
standard frequency transmission. relational constituents ' whilst retaining th.e
(2) Undesired Outputs morphology T. The procedure has thus been
Heat, acoustic noise, electrical noise on termed the morphological approach.
valtage outputs, light reflections.
(3) Inputs
*The design and fabrication of this Instrument
Purposeful formed a joint project for diploma students at the
Physical (power, 240 V, 50 c/s floating Royal College of Art (School of Industrial Design)
and at the N orthampton College of Advanced
battery). Technology (Instrumentation and Control Engineering
Human (control, adjustment, maintenance). Group).
The first step consists of recognizing that MACHOL, 1957) system: the motion of the
the proposed artefact A belongs to a class "C hands on the dial at the 'output' of the system
of devices which have the same functional may be traced back through intermediate gears
attribute R. Thus, if the design project is to the balance wheel and escapement
concerned with a device which indicates the mechanism as the primary source of the motion.
passage of time as its characteristic, A belongs It would then be necessary to induce the
to the same class as watches, sundials, clocks general classes of oscillator, intermediate
and so forth. If A is a car, it belongs to the means, and indicator. Adopting the motor car
set of devices having the common property of as an exemplar in the class of vehicles, the
transportation. In general A E <if, where dominant output is where the tyres interact
with the environment at the road surface. The
rc = {Ai I Ai = 'has the characteristic R' } dynamic processes at the interface, are
concerned with supporting, driving (propelling),
The second step is to determine the controlling (steering) and stabilizing the car.
relational constituents and morphology. This These processes are seen to be necessary and
may be accomplished by posing the question functionally distinct: the morphology is as
as to what are common sub-systems (or shown in Figure 11.5(b). The processes need
processes) amongst all exemplars in the class. not be associative at one confined region of:the
In the example of clocks, etc., the essential interface, indeed they are not in an aeroplane.
sub-systems are a periodic oscillator, a means The difference between the two examples is
of indication, and an intermediate means. that, in the case of the watch the object was
analysed into major sub-systems, while in the
latter case, the interactions of the object with
its environment were analysed. Since A is
always composed of sub-systems and is at the
(a)
same time a system in an environment, both
analyses may be carried out. For the watch,
Support the analogous dynamic relationship to the road
Drive surface to tyre interface, is the ergonomic
A Control one between dial indication and human
Stabilize interpreter. For the car, one analogaus single-
.._ _ ___j-..:c!O==-
thread system would be obtained by tracing
(b) back the steering motions from tyres to driver.
The third step consists of listing devices
Figure 11.5. (a) A single-thread functional or processes with the same attributes as the
isomorph; (b) vehicle morphology consti tuents. The dass of oscillators X 1
includes balance wheels, quartz crystal, tuning
These are the characteristics which any device fork, magnetostriction oscillators, planetary·,
in this class 'must have'. The sequence in stellar, and atomic motions, body rhythms and
which these three sub-systems interact so forth. A class X1 of intermediate devices,
determines the morphology as shown in and a class X 3 of indicators are also
Figure 11.5(a). The system of Figure 11.5(a) enumerated. Alternative means of support of
is the functional isomorph in all A E ~ vehicles are land, air, water, and so on. In
One exemplar in this class might have general, several sets up to Xn may be formed
been adopted for analysis. Thus a watch may but it is advisable to reduce the nurober as
be regarded as a 'single thread' (GOODE and much as is consistent with carrying the
concepts. Often these sets may be intuited functional isomorph to be a simple spring-
thus obviating the formalization of the damper positioning device with load-spreading
preceding steps: whether ~ can do this will and cushioning members. The example of
depend upon experience and the object of vehicle characteristics previously mentioned
analysis. is taken from the same source. Alger and Hays
The fourth step is to combine members, describe a morphological analysis of clothes-
one out of each set. When the morphology is drying as depicted in Figure 11.6(a). The three
~
u
Wtrmg Verttcal Hor•zontal Press•ng
OSCtllat10n OSCillat10n
Drytng me<.tlantsm c
~Gas ~ Aor
c ----~-----+-----+----~-----~
~LIQUid 1:' Notrogen
i Vacuum l Oxygen
~Solod "' co,

~
~ (0) (C)
UJ
Figure 11.6.
(a) Morphological analysis of clothes-drying;
Gas Gas
(b) concepts of clothes-drying; (c) design
concepts for electric tumbler drying. (From
(b) Algerand Hays, by courtesy of Prentice-Hall)
thread-like, the order of combination is to be dimensioris of the model are equivalent to

preserved. In set-theoretic terms, one forms classes xl I XI and X,, with the various means
the Cartesian set of attainment as 'parameters' or 'values'
defining the ranges of the dimensions.
X1 X X2 X X3 X ••••••• xn Figure 11.6(b) shows two concepts chosen for
which yields potential solutions as ordered evaluation, and Figure 11.6(c) shows one of
n-tuples. If, to pursue the first example, the these being subjected to further analysis at
designer selects a transistor-maintained tuning a more detailed level of sub-systems.
fork, frequency-dividing circuits and dial The reason for suggesting that the number
indicator, he has a concept for a (once novel) of sets should be reduced as far as possible
watch. can now be exposed: if selection of design
The morphological approach has been concepts is to be carried out and one or more
described in the literature. NORRIS (1963) shown to be not feasible at a given level of
gives an amusing and instructive example in generality they may be discarded at this level;
which the specified function is to 'take the ~ then has no need to proceed to more detailed
weight of one's feet': he recognizes the level of analysis of those concepts.
Evaluation and Decision but with some obvious loss of generality, names
The first phase of the design process is of components belanging to three sub-
concerned with feasibility, and evaluation is assernblies have been appended. Each link
directed to deciding which of the design between two nodes represents a possible
concepts are capable of physical realization decision to use the combination of components
taking into account constraints of time and represented by those nodes: any path through
money. In later phases, evaluation is for the the tree thus represents a design strategy and
purposes of selection of optimum design, for !':::. is to decide which of the available
strategies meet the criteria, and ultimately
which is the best one with respect to an
3 2 3 optimizing criterion. The consequences of
x, Materiäl--; I
adopting any binary combination thus have to
I
be compared with the consequences of adopting
§ub- other combinations in the same 'row'. These
X2 Thickness 1system
1(1} consequences can be either advantageaus or
I
I disadvantageaus (with respect to performance
x3 Fi~i~~ _ ~ criteria), and !':::. seeks to assign values to the
links by weighing the consequences. Decisions
may be made upon comparison of the value
X4 Material ISub-
products for the various paths through the tree.
~ystem
I (2) Link values are normalized so that they sum to
X5 Th1ckness: unity in any one row to preserve proper
---Sum=1 i comparison. As described, it has been assumed
I
X6 Fifli~t} _ ~ that !':::. is indifferent to which of the x,
components should be chosen: often 1':::. has
I
I preferences amongst the initial components
X1 Material 1 and assigns values to these, afterwards
:Sub-
n>ystem incorporating them in the path-products.
: (3)
X8 Th1ckness: In the general case, the number of links is
II\ Ii\ I unmanageably great and !':::. is compelled to
I I \ I I \ I
fragment the system: it is perhaps this activity
I I \ I I No. of paths= 2,880
which makes the most demands on the
perceptive faculties of the designer. Some
Figure 11.7. Design tree
variables are relegated to the status of low
relevance to the particular evaluation or the
ensuring stability, and in general for ensuring system is divided into sub-systems which are
compatibility with the performance criteria separately evaluated. Bearing in mind the
relevant to the particular situation. definition of system, either method of
It may be supposed that, by a morphological fragmentation is not without its dangers: !':::. may
or intuitive technique, one of the required lose track of an advantageaus interaction or
functional attributes is satisfied by any interdependency or perhaps of a catastrophic
member x~ of a set X, of components, another one. For example, referring to Figure 11.7
by X 2n E X 2 , and so on: by a component is here suppose sub-systems (1) and (2) are separated,
meant a technical means of attainment of the and adoption of link Lp implies such rapid
function. Figure 11.7 shows a design tree corrosion that a reliability criterion could not
formed of several such sets: to help fix ideas, possibly be met, then a zero value would be
assigned to Lp. Any path containing Lp must ing solutions and decisions and that these will
then necessarily have a path-product value of vary from one desi gn situation to another.
zero: it is as though this prospect had not It has here been possible only to outline
been available in the first place. But now the kinds of activities which 1'::. undertakes as
suppose that with link Lq of the connected an evaluator and decision-maker. Evaluation
tree, L p implies cathod1c protection, the implies measurement (FINKELSTEIN, 1963;
interaction L .Lq would probably be worth a CHURCHMAN and RATOOSH, 1959) i.e. the
value approaclling unity. An analogous instance assignment of numbers to entities to express
in the evaluation of propositional control facts and conventions about them. If the
systems is the necessity to recognize designer is to be objective in his decisions he
autonomous sub-systems for separation. must necessarily refer to past history for bases
The weighing of relative advantages and of comparison: but the future is not beholden
disadvantages of selecting one link rather than to the past and advances entail a probabilistic
another, poses problems of the relative extrapolation. The results of past measurement
importance of performance criteria: one solution lie in information of similar or equivalent prior
may afford greater reliability at a greater cost, systems, in experience, and in having
and another greater maintainability at less cost. evaluated a real model if the stakes are high.
The decisions here are concerned with solving To extrapolate with the greatest chance of
further problems: for example, will the success suggests the use of decision and
maintenance required if the second strategy be value theory, and of optimization and dynamic
adopted cost more in the long run than if a more programming techniques (Mesarovic; Asimow;
reliable but more expensive strategy be chosen Starr; Go s 1 in g; Go o d e an d Macho 1;
in the first place? For 1'::. to come to an FISHBURN, 1964).
objective decision he must have access to
case histories and invoke probability theory to Acknowledgment
arrive at a quantitative assessment for the I wish to thank Mr. L. Bruce Archer,
particular circumstance: obviously maintenance Director of Research, School of Industrial
labour costs on user sites is a factor of Design (Engineering), Royal College of Art,
importance in relating maintainability to costs. London, and also Mr. L. Finkelstein, Senior
However, this discussion stemmed from a Lecturer in the Department of Production
consideration of link evaluation in a design Technology and Co n t ro 1 Engineering,
tree: in general it may be expected that with Northampton College of Advanced Technology,
each link are associated sub-problems requir- London, for their encouragement and criticism.
Chapter 12
INFORMATION FOR DESIGN
J. Farradane
IN Chapter 11, Watts presents the need for a processes are within the field of psychology:
rational approach to the processes involved in how people think and combine concepts, and
engineering design as an intellectual activity. how they solve problems. Difficult though such
There is the morphological approach of NORRIS studies may be, there is evidence that the
(1963) to provide a conscious framework (as mechanisms of thinking are less complex than
opposed to intuition) of all steps in the design might be supposed. Such mechanisms are the
process. Parameters of essential qualities and means by which people manipulate concepts in
quantities, and parameter steps of possible relation to one another, a concept may be a
solutions for the requisite parameters, are single thing (as it is named) or a complex of
identified so that the stages of identification, things to which a single name can also be
analysis and synthesis can be examined and applied; the complexity of the concept does
systematically treated. J ON Es (1963) developed not alter the methods of handling it by human
a more closely controlled logical approach intellectual processes. The need is to be able
for identifying factors and interactions of to analyse a problern so that the concepts
factors in stages which can be examined in involved are clearly defined and capable of
matrix form through partial solutions to being brought into use, with due regard to
combined partial solutions and on to a final their interrelations, in the subsequent stages
design. A system of logical analysis devised of synthesis in solving the design problems.
at the Atomic Weapons Research Establishment
at Aldermaston shows a somewhat different Put into more everyday terms, these
logical system in which at any stage the concepts are the items of scientific and
problems are to be examined in the light of technical knowledge which constitute the
six basic elements: stage of achievement, state of the art, the facts which must be taken
place of operation, time factors, resources, into account such as materials data, and
methods of achievement of the design, and the theoretical factors such as dynamic and static
justifications of the design. Considerations of properties, and knowledge of processes, etc.
methods of improvement are then to be applied At every stage of the design process, except
at each stage. In the USA, McCRORY, perhaps that of the mathematics of optimization,
WILKINSON and FRINK (1963) have compared adequate availability of the requisite infor-
scientific research methods with the steps of mation is essential. This information is to be
determining need, need analysis, design found in the literature, the scientific and
conceptualization, determinations of feasibility, technical papers published in the joumals,
and final production. in data sheets, research reports, conference
It is thus clear that many people are papers, etc., and only too often unrecorded
striving to bring the complex and often know-how of individuals or organizations. To
apparently intuitional processes of design into obtain the recorded information it is necessary
sharper focus, so that the steps involved can for the engineer to read, organize (classify)
be identified and formalized. Many of these the knowledge obtained, and store it in a
97
manner such that it can be retrieved when become information-minded and demand adequate
needed. The methods of storage and retrieval sources of information; otherwise he is
must also be suitable for converting previously dependent on the stock of knowledge acquired
unrecorded information into a retrievable form; during his training or obtained by chance
this means that the methods must be able to consultations with others. In other fields,
handle detailed items of information, and not notably chemistry, there has always been a
just books or bibliographies. Finally, there traditional respect for the literature, for the
are the problems of isolating, understanding avoidance of duplication of work, for the need
and perhaps formalizing the processes of to climb on the shoulders of others. Engineers
synthesis in design; for an understanding of have lacked an adequate number of funda-
such problems at present only such methods mental joumals and have instead had too
as those of Jones' interaction matrices can be many indifferent trade joumals which tend to
relied on to stimulate individual insight; contain either low-level generalized articles
further progress may depend on research on or a plethora of snippets of information, all
the psychology of creativity, but some lines interspersed with an excess of advertisements.
of attack are becoming evident. If an improvement is to be obtained, the demand
The first problern is therefore the avail- must come from the engineer himself.
ability of information. The engineer has tended The individual cannot however be expected
to pay insufficient attention to the literature: to survey any large range of original literature.
partly because his traditional 'practical' What is needed is a comprehensive record
approach has engendered a distrust of other of available information, and this is provided
work; partly because the lack, in many fields, only by an abstracts journal. Unfortunately,
of a growing body of unifying theory has made engineering is even worse served in this
it difficult to see how much available know- respect. There is no comprehensive abstracting
ledge fits into a pattem; and partly because service in English; the Technisches
of the emphasis often placed on the apparently Zentralblatt, in German, is fairly good. The
overriding factors of local conditions and the Engineering Index is still far from complete
possibilities of new materials. This encourages and provides almost only the titles of papers,
specious arguments to the effect that each indexed under a large number of alphabetically
design is essentially a new problern to which arranged headings which, even with cross-
the literature cannot be adapted, that pre- references, are tedious to search unless the
liminary reading will prejudice creative subject is quite straightforward. The Bri tish
thinking, and that theory is inferior to practice. T echnology Index covers some four hundred
These attitudes are undoubtedly changing, British journals, but is again an index (even
but not fast enough to meet the demands though well permuted) and not an abstracts
of competition. Rapid technological advances joumal. Other abstracts joumals cover, with
are being made in all countries, and only an varying efficiency, some specialized fields,
adequate knowledge of such work will provide for instance Electrical Engineering Abstracts,
the necessary basis for advance in Britain. Nuclear Science Abstracts, etc. Compared with
It is only an aceidentel situation when practical the comprehensiveness of Chemical Abstracts
progress is not based on sound theory. It is (which of course covers much chemical
well recognized that engineering design is engineering), the resources at the disposal of
the spearhead of technological advance. the engineer are quite inadequate. The only
Creativity is essentially a process of making remedy for this situation is for the engineers
new combinations of known pieces of know- themselves, through their professional insti-
ledge; a new idea is not just imagined, it is tutions, perhaps, to demand the necessary
produced by synthesis, or at least by analogy services; with sufficient demand the cost
with known facts. The engineer must therefore to the individual will not be too great.
THE ELEMENTS OF DEMGN 99
If good abstracts are not yet possible, of single words (keywords or descriptors)
either for organizational or financial reasons, whose subsequent conjunction in response to
then at least greater attention should be paid a similarly represented question is expected
to the use of data sheets and similar sources to impart meaning, can be considered adequate.
of information, as available, for instance, from
Even the simplest item of information
the National Engineering Laboretory, the
is however more complex than the mere conjunc-
Royal Aeronautical Society, etc. Even these
tion of isolated words. Meaning is present
do not seem yet to be as generally used as
only in statements or other forms of structured
they might be.
information. The simplest representation of
Assuming that the engineer has become structure of this kind is to be found in methods
'information-conscious' and that an adequate of classification. The placing of terms in
amount of detailed information is being groups implies close relations between terms
collected from all useful sources, both in this within one group, and other relations between
country and from abroad, the next stages are higher and lower terms or between one group
those of storage and arrangement so that as and another. The methods of morphological
design requirements become clear it will be analysis of design problems, with subsequent
possible to retrieve any desired items from steps of synthesis, are similar to those of the
the store. The great increase in the amount of necessary analysis for classification and of
scientific and technical Iiterature in recent recent methods of classification by synthesis,
years has led to the realization that the waste- known as facet analysis. This method consists
fulness and possible duplication of research in analysing a subject field into a suitable
if the knowledge cannot be recovered when number of groups (facets) whereby each group
needed is retarding progress; people speak comprises terms or individual concepts with
of the 'literature explosion', though it is the same general character, e.g. materials,
probably a matter more of the amount of know- properties, processes, devices, theoretical
ledge available, in proportion to the number concepts, and perhaps more generat groupings
of people to whom it is useful, than of any (time, location, uses, etc.). There is no Iimit
sudden increase in the amount of knowledge to the number of facets which may be
or papers written. Two conferences with inter- established to cover a given subject field,
national support (in 1948 and 1958) have but usually something between eight and
emphasized the problems of information twenty is sufficient. The headings may be
retrieval. There are, basically, two requite- made quite specific, e.g. raw materials,
ments: how to analyse information reproducibly, secondary materials, lubricants, etc., inter-
and how to select relevant information mediate products, testing equipment; separate
accurately on demand. The first problern is facets may be made for more detailed
that of semantic analysis and classification; distinctions, or these may be made separate
the second that of using cognate methods to parts of one facet as long as no illogicalities
process a given question and obtain the are introduced. A complex subject is then
required answer from the store. In the USA, specified by citing the requisite terms from
there has been an exaggerated faith in the the facets, which have been arranged in a
capacity of the Computer to do both tasks, suitable order of successive subordination
but the methods have not been strikingly or other logical arrangement, and combining
successful. Many attempts have been based on them in that predetermined order; with suitable
the use of information as set out in written coding of the facets, a unique classification
language; but researchers and technologists code is produced for each complex required.
do not necessarily write weil or clearly, and The construction of the facets entails methods
neither linguistic analysis (as attempted with of analysis similar to those proposed by Jones
Computers) nor the mere selection of a group for defining design factors. Facet analysis
100 THE DEUGN METHOD
could do even more, since one could handle is a Straightforward process. Such classifi-
both materials and properlies and also inter- cations have been found to work weil within
action factors, known partial solutions, etc. relatively restricted areas of knowledge.
The primary use of a classification of A good description will be found in a book
this kind will be to record the available by VICKERY (1960).
information of all types. Thus there might be The identification of parameters and
facets for materials (e.g. iron, steel, oil, parameter steps in morphological analysis
plastics), static properties which could is equivalent to a second stage of faceted
include shape and form (e.g. wire, tube, rod), classification having more abstruse facets of
dynamic properties (e.g. Ioad, strain, stress, design and performance specifications details,
resilience, strength, stability), scientific so that synthesized possible partial solutions
Table 12.1. The Nine Categories of Relation
Temporary Fixed memory

Cognition
memory or evaluation
Concurrence
Concurrence Self-activity Association
or
recognition /e !• /; Increasing
conceptual
Not-distinct
Dimensions
or Equl valence
States Appurtenance clarity
convergent /=
/+ /(
thinking
Distinct or
Distinctness Reaction Functional
divergent
/) I· dependence
thinking
/:
Increasing associative memory -----
properlies (e.g. density, elasticity), mech- can be compared with the coding of the general
anical processes, devices (e.g. bearings, performance specifications. Such methods
cranks, shafts, valves, gears, screws, springs, will not of course eliminate any of the intel-
condensers), theoretical concepts (e.g. fatigue, lectual steps of design, but could at each
factors of safety, Poisson's ratio), and also stage assist creativity by providing the facts
perhaps a general facet to cover such terms as in organized form, so that constant reference
tolerances, standards, specifications, etc. could be made to all known information.
It will be seen that the classification can be The simple faceted method is not suitable
devised to fit any given set of requirements for more complex fields where the subject
and will enable accurate classification to be matter covers different disciplines; in such
achieved not only for objects, but also for areas, a concept of lesser importance in one
complex devices, processes and more abstract context may acquire major importance in
types of information. The retrieval of desired another context, and the synthesis of a complex
information, by a similar process of coding item becomes too difficult. Research is now in
the requirement and searching for a match, progress by the author, and by another group,
to discover new principles for dealing with I= Al= B expresses some degree of
more general fields. The facet principle is equi valence, up to full synonymi ty, or
confined to arranging concepts initially as the application of the concept A as
things, processes, abstract terms or properties, the concept B (e.g. a fuel cell as a
and distinguishing Ievels of complexity; such power source)
groupings then allow for the organization of
Ievels (in a different dimension!) of class I+ Al + B expresses the posi tion in time
terms, and, in separate schedules, of hetero- or space (e.g. A is at a given time or
geneaus group terms. Complex items of know- place or position B), or the state B of
ledge will again be produced by methods of A (e.g. a salt in solution, a meta! at
synthesis, but this time with the insertion of a given temperature, a plate with a
specific relational signs between concepts positive electrical charge)
(instead of relying upon a predetermined order I( Al( B expresses that B is a physical
to imply the relations). These relations have property of, or part of, the concept A
been deri ved from considerations of experi- (e.g. steel strength, stator winding)
mental work on the psychology of thinking, and
it has been surprising that the categories of /) Al) B expresses the distinctness of B
relations by which people construct (or synthe- from A, as in the case of substitutes,
size) complex ideas are in fact relatively imitations, etc. (This relation is
simple, only nine basic categories being rarely needed)
required. These categories of relation derive
from the interaction of three stages of mental /- Al- B expresses any action of B (a
memory and three stages of clarity of per- process or thing) on A (e.g. plastic
ception, as shown in Table 12.1. subject to moulding, Iever acted upon
by a spring)
The names suggested for the categories
of relation are arbitrary, and the typewriter 1: Al: B expresses A causes B, or B
symbols (partly mnemonic) are for convenience arises out of A (e.g. fatigue causing
in diagrammatic representations. The symbols fracture)
can then be used between concepts to
represent meaning in s t r u c tu red form It is thereby possible to write any complex
(F ARRADANE, 1963). statement or subject in diagrammatic form with
The relational symbols have ranges of concepts interlinked by the relations. The
meaning which may be exemplified (but not result may be simple linear representation,
defined) as follows (A and B are in each case or branching or circular structures. There is
two concepts to be related): psychological evidence that the mind does
not deal with more than at most four relations
18 Ale B can mean A in the presence of of one concept with other concepts, under
B, or A in the bibliographical form of normal circumstances, and this means that a
B (e.g. engineering dictionary) two-dimensional network pictured on paper is
adequate for delineating even complex sub-
I* Al* B expresses intransitive activities jects. In diagrams, a relation set vertically
(e.g. man walking, bridge collapsing,
between two concept words implies that the
lamp shining)
upper word is in the A position to the lower
I; Al; B expresses an indirect or ab- word (B position) as in the examples of
stract property B of the process A relational meanings given above.
(e.g. design symmetry), or the agent B A Straightforward example would be
of the process A (e.g. compression 'apparatus for fatigue testing of 7075-T6
with a piston) aluminium alloy in vacuum', which is analysed
102 THE DE~GN METHOD
as in (1) below (note that logically it is the consultation; this classification may be the
alloy which is in vacuo, not the testing, and same as that used for Controlling concepts
that fatigue is an indirect and not a normal for the more complex methods of analysis and
physical property). recording of information, as described above.
A more complex example would be 'com- In the examples analysed above, the information
parison of fixed-bed and fluidized-bed heat store will enable this (and no doubt other)
exchangers for utilization of the lew-potential type of fatigue testing apparatus, or heat
heat of waste gases', which is analysed as exchanger, to be located (in a paper) for
in (2) below. information. At higher stages of design syn-
The actual terms used would of course thesis the types of information, though more
be drawn from the pre-established classification, complex, will still be analysable, and storable,
and coded as desired. For retrieval of desired by the relational indexing method. It is possible
information from the material stored, the also to introduce numerical data into these
enquiry would be coded and a match sought; analyses, to some extent.
the enquiry might prove to be part of a more A more sophisticated application of the
complex subject stored, or vice versa, It should classification is also possible. Where a given
be possible to extend the search by 'browsing' desired item of information is found not to
at any point for higher or lower generic terms, exist among the stored records, a close
or cognate terms, if the initial search did not parallel, which may suggest a useful alter-
produce exact results. Rules are also being native, may perhaps be located by changing
sought for equivalents of condensations of one concept at a time in the structured analysis
meaning or logical jumps such as are often of the information required, the change being
found in the way questions are asked, since to a higher or lower generic term or to a
less important terms in a statement tend to cognate term, and repeating the search. Organi-
be omitted. zation of the store of information on a Computer
The design process, after need has been will of course enable various types of search
established and the requirements have been to be made very rapidly.
analysed as far as possible, then consists of The need for information does not how-
a series of steps or levels of synthesis and ever stop at the primary stage of initially
evaluation. At each step, information is available knowledge, or state of the art. The
required, the type of information becoming stages of synthesis of a design will succes-
more complex as the processes of synthesis sively reveal complexes of various sorts
of design proceed. The storage of the simplest whose compatibility of components and
information (e,g. properties of materials) may feasibility need examination; the existence
be satisfactory in the form of alphabetical of previous werk on such complexes or on
erdering of data sheets; if the possible range nearly similar or analogous situations must
of say devices (e.g. gears, heat measuring also be traced, if possible, so as to save
devices, etc.) is not initially known, then a duplication of effort. Only a highly sensitive
classification system must be available for method of analysing and storing information
(1) Aluminium/:Alloy 7075-T6/; Fatigue/-Testing/; Apparatus

/+
Vacuum
(2) Gasesi;Heati;Potential, low

I= /-
Waste '1' . I H E h I Fixed-bed } 1- Comparison
Uh tzahon + eat xc angers ; Fluidized-bed
will cope with these more complex situations. according to the relations involved. This
There is furthermore the possibility that would be equivalent, eventually, to a new
the methods of analysis of information may be mathematic 'logic' of induction or creativity,
adaptable to simulating steps of synthesis, It might be possible to show quite simply
at least for early stages. Research is at whether devices or processes were compatible.
present being envisaged to establish rules, To conclude, therefore, the problems of
on the basis of psychological investigation the design process are those of creativity in
in the field of decision making, problern solving general; a formalization of the creati ve task
or logical reasoning, for combining two known can be achieved only by understanding and
items of information. Thus if a piece of inform- imitating mental mechanisms. This assumes
ation analysable as A/ BIC (i. e. the terms A, the adequate exploitation of available infor-
B and C in some relational structure) is mation; full awareness of the need to obtain,
known and if C/ D/ E is also known, then store, retrieve and use existing information
rules for combining them will be sought, for is therefore essential.
example as AlB/ClDlE or in some other way
Chapter 13
PATTERNS
A.D. Newman
lntroduction Computer Programmes as Graph Nets

A design could be said to be a pattern Now a computer programme can be looked
created by man, with the purpose of meeting upon as a graph of a special sort. In general it
some specified requirement; such purpose, consists of a number of 'node' points linked by
perhaps, being aesthetic, as in artistic designs, directed connectivities. Usually in a computer
or severely practical, as in the case of the logic programme most nodes are connected only to a
of a digital computer - or a combination of the pair of others: one by a connection to it and
two, as is the case in architecture. But what- one by a connection away from it. But always
ever the requirement a design is essentially the some nodes have several others directed to
result of a creative act. them, and some, two directed away from them.
Since a design is a special class of pattern, This is shown in Figure 13.1. Here the network
there is relevance in speculating on the nature is active and sequential. At any one step one
of the pattern. It is a recursive thing, a hier-
archical thing. Any pattern in general is built -------
from other, lesser, patterns, each having a
relationship with each other. A relationship
itself has something to do with structure and is
.,.._ ___ _
----
surely highly restrictive. The final judgment,
for example, on the distinction between a mean-
ingless set of marks and a pattern is the prero-
gati ve of man, so evidently the relationships in
the structure of a pattern are in some ways an
analogue of human thought processes. Hence Figure 13.1
one passes from design to pattern, and from
pattern to thought processes. node is active; at the next step this node acti-
vates the node to which it directs. Where a
Thought Processes node points to two nodes, only one is chosen to
CHOMSKY, in his latest (1959) approaches be active at the next step, the choice depending
to an understanding of the grammar of human upon certain conditions.
language, makes the obvious yet profound Another setting of the programme net is
reflection that speech is clearly the most obvi- shown in Figure 13.2. When a node such as A
ous pointer to the nature of thought - much more is active, information passes through it to a
so than written communication. Speech is store, either directly from a store or from sev-
essentially serial and would appear to be the eral stores via some sort of computing device.
kind of structure known as a process. Such a A node such as B is vitally different. In this
process is related to modern graph theory and case no information pass es but the next acti vat-
again to modern computer programmes. ed node is either X or Y dependent on the
105
106 THE DESffiN METHOD
Stores Stores (3) The important kind of node is the kind

A called B in the diagram - i.e. a node which
when acti ve leads to one or several, according
to the input criteria.
(4) It is not necessary to have the 'double'
network.
An active graph net can be designed with

connections of more than one kind which can
undertake any possible computing task. In this
case the information storage, the programme,
and the computation are all one entity. Figure
13.3 shows such a system. In this the connect-
ivities are +1, -1, and +Y:l. Anode becomes
acti ve if the sum of the connecti vities from
active nodes is 1 or greater. If the A's and the
X y B's are made active according to two 3-digit
binary numbers, then, in due course, (actually
Figure 13.2 eight steps later) the S's will become active
according to their sum. Thus, if the starting
states are
nature of the infonnation reaching the node from
some other store. Four points should be noted:
(1) It is not essential for only one node to B 1 = 1, B 2 = 1, B 3 = 0

be active at a time.
(2) The nurober of paths 'fanning in' to a the final S states will be
node or 'fanning out' from it need not be limited
to two. sl = 1, s2 = 0, s3 = 0, s. =1
: :
1 1
r
1 1
• • • •
I I
Al
t1
:~
54
A2
AJ
81._----------~
82.._------~--------~------.-------_.
83.-----~------.-----~------~----~~-----.----~------_.
Figure 13.3
Although different things occur if other nodes supposed that he does this by means of inter-
are marle active as starting nodes, the process- connected neurones as nodes in a programme
es which occur as the result -of activity in thisnet of the sort discussed above. If this were
net are always closely related to addition. the case, the programmed procedures would be
The net of Figure 13.3 is not recursive, determined by the directions and strength of the
but there is no reason why s., s2 and s3, for connectivities. Learning a new procedure would
example, could not lead to A" A2 and A 3 , and be done by making new sets of connectivities,
it is possible to have a system in which any and procedures adapted to new purposes by
node, in general via several others, points to adding or modifying a few connections only. It
every node in the network. In such a case a would be expected that in a living creature a
given activity state can be the result of a number of Straightforward separate procedures
considerable range of input states. On the would be developed first, but as time passed
other hand, since the effect of an input pattern these would become more knotted and inter-
to the system willbe dependent uponits activity connected, and links would occur between the
state within the network at the time of input, knots. Thus the output states produced by
acti vity of input nodes would depend more and
• • more on the activity of internal nodes and hence
on the content in which the input occurred.
• Now it is not necessarily implied that the
brain actually operates in the manner of this
system, but it is suggested that the system 's
behaviour is such as to allow reasonable specu-
lations about concepts such as meaning and
likeness.
'' All the nodes in a tightly interlinked recur-
'
" •
sive part of an active net are very much of a
definite whole. Tagether they constitute an
entity closely analogaus to the Jungian idea of
a complex - in a sense a tightly knitted part
• • • of a net constitutes a 'concept'. The 'meaning'
Figure 13.4 of a given input set can, in this model, be
looked upon as the complex with which the set
a wide range of output activities could be pro- is directly associated. But in a real sense this
duced by a given input, depending upon context. depends not only on the position of the initially
The net of Figure 13.4 clearly has two activated nodes in the general net, but also on
parts, A and B. Within A or B there is heavy the state of activation within the net at the time
interconnection, but there is a loose A to B of input.
connection. If the dotted link were marle there In this way, meaning is not absolute. It
would be loose A to B and B to A connections. depends on the net and the relationship of the
Active graphs representing meaningful proce- inputs to the net which in turn depends upon the
dures break up in this way. In practice there experience and the general context in which
can be a hierarchy of steadily larger groups, the stimulus occurs. Meaning therefore becomes
the connections becoming progressively more a private thing, an ephemeral thing, and com-
tenuous. munication between men and their colleagues
relies upon basic similarities between different
The Concept of Meaning men and their past experiences. Likeness can
The concept of 'meaning' may now be be looked upon in a similar way, for two input
examined in terms of the net programme. Clearly, nets can be said to be similar if they communi-
man carries out procedures, and it could be cate with largely common parts of the net.
Outputsets can be defined in an analogous way. design is a pattern having the purpose of meet-
Looked at thus, 'likenesses' depend very much ing some specified requirement. Since the
on context, and are again essentially 'private'. requirement is man-specified it will have a
Likeness, like beauty, is in the eye of the structure and can be further specified in the
beholder, and the eye changes continuously. form of a programme process. Once this has
If a pattern must essentially be meaningful, been done, the design is in a sense complete.
it follows that what is a pattern to one man may The problern then is how to specify requirements
be a meaningless jumble to another, or perhaps in the form of a programmed process, and arises
even to the first man in a different time or in a from the fact that the specification will not be
different context. It is perhaps because of this given in the form of the structures, but rather in
that any new form of art tends to be 'meaning- the form of words or drawings which, as input
less' or incomprehensible to people who have to, or output from, the net are tightly linked to
not followed the developments that led to it. certain complexes, and which therefore label or
Any pattern that is created as the output of a represent the complexes. The actual designer,
net programme is evidently closely related to whether man or machine, is beaten from the start
the structure of the net, and hence very heavily if its make-up of experience does not contain
restricted in possible form. Even so, the total similar complexes 'labelled' in a similar way.
number of output patterns that can be produced
The next thing is to discover whether the
from a stable, active, net containing 10 11 nodes
various sub-structures can exist meaningfully
and several orders more of connectivity, is together - that is to say whether they are link-
immense. If the connectivities evolve (change
ed, or can be linked in a way analogous to the
with time) the possible output variation is even designer's thought-structure net. This can be
more tremendous. discovered in one of two ways, either by experi-
Thus even if the 10 11 neurones of the brain menting with context, or by a small evolution in
formed a stable set, the brain would be able to the thought-structure net. Either way, of course,
produce a great many behaviours, and it would calls for a degree of random change. But if the
be virtually impossible for its activity during design pattern is worked out as a hierarchy, the
two different periods to be identical; thus any required design will involve perhaps a small re-
pattern produced by such a system is truly a arrangement of concepts or complexes already
'creation'. In another sense, however, it is not existing in the designer's thought-structure, and
new in that it has arisen almost entirely out of at most small changes in the sub-patterns.
past experience. A little thought shows that, Design is possible in a reasonable time simply
looked upon from the point of view of a struc- because the required amount of trial and error is
tures procedure, the pattems of many designs
very small, and only possible for a designer
are remarkably similar. Thus a design for a with a thought-structure adequately developed
hydraulic system will be found to have a very
as a result of a suitable experience.
close parallel on the one band to some mechani-
cal system, and on the other to some electronic Where the purpose of the design is to please
system. It would also seem that the invention aesthetically, different problems arise. For
of a remarkable new design will be found to instance, the design must be meaningful to the
have introduced a very small, though vitally observer, and must thus match the structure of
important, change into a previous system. his thought processes. Clearly it must do more.
Design can, perhaps, only be truly carried out As a conclusion, a postulate about this 'more'
by analogy. is considered.
The Tasks of Realization The Aesthetic Postulate

The above ideas suggest the ways in which The brain is frequently thought of as a
design could perhaps ultimately be carried out device essentially to match motor or autonomic
by a machine. It was suggested earlier that a activity correctly to the stimulus. Correctly,
THE ELEMENTS OF DEaGN 109
in this context, means 'in a way conducive to system bringing news of more subtle alarms. In
survival'. This in turn is related to correct addition the brain needs information that 'all is
body functioning, which once again is measured well' and that a given problern is being solved.
in terms of activity in the autonomic nervous What would be the best measure of 'all being
system. well'? Surely a lack of nodal acti vi ty in an
A sensible way to design a brain therefore alert brain. Thus it would seem plausible that
would be to make adaptation of net structure any pattem which was not only meaningful in
dependent upon information received from the matehing thought processes, but also was of
autonomic system. The most obvious and such a nature that it minimized nodal activity
pressing information would clearly come from would be 'pleasing' to observe. If this were so,
darnage occurring to the outer structure of the beauty would truly be in the eye of the beholder,
body - this, of course, is pain. Other inform- for appreciation would depend basically upon
ation reaches the brain from the autonomic the actual thought- structure.
Chapter 14
CREATIVITY
Geoffrey H. Broadbent
lntroduction concerned more with the improvement of

The inherent difficulty of heuristics, the existing systems than with the design of new
study of design method, is that one becomes ones (ACKOFF, 1961). It will therefore he
fascinated by means, rather than with ends. appropriate for certain classes of design
One tends to think of 'problem-solving' instead problem, as systems engineering will be for
of 'design' and to concern oneself with rational certain others, but neither will be particularly
procedures instead of creative acts. All design effective when innovation is required.
processes are ba~ed on scientific method. For On the whole, design methodology has
some time now, the aim has been to devise tended to shy away from innovation. It is
rationalized procedures analogous to the comparatively easy to devise a design process
mathematician's algorithm, sets of instructions which will lead to the optimization of accepted
for solving particular problern types which methods, but innovation involves uncertainty,
require no access of creative thinking and therefore seems less amenable to
whatsoever. Some theorists such as DA VIS systematization. There are other reasons too
(1958) have suggested that if no such for resistance to innovation. It implies threats
algorithm exists, then the problern is to security, often in very real terms - earning
unsolvable. As GosLING (1961) has pointed capacity, established working patterns, even
out, design methods have their vogue and then the jobs of those who are resisting change.
fall from esteem. If they really are of value, N aturally, if they have helped to develop
their development is continued into a existing systems, and this includes rationalized
specialized branch of knowledge, which will design ptocedures, they do not wish to see
then be useful in the understanding of them superseded, with consequent loss to
particular classes of problems. themselves of prestige, status and recognition.
As such a vogue recedes, others advance
to take its place, and recently there has been
an increasing emphasis on human values in Creativity in Design Method
design. In part, at least, this has been Yet it seems to me that current design
encouraged by Operational Research (OR), a methods are less efficient than they might be,
discipline which is complementary to the because they fail to build in the 'unreliability'
Systems Engineering which Gosling advocated. of the human operator, which is the only real
Both are concerned with the design of systems source of creative ideas, good and bad. The
in which humanoperatorswill control machines, most important innovations in the history of
but OR concentrates on the analysis of ideas and inventions have been achieved, on
activities and procedures; therefore it places the whole, by an access of creative thinking
particular value on the contribution of the into some version of scientific method. It
human operator. Yet on the whole, OR is should therefore be possible to devise further
111
types of design process, additional to systems be superior whenever value judgments, form
engineering and OR, like them akin to recognition, association of ideas and the
scientific method, but which will have built-in generation of unpredictable relationships are
mechanisms to encourage creative innovation, concerned (WALTER, 1953). In other words,
where this would be appropriate. the most effective design process will utilize
If DEWEY'S dassie analysis of scientific the brain and the computer, working tagether in
method is taken, and broken down into its a symbiotic relationship, each acting in ways
constituent parts, the fotlowing five steps are which are not accessible to the other. And
obtained: provided that steps (4) and (5), rational
elaboration and corroboration, are carried out
(1) The occurrence of a difficulty. with the utmost analytical rigour, it will be an
(2) Definition of the difficulty. advantage if step (3), the creative leap, can be
(3) Occurrence of a suggested explanation as wild and imaginative as possible.
or possible solution.
(4) The rational elaboration of an idea.
Creativity in Art and Science
(5) Corroboration of an idea and formulation
There is an extensive literature on the
of a concluding belief.
subject of creativity in art and science, which
traces the ways in which the mind prepares
Dewey was concerned with an analysis of the itself for this leap (see GHISELIN, 1952).
complete act of thought and, often, he has been WALLAS (1926) outlines four stages in his Art
misrepresented as advocating a totally rational of Thought: preparation, incubation, illumination
approach to the subject. But he is careful to and verification. In the preparation stage, the
poin t out that his third step, explanation or brain is programmed with all the facts relevant
solution, is a matter of inference, which to the problern - these, of course, will be avail-
involves 'a leap, a jump the propriety of which able to the brain from the computer after Dewey's
cannot be absolutely warranted in advance'. step (2). Once the facts have been assimilated,
In OR terms (as in Ackoff), the parts of an the problern is dismissed from the forefront of
operation are designated as 'black boxes'. the mind, and incubation takes place at an
OR is not particularly concerned with the unconscious level. During this period, one will
contents of these boxes, but by thinking of be engaged in a variety of other acti vi ties,
Dewey's method in operational terms, the types mental and physical. At times, of course, the
of black box which will be appropriate for each problern may emerge into the conscious mind
step can be defined. Steps (2), (4) and (5) again, or into a threshold state between
clearly are strictly rational. The most efficient consciousness and subconsciousness; one will
device available for operating rational be thinking about and araund it, but the
processes is the computer, which will be creative 'leap' may occur unexpectedly at any
entirely appropriate for these steps. But step time. Illumination, in fact, is likely to strike
(3), according to Dewey is irrational, and at an unguarded moment, when mind and body
therefore it requires a different type of black are completely relaxed. This fact is confirmed
box. It is precisely the ways in which the in manv farnaus accounts, such as those of
human brain differs from the computer which Poincare, Gauss, Mazart and Helmholtz
make it the most effective device for use at (HADAMARD, 1949). After the leap, of course,
this stage. In terms of leaming capacity, the solution is subjected to rigorous analysis
mernory, predictability and the operation of an and verification by experimental test.
algorithm, the computer can always be made These stages are common to artists and
more efficient than the brain. But the brain will scientists; the great creative acts which have
shaped the twentieth century, and given notes is repeated in various ways, but always
contemporary culture its unique flavour certainly in the same order (STUCKENSCHMIDT, 1959).
have them in common. I take these to be the 'I believe', he wrote 'that I have found a key'
discoveries of Relativity, Cubism and the (STEIN, 1964). The flash again, and
12-note Method in music. significantly enough, Schoenberg's flash is
Einstein 's struggle is weH documented analogous to Einstein's, for he says (1950):
(see, for example, WERTHEIMER, 1945). He 'The two-or-more dimensional space in which
worried for six years and more - not over the musical ideas are presented is a unit.' The
solution to his problem, but over the actual manner in which Schoenberg devised his Method
nature of the problern itself, which was how is an object lesson to anyone who is engaged
to reconcile Newtonian mechanics with in the methodology of design. There was no
Maxwell 's field theories. At an intermediate question of deriving an algorithm for composing
stage, during incubation, he realized that music; Schoenberg wrote intuitively, with
'light is the key', and then fina1ly, in the conscious avoidance of the tonal system, and
seventh year, the flash came when he then analysed what he had clone. Only then did
understood that time is the fourth dimension. he devise a set of rules for composition, and
After this, the rational elaboration of his idea their validity has been proved since by the
took only five weeks, and the experimental crucial effect the Method has had on the
verification was left to others. development of music in our time.
There are curious para1lels in the It may be argued that these epoch-making
discovery of Cubism, not least in the fact that, discoveries in Physics, Painting and Music
at an intuitive level, the painters also were have little to do with everyday activities in
concerned with a fourth dimension (BARR, design, that their authors were different in kind
1933). After years of imbibing 'influences' from from everyday designers. That may be true,
the entire history of art, they realized, during certainly one doubts that Einstein, Braque or
the incubation phase, that Cezanne was the Schoenberg needed a systematic design method
in rationalized terms. However, the use of
key - and then the flash came when, for the
computers in design can raise the efficiency at
first time, Braque was confronted with
Dewey's steps (2), (4) and (5) to such levels
Picasso's Demoiseiles d' Avignon. He saw that
that, if a balanced operation is to be achieved,
a painting need not look like anything at al1,
the creative leap at step (3) must be made
but could be a completely new thing in its own
right (COOPER, 1956). So the Cubists began equally effective.
their tremendous exploration into the nature of Creative Techniques
visual reality; the rational elaboration of their Many techniques are available for
idea took about a year, and its validity has enhancing creative capacity. One tends to view
been corroborated in the history of the visual them with suspicion because they derive from
arts ever since. the advertising industry and savour too much
The musical equivalent fo11owed a similar of a Madison Avenue approach. BAKER'S
pattern. Composers such as Wagner and Y out Key to Creative Thinking is fu11 of
Debussy had been breaking down the old tonal practical tips in ideas-generation, bu t the
system and Arnold Schoenberg realized that examples he quotes, such as formulating a
there were inherent difficulties in writing music detergent of two-colour granules, packaging it
in no key at all. For six years or so, he stopped in different ways for different purposes and
composing altogether, letting the problern selling it with gift offers, are precisely the
incubate, and then, in 1923, he wrote a little techniques one has learned to distrust from,
piano piece in which a basic row of twelve say P ACKARD'S The Hidden Persuaders.
The psychology of creativity is not as views on creati vity were criticized severely by
helpful as it might be either, because it tends Wertheimer in his formulation of Gestalt
to be theoretical rather than pragmatic. The psychology, based on the concept of
standard tests on creativity hardly ring true structurization. In Gestalt theory, a problern
with those who are engaged from day to day in consists of an incomplete structure, and to
creative activity. Typical tests, as listed by solve it, one must comprehend the relationships
HYMAN (1963) take a form such as: 'How many between parts of the structure. The gaps are
definitions can you gi ve for the word "holt"?' then closed by drawing relevant material from
or 'List all the uses you can for 5,000 used red one's previous experiences, which are stored
bricks.' Other tests might involve the relation- in the brain (KOFFKA, 1935). Gestalt draws on
ships of simple geometrical shapes, the the idea of the Schema, which was conceived
completion of a story in certain ways, or by Head initially and developed also by
problems in numerical manipulation (GETZELS BARTLETT (1961). Many psychologists find it
and jACKSON, 1962). Several experimenters inadequate now, as a model for mental
have claimed, according to Hyman, that on test, processes, probably because of its Gestalt
people who were told to 'be creative' performed associations, but in Bartlett's form at least i.t
as weil as those who had been educated to it. explains satisfactorily, at pragmatic level, the
This seems to me inevitable when the tests techniques which are available for tapping
themselves are concerned largely with one's creati ve potential.
ingenuity, which I take to be only one BartleU was concerned essentially with
component of creativity. memory and the ways in which, over the years,
In fact, a clear definition of what the one's ideas develop and change. He took the
creative act comprises, in terms of mental schemata to be arrangements within the brain
processes, is lacking. Three views are held of past responses to stimuli and believed that
generally in psychology each of which may memory changes result from the interactions of
represent a part of the truth. The first of these the schemata with each other, and with new,
is the Determinist view that thinking is amatter external stimuli. The schemata, for Bartlett,
of logic, concerned with the progressive were organized according to one's appetites,
altemation of hypothesis and test. In its instincts, interests and ideas. These were
developed, Renaissance form, this is the basis concerned with the senses, but also with sport,
of scientific method; gather the facts, observe literature, science, philosophy - any unifying
relations between them, formulate assumptions interest into which ideas might be organized.
and test them. Much thinking in the field of When a new stimulus excited th·e senses, it
design method up to now has been rigidly was tested against one's existing schemata.
determinist, and therefore it has excluded the The act of perception, therefore, became a
possi bili ty of more sophisticated techniques. two-fold process: first the physical stimulus
The early psychologists added to determinism formed a sensory pattern of the 'real' world,
the Associationist view of stimulus and which was then modified by interaction with
response. They believed that ideas are one's schemata. At the same time, the schemata
associated in the mind when they first arise, also were modified by interaction with the
and that the creative act consists of drawing sensory stimulus. This led Bartlett to suggest
on these associations in rapid sequences of that imagination consists of free constructions
trials and errors. RUGG (1963) states that on one's schemata, and any technique which
Dewey's aim initially was to reconcile these encourages creative activity will be based on
two different approaches. such free constructions.
Both the determinist and the associationist In many ways, Bartlett's model is
appropriate also for the centrat phases of the Reference '· Typical questions are, in the
design process itself - analysis and synthesis. category of Technological Stretching: 'What
The output from analysis will form the stimulus happens if we push conditions to the limit?
and may be just as 'real', in physical terms, Temperature, up or down? Pressure, up 01
as the sensory stimulus in the act of perception. down? Concentration, up or down? Impurities,
But just as in perception, the sensory pattern up or down?'
is modified by interaction with one's schemata, Classification systems, such as those
so in conception, the analytical pattern will be used in libraries (UDC, Dewey) and the
modified in similar ways also. GREGORY and architeces office (SfB) may be useful in this
BURDIS (1965) have put forward a formal model context, but the most accessible list is probably
whose origin may be traced to similar lines of the Tabular Synopsis of Roget's Thesaurus.
thought (see Appendix to this chapter). Roget juxtaposes positive and negative ideas,
The techniques which are available for grouped in pairs, and the most useful stimulus
increasing one's creative capacity all conform words will be found under the generat headings
to this perceptual model, whereby an analytical of Abstract Relations, Space and Matter. As an
stimulus is forced into interaction with one's example, Abstract Relations are sub-divided
schemata. The simplest of them rely on into Existence, Relation, Quantity, Order,
stimulus words to trigger off ideas, and there- Number, Time, Change and Causation. Under
fore they take the form of check-lists. The each of these headings, there are dozens of
Gregory-Burdis model was related to a range words which may suggest appropriate
of stimuli, including 'challenge'. modifications to an object in design. For
example, under one sub-heading alone one
Check-lists reads:
The danger with check-lists is that the
CHANGE I. SIMPLE CHANGE
questions become too vague: 'How can we 148. INTERCHANGE- N. inter-, ex-change; com-,
improve our product?' It should be possible to per-mutation; reciprocation, transposal,
work out particularly meaningful sets of transposition, shuffling; reciprocity, etc.
questions for any particular design-type, but
even general check-lists can focus attention Each of these, of course, might be a powerful
wi th surprising precision onto the difficult stimulus word, and the possibilities in Roget
aspects of a problem. range from the simplest and most concrete
ÜSBORN (1963) and GREGORY (1963) have ideas to the highest flights of abstraction.
compiled useful check-lists which suggest new Another approach to check-lists is
ways of looking at the problem. Osborn's main suggested by VON FAN GE (1959). He puts
headings are: 'Putto other use? Adapt? Modify? forward the idea that a person tends to order
Magnify? Minify? Substitute? Rearrange? the past in terms of a few major experiences
Reverse? Combine?' As an example, under the which have. affected him in each year. These
heading of Rearrange? he suggests the should be listed, and when he is faced with a
following supplementary questions: 'Interchange situation which demands creative thinking, this
components? Other pattern? Other Iayout? Other situation can be compared with the recorded
sequence? Transpose cause and effect? Change experiences, one by one. For this reason also,
pace? Change schedule?' it is useful to reread old papers, notes, diaries
Gregory lists his questions und er functional even, which will stimulate far more ideas than
headings: 'Economics, Understanding, Practice, are actually written down. Further useful
Technology, Technological Stretching, Cross- check-lists will be found in MA TOUSEK (1963)
fertilization, Guessing the Trend, New Axes of and TA YLOR (1961).
s
Interaction Techniques start by considering the relationship between

Other techniques for stimulating creative desk and chair, in terms of similarities,
associations derive from the use of interaction differences, analogies, cause and effect. This
charts. In the simplest form, one set of factors would lead to a chain of free associations
is plotted vertically, and another set which might suggest a totally new concept in
horizontally; associations are then generated office furniture design.
by plotting all the possible interactions between
these vertical and horizontal sets on a chart. Group Activity
The most potent techniques for ideas-
4ttribute Listing generation utilize the concept of free
This method, dealt with by Taylor, was association, or 'schema scrambling' in a more
devi sed by C RA wF o R D (1954) who abstracts Freudian context. The two best known,
from the object in design a !ist of its physical Brainstorming and Synectics, were both worked
attributes. For a screw driver, these might out initially in terms of group activity. This
include wooden handle, steel shank, wedge- has advantages, of course, in that any group
shaped end for engaging screw, etc. These are can draw on wider ranges of schemata and
then plotted vertically and possi ble variations associations than any individual.
plotted horizontally, in a systematic manner,
against each attribute. In a good field, there Brainstorming
will be no obvious vertical links between This is the brash, Madison Avenue
individual steps in adjacent rows, so that a technique in excelsis. It was devised by Alex
new solution will be generated by selecting Osborn, himself an advertising executive and
one independent variation on each of the the intention is that the members of a group
attributes. shall vie with each other in generating a rapid
succession of ideas. Osborn lays down four
Morphological Analysis basic rules for brainstorming:
This was developed by Dr. Allen Long of (1) Criticism is ruled out.
Long Beach College and Professor Fritz (2) 'Free wheeling' is welcomed.
Zwicky of California Tech; like attribute (3) Quantity is wanted.
listing, it is concerned with separating out the (4) Combination and improvementare sought.
independent variables of a problem. Instead of
taking the actual physical characteristics, the It is essential that group relationships be
designer abstracts the relevant design amiable and relaxed; for this reason, Osbom
parameters for the object. For the screw driver, recommends that the session start over a good
these might be: drive element, transmission lunch. The leader states the problern in basic
element, screw engaging element, etc. NORRIS terms, focused on a single point, and through-
(1963) gi ves good examples. out the session he is at pains to suppress the
criticism of any idea, however crude and
Forced Relationships irrelevant it may seem. The aim, in fact is to
Charles Whiting's method (Osbom; Taylor) stimulate competition in ideas-generation, by
is concerned with the relationships of whole free association, the wilder the better.
objects, rather than with individual attributes Much of the success of brainstorming
or parameters. He quotes the example of an depends on leadership and difficulties may
office furniture manufacturer who might be arise if a group is asked to rework old problems.
concerned with desks, chairs, desk lamps, If these are desperately familiar, von Fange
filing cabinets and book cases. Whiting would reports that the group will simply restate old
attitudes. This is the chief defect of brain- acting on me? What is my attitude to the
storming, but conversely, a good group may supports? etc.'
generate so many ideas that evaluation becomes Direct Analogy - The problern is compared
a major task. Osbom insists that all ideas be with known facts in another branch of art,
written down, and fhat afterthoughts are also science or technology. Synectics quotes the
added to the !ist. This is then submitted for example of Brunel who, faced with the problern
scrutiny, preferably by assessors who took no of building underwater constructions, observed
part in the original brainstorming. The most a shipworm forming a tube for itself as it bored
promising ideas are then put forward for into timber. From this, Brunel concei ved the
verification. idea of the caisson.
Symbolic Analogy - The designer tries to
Synectics
penetrate to the essence of special meaning
In contrast to brainstorming, Synectics on
which he attaches to the problern by means of
the whole is a quiet, contemplative activity, in
some personal symbol. This may be verbal,
which ideas are generated in a purposeful way,
visual or conceivably could take some other
and evaluated, as far as possible, during the
form. In one Synectics session, the group was
session itself. Synectics was devised by the
concerned with detecting the presence of an
Invention Research Group at Harvard Uni versi ty,
unwamted flame in some complex piece of
under the leadership of GORDON (1961). It has hardware. They asked the question: 'What is the
some features in common with brainstorming: essence of flameness?' and eventually, thought
it is primarily a group activity in which, during of it as a 'ghostly wall', which opened up a
a session, personal criticism is ruled out. It whole new range of feasible solutions.
can also act as a great stimulus in the
Curiously enough, the three key twentieth
individual creative act because, more than any
century ideas of Relativity, Cubism and 12-note
other method, it is designed to draw on the
Method share a common symbolic analogy in the
resources of the whole personality. Synectics
space-time concept.
is a complete design method, including analysis,
which is called 'Making the Strange F amiliar', In practice, a Synectics session is
and creative synthesis or 'Making the Familiar conducted systematically by a chairman, who
Strange'; in other words, seeing the familiar introduces the problem, which is then analysed
problern in a new light. This is achieved by a and disc.ussed. At a key stage, there is a
system of analogy-generation, which is the 'purge of immediate solutions', after which
most striking feature of Synectics. Three types attention is narrowed onto one particular aspect
of analogy are identified, (Gordon; GITTER, of the problem. The chairman then asks an
GORDON and PRINCE, 1965; T. ALEXANDER, 'Evocati ve Question ', which will force answers
1965): (a) personal analogy; (b) direct analogy; in · terms of one of the analogy-types; once a
(c) symbolic analogy. ·
fruitful analogy has been generated, its
Between them, these analogy-types are implications are examined in detail. Like all
capable of tapping the entire range of human creative acts, a Synectics ses_sion is cyclic.
experience, whieh is why they are different in If no new viewpoint can be established from the
kind from each other - personal, concrete and chosen analogy, the chairman will guide the
abstract. discussion back to an earlier phase, and try a
different approach.
Personal Analogy - The designer identifies Synectics draws on the whole creative
hirnself with the object in design: 'If I were this capacity of the brain. It is concemed with far
beam, how should I feel? What are the stresses more than mere ingenuity, because analogy
generation, is a very personal thing, depending Einstein seems to bear this out, in a
on the stored associations or schemata which famous statement on his creative processes
have been built up in the brain over the years. quoted by Hadamard:
The brain may very well make apparently
irrational connections which lead to supremely ' .... words .... or language, as they are
rational solutions because, however curious written or spoken, do not seem to play any
they may seem at the time, they have been role in my mechanism of thought. The
subjected to the censoring mechanisms by physical entities which seem to serve as
which the schemata control the input of ideas elements in thought are certain signs and
into the brain, their associations and more or less clear images which can be
subsequent output. The strength of Synectics, "voluntarily" reproduced and combined .....
which is shared to some extent by the other The above mentioned elements are, in my
m ethods, lies in the fact that it taps precisely case, of visual and some of muscular type.'
those thought processes which are inaccessible
to the computer. At one level, the computer For this reason, more than apy other, it seems
might be programmed to 'brainstorm' itself; it clear that, within the design operation a central
could throw up an enormous range of random 'black box' will always be needed which
associations, but the problern of evaluation consists of the human brain, working in
would be greater even than with a human symbiosis with the human body. There seems
brainstorming session. Certainly, it could not to be no other way of building into the design
be programmed to draw meaningful analogies in process the mechanism for ensuring that, when
the manner of Synectics. needed, a solution will be achieved which is
different in kind from the tried and accepted.
The reasons for this become clear if the
The remarkable thing is not that this should
implications of 'whole personality' are consid-
be so, but that for so long, people have allowed
ered in this context. It is well understood now
that creativity is not just a determinist, rational an outworn philosophy of Determinism to lead
them into trying to e 1im in a t e human
matter, but even if the computer could be
'unreliability' in the design process.
programmed to make irrational connections, it
could hardly achieve the interaction of body
and mind which, increasingly, is being seen as APPENDIX
essential to the creative act. One wise and The crude model of creati ve behaviour
perceptive writer on this subject, Harold Rugg, suggested by Gregory and Burdis is shown in
had almost formulated a theory of creativity in Figure 14.1.
these terms, when he died in 1960. He believed
that the creative act involves the whole
personality in a two-phased interaction of body
and mind. This is known to be true of
perceptual acts - seeing, touching, hearing and
tasting - which involve physical as well as
mental processes. It is also believed that the Figure 14.1
creative act consists of drawing together
associations which have been stored in the
mind as a result of these perceptual acts. How ........_ __ _....
then should it be supposed that the conceptual
act also does not draw similarly on mental and The black box is an individual, or a group,
physical processes? or a human-computer hybrid. Input for the
oeeasion is /. This is information of some kind, Some Practical Situations

transmitted in any pereeptual system. The
(1) A highly developed art gives a high value
resourees Ri eontain, amongst other things,
of Z.
aecumulated information. This is the memory
aspeet of Ri. Specifie operations on I and (2) New country provides a low value of Z.
aecumulated inputs are represented by T s• a (3) A bureaucratie structure, according to the
systemalle proeedure whieh ean be programmed. number of decision points, provides a high
Te represents a ereative proeedure. value of Z.
Considering only a systemalle proeedure ( 4) Cross-fertilization increases (Ri + Re).
0 = I, IT 8 , 12 (Ri + Re) (Mi +Me> (5) Increased ego-strength provides an
Value of operation is increase in (Mi + Me ).
(6) Challenge inereases (Mi + Me>.
V2 x I,IT 8 , 12 (Ri +Re> (Mi+ Me>- v,I
(7) According to Maslow, as interpreted by
If the systemalle proeedure is unsatisfaetory T s Repucci, problem-solving takes 'energy'. If
is zero or V2 is too low. A ereative operation there is a low need-fulfilment, energy is taken
is then required away so that problem-solving becomes difficult.
13 /Te, 14 [(Ri +Re>(Mi +Me)- Z] (8) I may come from inside; so also may
changes in Ri and Mi.
where Z is a quantity whieh has to be supplied
before any useful result ean be obtained. This (9) Z may be cireumvented by using special
is some funetion of the state of the art and also 'unlocking' techniques.
of the mieroeulture and maeroeulture. The over-
all value produeed, Vnett• is Validation Status
This model with non-linearity included
v2 X I, ITS, 12 (Ri +Re) (Mi + Me) + qualitatively summarizes most of the relevant
material presented at the Foundation for
vl X 13/ T e• I. [(Ri + Re) (Mi + Me>- Z]- V, I Research on Human Behaviour: DYN AMAR
Seminar, 1965.
Under these eonditions V 3 is likely to be mueh Possibilities are seen for quantitative
greater than v2. investigation in some aspects.
Chapter 15
CHOOSING AND EVALUATING

A.G. Pleydeii-Pearce
lntroduction the evaluation of any particular action is either

Design involves choice. Choosing involves instrumental or arbitrary. In the latter case the
acting. Choosing can be capricious, but a per- concept 'value' has no real application. Actions
son usually believes the choices he makes to are sometimes accorded value because they Iead
be capable of justification. Sometimes he to or are believed to Iead to a desired state of
thinks his choices right; sometimes he comes affairs. Notice I say 'accorded' value and not
to think them wrong. 'Right' and 'wrong' in this 1have' value; although once accorded value, the
context are customarily held to be'value terms', actions have it in the only intelligible sense.
of which the right is to be preferred to the Sometimes they are accorded value because
wrong. Choosing constitutes a 'critical step' in they exemplify a desired state of affairs. The
the human activity called designing. Such a value of the state of affairs itself is not cap-
step may in practice be extremely complex. This able of justification. To justify something is
complexity is apt to obscure the issues involved either to give a reason for it or to say what
in the decision-taking situation. Psychological caused it. Thus 'justification' is u.sed in two
involvement adds its own quota of confusion. different senses, which I shall refer to as
I intend, therefore, to Iook in an abstract way at 'justification (1)' and 'justification (2) '.
the concrete problems of designers working in Justification (2) represents one form of
the field. This method of proceeding, since it explanation, but the two concepts are essen-
avoids immediate concem about a particular tially disjunct; they cannot be reduced one
situation, is less likely to divert attention from to the other. Justification (1) is the primary
the principles elucidated. In addition the sense of justification. If an infinite regress is
increase in simplicity may be paralleled by an to be avoided some reasons must be reasonless.
increase in comprehension. In any event, if the Cause is a fundamentally confused concept,
simple is obscure, the complex can hardly be but for reasons I shall give, the attempt to use
expected to be clear. Of course, as Nelson a causal account to justify an action taken (at
Goodman once pointed out, attempting to make least in one of the central cases of 'cause') is
the obscure obvious is apt to be unappealing: bound to fail.
failure brings confusion, and success banality.
But in a field as obscure as that of value theory Some Distinctions
the attempt at clarification is unlikely to be I shall begin by making a number of
wholly successful. Some areas of interest will distinctions: I shall distinguish between
doubtless remain. actions and sub-actions; acts and projects;
I want then in this chapter to do two things: causes and reasons; conscious and precon-
firstly to discuss what is involved in choosing scious choices. On the basis of these
to perform one action rather than another, and distinctions I shall argue that the only problern
secondly to make some general points concem- for those interested in the design of a value
ing value theory as such. I shall contend that system is what, given a projected end, is the
121
most effective action or actions to undertake. changes in the extemal world. Of course,
This will often resolve itself into a concem when covert actions are performed changes
with sub-actions whose justifiability, once in the extemal world, such as physical
the action has been clarified, can be given processes in the brain, do take place. But it
broadly in terms of efficiency. It may be would, I think, be misleading to describe an
possible to reduce the actions themsel ves to agent as bringing these about, as if he set
sub-actions in relation to some more funda- hirnself to do this. In thinking out the solution
mental goal. If it is possible, then these sub- to a puzzle he does not bring about changes in
actions can be given instrumental justification his brain state. Such changes are part of what
like any other sub-action; they will hence be
it is to be thinking at all.
given value in the only non-arbitrary sense. An action may be composed of both overt
If not, they must be seen as exemplifications
and covert features; for instance in writing
of some desired project; they will then not
only not be in need of justification, they will a chapter, one may rehearse mentally a nurober
also be incapable of i t. of arguments before committing anything to
I shall mean by action any human behaviour paper. An action may also be composed of a
based upon and intended to bring about a nurober of sub-actions. In many circumstances
projected goal or end. Included in this definition the same projected end may be brought about
are what might be called negative actions, in a nurober of different ways. Agent A may
which occur when an agent deliberately brings obtain a desired fortune by gambling on the
about, or attempts to bring about, a future stock exchange, by poisoning a rich relative
state of affairs by refraining from action. He who has noted him favourably in her will, by
takes the conscious and deliberate decision rohhing a bank, by honest labour... and some
not to act. Negative actions will not be ways will be more effecti ve than others. Some
specifically discussed, but for the purposes may clash with other projects that he also
of the present chapter remarks about actions wishes to bring about; or he may think that
will be taken to apply mutatis mutandis to they do. Such actions may be simple or
negative actions also. Care must be taken not complex. A may write a letter by taking a pen
to confuse actions with mere movements; and making appropriate marks on a sheet of
concepts such as aims, purposes, ends and paper, or he may type it; but he may start in
projects are needed to distinguish between one way and finish in another. In each case he
them. (About movements I shall have nothing will be performing the same action. I shall
to say.) The importance to actions of the say however that the sub-actions are different.
listed concepts is furthered by my contention In the example given neither sub-ac;tion is
that non-arbitrary evaluation can be introduced strictly necessary for the completion of the
only by reference to them. This is one reason action. He could have used a pencil, or
why the concept of action should be clarified. dictated the letter.
Actions may be overt or covert. Examples A completed action will be termed an act.
of overt actions would be cleaning a car, An act may or may not be the one intended by
running for a train, holing in one, etc; in the agent. B intends to run a four-minute mile.
general terms, in acting overtly I bring about He runs a mile, but in five minutes. An action
a change in the extemal world. Examples of is completed, but the act is not the one
covert actions would be mentally carrying out projected by the agent. Thus to act is not to
certain mathematical calculations, composing guarantee success, which is perhaps obvious.
'in one's head' a melody, solving mentally a An action may often, perhaps usually, be
problern in science, etc; again in general explicated in terms of a nurober of sub-actions.
terms, in acting covertly I do not bring about How can these two be distinguished? I think
in terms of ends. Actions can be characterized sub-actions, and have suggested the form that
by reference to projected or actual ends. To this distinction takes, but this does not mean
know what action an agent is performing one that the distinction can always be unambigu-
must know (i) what state of affairs he intends ously applied. in a particular case, the
to bring about, and (ii) what state of affairs he observer, and for that matter the agent,may have
is actually bringing about. When the act is difficulty in distinguishing one from the other.
successful (i) and (ii) coincide. When they do This is partly because actions and sub-actions
not, it can be said that the agent thought he are similarly structured, yet when the agent
was doing x but that really he was doing y. chooses how to act the distinction may be
lf the projected or actual ends are different important. A may wonder whether or not the
the action is different. To change one would be establishment of missile bases near his
to change the other. Thus if I change my territory constitutes an act in the sense
action, or do a different action, I necessarily previously defined. Is it, for example, the end
change its end, which is not the case with product of a series of sub-actions designed to
sub-actions. I am not, in the appropriate sense, bring it about (as if agent B has said 'Let's do
doing something different if I type a letter it and see what happens'); or is it a sub-
rather than write it in pen and ink, nor if I action in a !arger purposive plan? A's choice
start it in one way and finish it in another. of action will be influenced by the answer he
There are some cases where prima facie this gives to this question.
might be so (when, for example, a typed letter
is part of the end of the action), but even here The Larger Goals
there is a choice of appropriate sub-actions. Now it might seem prima facie that all
It makes no essential different whether I type acts may be regarded as sub-actions in relation
the letter or whether I dictate the letter for to the !arger goals that a person may set
solneone else to type. It may be, of course, himself. But this hypothesis is, I think, a
that an alteration in one or other of the sub- mistake; not because it assumes that people
actions will in fact lead to an alteration in entertain long-term ends of the requisite kind,
end, and hence to a different action. Thus in nor just because it assumes this (though if
one sense, an observer cannot know what a people did not entertain these ends it would
person is doing until he sees what is actually be). I shall indicate later the kind of mistake
accomplished. If agents were always successful it is. Here it need be said that the agent does,
in the projects they set themselves, an in practice, make the distinction outlined
observer would know what A is doing when above, though it should be added that, if
he knows what A intends. Often he only knows pressed, he will make it at a nurober o[
what A was doing. Choice, however, is different stages. There is a clear sense in
concemed with what a person intends, and is which I can, and perhaps will, say once this
characterized by reference to projected ends. chapter has been written 'Well, that's done! ';
Caesar chose to conquer Gaul. Success or just as there is a clear sense in which it would
failure makes no difference to his choice, have been inappropriate to have uttered the
though it makes a difference to him. Since phrase as I successfully picked up my pen to
this chapter is concemed with choosing I commence writing. But would it be equally
shall consider actions only in terms of their inappropriate to think of the writing of the
intended ends. In this context sub-actions can chapter as a sub-action intended to bring about
be thought of as the form or forms that a some further end - participation in a sym-
particular choice takes. posium, for example - and to think of this in
I have distinguished between actions and its turn as instrumental in fulfilling further
aims and purposes? Here the kind of mistake justification for choosing y will be given in
involved begins to become clear. The ogre of terms of x. But if one is to choose at all one
infinite regress appears. Not every action can must act. (Here choosing to do nothing can be
be clone for the sake of some further action. characterized as acting negati vely, as distinct
Acts, like Tennyson's time, must have a stop. from not choosing and doing nothing.)
ARISTOTLE noted this at the beginning of The action performed must be characterized
the Ethics, but I shall not go on to draw the in terms of its end, but this does not mean that
conclusions that he drew. I shall indicate the agent must be able to give a precise
later that acts may come to a stop in two description of either the end or of what he is
different ways, though perhaps only one of doing. If a child teils an observer that he is
these is typical. counting the number of bricks in a box, the
The infinite regress argument is trouble- o bserver knows what he is doing, and so does
some in another way. The only effecti ve way the child. It is not merely that the observer
to avoid it is to argue that sub-actions are can name what the child is doing; he can do
always capable of justification, but that it himself. However if asked what I am now
actions never are. Human acts may then be doing I may reply 'Writing a chapter', but
accorded justification to the extent that they having said what my action is I have still
can be reduced to, or legitimately characterized given no precise description of it. Further
as, sub-actions. Those which cannot be so questioning as to the kind of chapter and what
treated must remain unjustified, and indeed it is about may help to fill the gaps, but a
incapable of justification. Ultimately their precise description may be impossible. Of
performance is arbitrary, and it is because there course I have some idea of what I am doing,
is a resistance to the arbitrary that the argu- since only then can I construct the steps which
ment is troublesome. One expression of this I believe to lead to the purposed goal, but at
resistance can be seen in the promulgation of some points along the way the goal is the
the traditional so-called objective theories merest outline, a schema the details of which
of value. I may not know myself until the action is
A prejudice against the arbitrary is not completed. Thus actions and sub-actions may,
the only difficulty encountered in this field. even for the actor, have only relative clarity:
Another is that many projected aims and they can be seen as lying along a continuum
purposes are often essentially obscure. Same- from those cases where the projected actions
limes the obscurity can be removed but some- and sub-actions are the merest sketch to
times it persists. Its persistence does not, of those where both can be clearly delineated.
course, prevent a person choosing; nor does it An example of the former is painting a picture.
prevent him acting, since choosing entails Here the goal may be only the vaguest schema.
acting. To choose x is to attempt to bring it The artist cannot in advance say what precisely
about, and in this respect choosing must be his end is, nor can he formulate the precise
distinguished from mere wishing. A person sub-actions necessary to bring it about, but
may wish for x, but if he does nothing to bring his behaviour with brushes and paint is not
this about he can scarcely be said to have entirely random .. It is directed towards a goal,
chosen it. The prisoner may wish to be free, however loosely schematized it may be. At a
but he can only be said to have chosen to be certain stage the painter may put down his
free if he takes some steps to achieve it. brushes with the awareness of having completed
x may be chosen deri vati vely, in that y may be a task not previously describable in detail.
chosen in order to brin.g about x. This is the 'That's what I was after' he might say to
normal relation of sub-action to action: the himseif. Here his behaviour may be likened
to a voyage of discovery into relatively un- reasons one may mean the reasons that a
known territory. By the very nature of the person is able or prepared to give; but one may
venture the voyager cannot, in advance, say also mean the reasdns he had for doing what he
what kind of territory he will find or what did, whether he was able and prepared to gi ve
precise features it will possess; nor can he them or not. The concept of rationalization
say just what the character of his behaviour has a role to play here, for instance when one
will be. At the other extreme the actions and wonders what A's real reason was. People
sub-actions may be clearly defined. An example are thought to deceive themselves about their
of this might be making a pot of coffee, where motives for action, though 'deceived' in this
past performances help to lay bare their context needs more careful unwrapping than
relevant details. Between these extremes it customarily receives. However, I want to
are any number of possible actions and sub- concentrate upon the diffetence between
actions capable of various degrees of prior reasons and causes. This difference was
clarification, as are the relations between characterized by ScHUETZ (1951) as the
them. It is one thing to set oneself to win the difference between 'in-order-to' and 'because'
love of a desired woman. It is quite another to motives.
know precisely what to do.
Actions then may be described with Causal Accounts and Determining Factors
varying degrees of clarity. Sometimes the This is not the place for an extended
obscurity is of the kind mentioned above, but analysis of the concept of cause, but some
sometimes it is rather different. It is arguable points need to be made if only because they
that some choices are preconscious. Note that are usually ignored. Take for example decision-
I do not say subconscious, since there seems taking and creativity - two classes of human
to me an important difference between the two. behaviour clearly relevant to the design
The part of this difference which concerns the process. On occasion one wants to influence
present chapter will subsequently be made decisions, or to increase creativity. One can
clear. By preconscious choices I mean choices ask in a specific case for the reasons that
capable of discovery by certain techniques of support a particular decision against another,
introspection, though I do not suppose that and one can gain some influence in respect of
these techniques are always or even usually the decision to be taken by producing reasons
available to all persons. I do not think that thought to be relevant. It is plausible to
preconscious choices can be causally con- suppose that decisions may be influenced by
nected with conscious ones, and this is the less rational means: for this reason Public
important distinction between them and sub- Relations and Advertising find employment for
conscious choices, at least as 'subconscious' psychologists. It is not so clear that creativity
is normally used. The use of the concept can be increased by the presentation of reasons
preconscious choice is to gi ve meaning to intended to show the need for such an increase.
many of the actions we perform. This is often Here an understanding of the nature of the
overlooked owing to the inherent ambiguity of creati ve acti vity would clearly be of use. If
the question 'Why did you do x? '. This ambiguity its cause could be explained, current techniques
shows itself in the different uses to which the for increasing its incidence might well be
term 'motive' is conventionally put. When vastly improved; but just what kind of ex-
asking for a person's motive in doing what he planation would this be? Once again the very
did one may be asking 'What are your reasons complexity of the examples tends to obscure
for doing what you did?' or equally 'What the points at issue. Concern with explaining
caused you to do what you did? '. And by particular instances of human behaviour may
lead one to ignore pdor assumptions, although It seems not, since he may choose the project
these have their own importance, at least for despite the disvalue of the punishment. A
those interested in the theoretical status of decision to pursue the project would be
'explanation' in the human sciences. What, evidence for this. But A may choose to be a
after all, does a causal explanation imply? thief in a society where he believes punishment
The supposition behind the traditional to be avoidable, and honest in a society where
causal account of human behaviour is that he believes it tobe inevitable. If he is taught
people act as they do and choose what they and believes punishment to be unavoidable the
choose, because of certain determining factors, choice of actions open to him is not obviously
which are briefly describable in terms of limited. This circumstance will not force him
heredity and environment. Such descriptions to evaluate punishment in one way or another,
take the form of reference to physical, psycho- but if he chooses it as something to be avoided
logical, social, and historical conditions. at all costs then his choice of actual actions
It cannot, I think, be denied that these do will be limited by his beliefs conceming its
constitute determining factors. To what extent inevitability. Some restriction will have been
an agent's beliefs are so determined, and to imposed.
what extent his beliefs determine his choice Beliefs, of course, are not held to be the
to act in one way rather than another, is a only determining factors. Emotional pressures
matter for detailed debate. It is sometimes said are also high on the list of effective causal
that A believes what he wants to believe. This, agencies. These in their turn are often given a
if correct, would mean that in some cases at causal explanation in terms of human physio-
any rate it is not the beliefs that determine logy (e.g. the chemistry of the human organism).
what A chooses, so much as what A chooses Here, too, a more detailed analysis is required
determining what he believes. But that beliefs before the relevant issues can be properly
have some influence upon behaviour seems clarified, but for present purposes such an
scarcely deniable. For A to act at all he must analysis need not be provided. Few, I think,
see some different state of affairs as possible. would deny that the choices people make are
If he has been led to believe, and does believe, conditioned in certain ways: the argument is
in the impossibility of x he is unlikely to more likely to be over the extent of the
attempt x. The nature of the influence of conditioning. But some would argue that the
beliefs upon choice however depends upon causal account in its proper formulation pre-
whether it is a choice of actions or sub- supposes the possibility of an account in
actions. lf A believes that punishmen t may completely deterministic terms. Explanations
follow a course of action there are two ways fail out of ignorance, not because the causal
in which this belief is relevant to what he account cannot in principle be given. Against
chooses to do. It may affect the sub-actions this others would assert an in principle
that he performs in carrying out his project; impossibility in the demand for such an
they may, for example, be designed to escape account. This takes the form of demonstrating
detection. It will not necessarily make him that the account fails through circularity. The
abandon his project, though it may do this. argument can be put in a number of ways. For
How inevitable he believes the punishment to convenience I shall sketch the account given
be will be a relevant factor, but it will not be by HUSSERL (1965) which I shall call the
wholly decisi ve. Suppose A does believe that 'natural science account'. Husserl argued in
punishment cannot be avoided (e.g. punishment effect that in assessing the value of this
by an all-seeing, omnipotent god). Will this of account one should not forget that doing
necessity cause him to abandon his project? natural science is a conscious human activity.
Indeed the term 'natural science' names such latter had largely symbolic content. Any
acti vities on the part of natural scientists attempt to assimilate physical theories to
(psychologists, sociologists, biologists, etc.) physiological hypotheses was an error. This
working in co-operation with one another. is not the place to repeat the arguments relative
These activities are amongst the things that to this position, nor am I here much concerned
the natural scientists are called upon to with its validity. I am concerned to point out
explain. Is it not then circular to use natural that seeing how science relates to design.
science to explain the 'event' natural science entails holding views about the nature of the
itself? Psychological and other laws of nature scientific enterprise. It is these views that
are called upon to explain their own production. dictate the character of much talk in this field.
Produced by the contributions of scientists Given that Duhem's claim may be made
they are themselves part of the problem. plausible, as I think it can, a number of prima
facie possibilities present themselves: a
A similar but different argument can be
particular science may be wholly descripti ve;
put as follows. Supposing that it were true
it may be wholly non-descriptive; or it may be
that everythin g people do is causally deter-
poised between the two. If the latter, it may be
mined, then everything they say would be so
predominantly descriptive, or it may be pre-
determined. Amongst the things so determined
dominantly symbolic. It may be neither. I do
are the utterances of scientists. Amongst the not take the issue to be closed, but I do take
things uttered are particular causal accounts the elucidation of the relation of science to
of what people do. Thus the reasons such design to involve taking up a position in this
scientis ts gi ve for any account cannot be the debate. In practice this may engage attention
real 'reasons'. The real 'reasons' are the
only when things go awry, but I would not on
causal agencies in virtue of which they cannot
this account regard the issue as of any less
belp saying what they do say. So why then
importance. That science and design are in
should any specific account be taken seriously?
fact related seems hardly in doubt, although
The force of the circularity account the nature of the relation is. Ergonomics
depends upon taking the causal hypothesis to presents a case in point, since it applies
be fundamentally descriptive. If it is taken as physical, physiological, and psychological
primarily methodological the circulari ty, whilst findings. But what is the nature of this appli-
it does not disappear, becomes irrelevant to cation? How like physics is physiology? How
the scientific enterprise. But then one should like either of them is psychology? Opinions
be wary of the tendency of this enterprise to differ, and the issue of descriptive versus
encourage one to prejudge the issue. The point symbolic has still to be settled. There are
is not merely academic. The causal hypothesis divergences of opinion as to whether or not the
lies behind most thinking in the fields of laws of psychology are nomothetic (like those
science, technology, and design. Yet its of physics) or ideographic. If we are theoreti-
precise status remains obscure, as do the cally concerned with the role of ergonomics in
relations between the disciplines it appears to desi gn, the account that we gi ve of it will be
support. The latter obscurity is aggravated by influenced by a prior account, as will the one
its own ambiguities. 'Science' appears to name that we will accept. This prior account will
a unitary activity, but how unitary is it? depend upon issues of the kind raised here,
DUHEM (1914) for example, thought that the although what would make this prior account a
empirical hypotheses of a science like physio- correct one is not prima facie clear. Even if
logy were radically different from the theories it were, the causal issue would still be in
of physics. The former were descriptive; the dispute.
Justification A may punish B because such an action

In these circumstances i t remains very exemplifies some aspect of the kind of society
much an open question whether any account of that he wishes to promote. This is the form of
'choice' in terms of because motives can do the retribution theory of punishment, which is
the job that it is intended to do. Even if the simply the view that i t is 'right' to punish what
causal account could be made to work it would is believed to be wrongdoing. It is not (as is
not do so in the required way. It would provide sometimes mistakenly supposed) an argument
a justification in the sense of providing an for punishment. The two views are not of
explanation - what I have called justification necessity incompatible, although they are very
(2) - it would not provide one in the sense of different. The first claims the action to have
justification (1). To explain causally A 's instrumental value; the second that its value
embezzlement of his company's funds would comes from its exemplification of an accepted
provide justification (2) for his action but it institution for which no argument needs to be
would not provide justification (1). In assess- provided. In each case the goals involved may
ing in this sense his action, the reasons he be either conscious or preconscious, and in
gave for acting as he did would be relevant to each case acceptance of the goals is necessary
this assessment. They would not be relevant, before any value can be predicated of the
or at any rate not relevant in the same way, to actions performed. Without human aims and
assessing the validity of the explanatory purposes the term 'value' would be devoid
account. As Freud noted in respect of religious of meaning.
beliefs, to explain why people hold the beliefs In terms of justification (1) only sub-
they do is to say nothing about the truth or actions can be defended. Such a defence can
falsity of the beliefs themselves; though given always be attempted in terms of efficiency,
the one it may be difficult to take the other expediency, availabili ty, etc., in respect of the
seriously. But if the beliefs are false they will completion of projects. The difficulties in
not be shown to be so by explaining how they practice are well known. The variables in any
came to be held. Analogously, to explain why choice situation are likely to be near astro-
A chose x is to say nothing about whether or nomical. The agent is likely to be ignorant of
not he was justified (1) in so choosing. But many of them. The projects of others, which
what sort of justification is justification (1)? the agent may or may not be aware of, will
In my view only instrumental. ideally need account taken of them. Projects
Now it would be wrong to think that it is can only sometimes be clearly characterized,
always possible to use instrumental justifi- but often they cannot. Yet the ability to choose
cation. Not all acts (i.e. completed actions) appropriate sub-actions is not independent of
can be seen as contributing towards the this clarity. Also, given the choice of sub-
realization of a project. Some can, but others actions the agent would ideally wish to de-
merely exemplify these projects. Thus A may liberate concerning their respective merits.
punish B because of the ends to which this This would entail making accurate probability
action leads or is thought to lead. This is the estimates concerning the chances of a sub-
form of the deterrence theory of punishment. action successfully furthering the proj ect in
B is punished in order to prevent others from hand. It would also entail an awareness of the
doing as he has done, as well as to prevent B relation of the sub-action to other things that
from repeating the offence. What the agent the agent requires: i.e. seeing how his various
wants is a society which will be free of the aims and purposes are related to one another.
kind of act that B has performed. Alternatively If the only way A can obtain x is by theft he
THE ELEMENTS OF DEMGN 129
may decide to go without x. Instrumentally then x has value. It would not be self-contra-
such deliberation would appear always to dictory to assert that x furthers y and deny
have value. Yet the apparently less valuable that x has value. I argue, however, that if A
may sometimes be preferred. In an emergency chooses the ultimate project y then if x furthers
situation a decision which is random but quick y, x has value; since if the above account is
may be preferred to no decision at all. The accepted it would be self-contradictory to
considerations of detailed application are assert that A chooses y and x furthers y and
legion. However, none of this affects the deny that x has value. [Though we should
general point that reasons can be given for note that this would not in itself justify (1)
sub-actions, but that they can be given for A doing x.]
actions only within a framework where these The contention that this would not be
can be reduced to sub-actions. Where this is self-contradictory presupposes the above
not possible their acceptance by an agent is account to be false. It does nothing to show
arbitrary. that it is. Thus if A chooses y and x furthers
It may be thought that the above account y, x clearly has some value. Has it, that is,
is guilty of a form of the 'naturalistic fallacy'. in the only intelligible sense in which value
If i t were I would not regard this as fatal, can be possessed: instrumentally. The mistake
since I do not believe the 'naturalistic fallacy' would be in thinking that to say x has value
to be fallacious in the required sense. The means only that x furthers y; or that only what
arguments for it work only if there is already furthers y has value. This is clearly not so.
x has value when anyone chooses y, and when
dissatisfaction with the conclusion that value
anyone chooses any other project that x
terms can be identified with those expressing
furthers; but it makes no sense to talk of the
natural properties. If it be argued that the value of the ultimate project (in this case y)
phrase 'x has value' doesn't mean 'x will itself. Here the concept value has no appli-
f1,1rther y' this is, of course, so. 'If x has cation. That this is not always apparent is due
value then x furthers y' is not analytically tothelarge measure of conformity over accepted
true. It is not self-contradictory to assert that projects in a specific society. Considering
x has value and to deny that it furthers y. the efforts normally expended to ensure
Nor is it analytically true that if x furthers y conformi ty, this is not, I think, surprising.
Chapter 16
DESIGN AND DECISION
S. A. Gregory
lntroduction The last fifteen years have seen a considerable

This chapter is concerned with knowledge expansion of interest in the topic, stimulated
of the design method and the benefits which in part by the successes of operational research
may come from its practice. One possible (BROSS, 1953), and in part encouraged by the
advantage lies in the fact that method can be hope that, within economics and some of the
taught. More important is the strong beliefthat behavioural sciences, it might become possible
by a methodical or systematic approach, the to extend greatly the sway of rationality. Some
chances of producing better designs may be of the decision procedures may, in fact, be
improved. traced back at least two centuries to the
Design may be seen as a perpetual struggle discussions of Bernoulli on the effects of
between fertility of ideas and choice, with probability and utility to the gambler.
choice in the form of decision having the last Designers have always made decisions,
say. It may be characterized as a sequential but, in most cases, they were certain to have
decision process. Therefore, concern with been 'intemalized'. If design decisions are to
method in design must also entail concem be brought out into the open and be made
with method in decision. But in addition, systematic designers cannot blindly take over
because many decisions have considerable decision theory as developed at large and
consequences it is important that they should expect that it will necessarily fit design. What
be laid bare ('extemalized') so that people in- is needed is more information about the actual
side an organization or system know the grounds practice of decision in design coupled with
on which decision is reached. The knowledge Suggestions about the way in which the corpus
of decision-bases is the full knowledge of of existing decision theory may be used.
constraints. Some students of decision are pre- Decision may be seen as a substantial
pared to see it 'as a means of making an point of difference between the design method
organizational response to a situation' (GORE and the scientific method.
and SILANDER, 1959).
It is not intended to imply that by a
Decision in Design
systematic method it is possible to arrive at
In accordance with what is now generally
all decisions; far from it. What needs to be
recognized is the limit to which system may be accepted as the design process it is possible
pushed and the place where rational approval to see each stage in the route from perception
must be accorded to non-rational procedures. of need to the commissioning of the product
Further, there must be recognition of the as terminated by a decision, whether formal or
differences between decisions possible in informal. Within each of these major decisions
closed and open systems. may be many nested decisions. Relevant to this
The conscious study of decisions grew up is HALL'S diagram (1962) in which he proposes
in management studies about half a century ago. partitions of the field of decision-making
131
(Figure 16.1). Within the general approach of In its simplest form design is seen as
this diagram, an 'ends' decision requires the finding the quantity or the dimension, the shape
detailed establishment of objectives which will or the material, to fit a carefully specified
usually include statements about: profit, market, requirement. Much of the teaching of design
cost, quality, performance, competition, falls into this area, including at the same time
compatibility, a da p t ab i 1 i ty, longevity, considerable emphasis upon certain restricted
simplicity, safety, legal and/or ethical methods of modeHing and communication. In
constraints, and intangibles. this, decision turns upon determining whether
In sequence from perception of need onward there is a match between the design and the
the decisions may be expected to vary in requirements or that some limit is not exceeded.
All decisions to be made
~
Ends or means . .. ( 1)
Ends I "Means
Number of d1mensions .. . .. . (~ ·" One dimension
n dlm~SIOnS
Measurability of dimensions .. ql--~--All on nominal scales
Mixed measures
I
"All on ratio scales
Rote of probabilit1es ....... (4)" • Certainty. p= 1·0
I
RISk, O<p<1
Number of individuals . (5) One individual
n 1ndividuals
-I
Type of decision cnterion .(6)""' •
""""" Automatie
I '" .Mathematical criteria
Trial and error t Appeal to authority
Value synthes1s
Use ot eth1cal systems
Figure 16.1. Partition of decision-making (alter Hal1,1962)
character. These variations may be due to the Whenever a model is used a decision has to be
technical content of the design, to the novelty reached about its content of uncertainty.
of the subject matter, to the prevailing This kind of design is characteristic of a
circumstances of the environment, to the non-dynamic market within which engineering
personali ty of the designer responsible, or to operates upon a range of well-known products
the way in which the designer hirnself is usin g materials whose properlies are
controlled. These Variations in approach turn, established adequately. This kind of design is
in the last analysis, upon some function of not adequate for present-day circumstances and
information and its appraisal by individuals. the last half century has brought an increasing
Decision is a choice among alternatives interest in optimization procedures for well-
made by the person responsible for the developed i tems, and techniques of prototype
outcomes. It is important to distinguish this and model testing for items undergoing
from any prior activity such as evaluation or the development. Optimization procedures them-
calculation of probable effects. The decision is selves may be said to have reached a high
not made by the calculation or the test run but state of development, even if they are not
by the individual responsible. applied as much as they might be because of
the Iack of awareness of the power of these one and the beginning of another being marked-
economic tools and because of the difficulty of usually quite dramatically at a recognizable
putting some of them into effect. In spite of instant - by a decision which chooses a
this, the optimization process is the principal particular solution from among a number and
aid to decision in the design of highly- sets out the sub-problern to be tackled in the
developed products; with new products the test next stage ... A number of features characterize
procedure is probably the principal arbiter. The the "critical" decision. First, it is taken by
rise of the computer has intensified interest in someone much higher in the executive hierarchy
linear programming for optimization, in than the designer who explores the
abstract models for testing designs, and in the consequences of proposed solutions ... Second,
search for models adequate for treatment by the decision is treated as if it were
perturbation to provide information on irrevocable... Thirdly, the acti vities of the team
optimization. undergo a marked change.'
Marples states that decisions are made by
Studies of Design Decisions
comparing the various proposals against a set
It is important to Iook at practical cases of criteria, but that these criteria are not
of design in order to be certain that designers, obvious. 'The list cannot be laid down at the
or other people, make design choices; that the outset, because, until the proposals have been
different kinds of decision may be found, and examined it is not possible to say in what
that the possibility may be studied of linking respect they differ.' It is possible to draw up
practical decisions with what theory may lists of criteria in the abstract (these would
suggest. constitute check-lists) (see MATOUSEK, 1963).
In spite of the very great number of designs Marples refers to 'general engineering values'
which have been made there are very few which are independent of a particular design,
readily available published studies which throw and to engineering values which belong to the
light of consequence on the practice of decision design. Any decisions made must take
in design as it took place in fact. Two principal cognizance of natural properties, engineering
sturlies of this type are those of MARPLES values and prior decisions. Prior decisions
(1960) and GR EGORY (1964). In his paper include time, manpower and expense budgets.
Marples reports on two designs which he Each decision takes account of these and of
observed as a non-participating engineer. The predictions regarding the outcomes of sub-
designs are largely concerned with mechanical problems. Tests, calculations, models and
embodiment. The paper of Gregory is broader drawings are made to help predict the future
in scope and deals with marketing, process behaviour with greater certainty and to reveal
design, mechanical embodiment, and production; su b-problems.
he writes as a participant. In summarizing his paper Marples states
Marples sets out his findings in the two that it is likely that his model applies only to
cases in the form of a tree diagram. This problems requiring novel solutions and not to
diagram begins with the greatest Ievel of those where the form of solution is known but
abstraction and becomes successively more conflict over the choice of parameters exists.
branched and carrying more detail. In his words: In extending this point made by Marples
'At the bottom we envisage detailed bits of it is important to note that he reported on
hardware made from particular materials.' At organizations engaged in designing equipment
each Ievel a number of alternative proposals for their own use. The requirement was certain
would be generated and investigated. Marples and there was no great competitive sanction on
sees design proceeding in stages, 'the end of product price. The principal uncertainty
admitted was in connection with the outcome of absorbing or dispelling uncertainty. He records
sub-problems. The strategy to deal with this particularly the deviation of the observed
was to choose a solution giving the greatest decision processes from what might be
ground for manoeu vre. suggested by th.e engineering literature.
Marples is supported in part by a paper
from BOOKER (1962) which, although not
immediately concerned with the study of Modern Decision Theory in Relation to Design
decisions, throws light on the decisions reached The last five years have produced two
in his study. He reports on a design problern clear currents in thinking about decisions. The
which he was able to follow as an outside rise of interest in system design has led to
observer, giving at the same time the design analysis of the problems of feasibility,
setting and the comparable work clone in other optimization, and values. This has drawn upon
organizations. He reports on the generation of the accumulated studies already made in other
possible designs for a given purpose by the disciplines (e.g. HALL, 1962). At the same
intensive utilization of precedent from any time workers concerned with disciplines having
relevant field already existing (i.e. morpho- relationship with the behavioural sciences,
logically similar solutions). In the course have tended to review decision theory in the
of discussion he refers to decisions made upon light of their own problems of decision (GORE
general engineering values and to engineering and DYSON, 1964; COOPER, LEAVITT and
values specific to a design, although not SHELLY, 1964; SHELLY and BRYAN, 1964).
employing this terminology. Booker writes Much of this kind of thinking has tended to
against a background which is aware of · show the essential limitations of what has been
competition in the form of alternative sources termed 'closed' decision processes.
of supply of hardware, although, in the Feasibility determination is a desideratum
particular case, much of the uncertainty of the for all decisions in design. Where the design
market was missing. Against this background is highly developed there is usually li ttle
he is much more conscious of the need for formal questioning of feasibility and the stress
conservatism in design, a conservatism which is on optimization. On the other hand, with
is satisfied by the use of precedent (leading innovative design, or completely new design,
to more certainty in design decision and, later, feasibility is of prior importance. In many
by the same feature, affecting customer cases, particularly with new design,
decision - a precedent eases innovation). optimization is almost impossible. The
There is subsequent appeal to model tests of important thing is that the design should work.
all kinds, from mathematical, through analogue, It is not always possible to test out a full-
to more obvious forms. scale design, as in the case of a bridge, so
Gregory 's paper is concerned with the that model testing may assume importance. In
design of an innovative plant system under some fields the model tests are extremely
competitive conditions. He connects the design complex and may lead to the pilot-plant. But in
with market requirements and carries it through view of the costs involved every effort is
to the conditions of service after sale. He is normally taken to undertake only that amount of
concerned with feasible methods and also with work which will provide sufficient knowledge
optimization procedures. At the end of the paper for the reduction of uncertainty to the point
he analyses the major decisions taken and the where a decision may be made. Page's strategy
considerable extent of uncertainty involved, involves this, together with the deferment of
even in what appears to be Straightforward decision until all reasonable alternatives are
optimization. Attention is drawn to the need for examined.
The introduction of thought about values, Subjective Probabi I ity

which is particularly associated with system Where statistical data exist or probabilities
design (Hall), arises from the substantial social may be calculated, it is possible to apply them
consequences of major system development to analyses of design provided that the
(LICHFIE LD, 1964) and with the extent of proposed product is to be manufactured in
man-machine interaction within the system. reasonable population quantities. Where a
Most of the thought is concerned with a search single item is concemed the estimation is less
for possible guidance from pr.ior thinking about obvious. The states of nature occur over a
values, particularly by philosophers. The entry range of values. One method of dealing with
of values into decision-making in design is not this kind of situation is to consider the range
by any means a conscious matter and the extent of values and make allowance for the posslbility
of uncons<;ious influence of values over design, variety of outcomes. This is a sensitivity
both directly and indirectly, on the part of the analysis.
designer and of the environment, is worthy of What is the situation with subjective
more study and has been touched upon else- probabilities? By this is meant an individual's
where in this book. Such unconscious influences estimate of likelihood in the absence of
may weil suggest that a system is closed when, objective data. How valuable is this information?
in fact, it is open, and decisions have to be What can be clone with it? There is a great deal
arbitrary. of current interest in this problem.
The application of formal decision It has been suggested that subjective
procedure recommended for design largely turns probabilities may be substituted for objective
upon the use of statistical decision (STARR, probabilities in payout matrix calculations;
1963). In this a payout matrix is constructed in that a group of people may be asked to make
which values are put relating possible designs subjective estimates which are then averaged
or alternatives to probable states of nature. in some way or which are used to provide the
After the application of any suitable sim- range of variation for a sensi ti vi ty analysis.
plifying procedure the best strategy is read off. This kind of approach has obvious
This kind of procedure is employed as a weaknesses. Clearly some people are better at
guide to decision-making under risk (DMUR). A making subjective estimates than others; indeed
risk situation is one in which there is sufficient they may be using cues which amount to some-
information about past behaviour to provide thing more than an absence of objective data.
statistical indication of the occurrence of In some cases creative people have the ability
particular states of nature. There is information to offer estimates which form the basis of seif-
which may be termed objective probability. ehallenge to fulfil a task (GREGORY, 1963,
Two other kinds of decision-making are 1965). How does the decision-maker grasp
distinguished: decision-making under certainty another's subjective estimate? How does the
(DMUC), and decision-making under uncertainty decision-maker choose the people for his
(DMUU). Authorities vary about the relationship assessment panel? How does he weight their
between the three kinds of decision-making. values? It is best in such a situation to be able
The simplest suggestion is that DMUC is an to rely on an individual's demonstrated ability
extreme form of DMUR, in which the probability and to make sure of his involvement.
of a state of nature approaches unity. At
another extreme is DMUU in which there is no Certainty and Paying for Information
information about the probability. To use the It is a general rule of organizations to
payout matrix certain assumptions have to be attempt to eliminate uncertainty from their
made. transactions in some way. The ideal situation
136 THE DESWN METHOD
is that in which decisions are made under (6) Precedent outsider full-scale design
certainty, and it is this situation which reported in literature.
managers attempt to create, particularly within (7) Precedent outsider full-scale design
the bounds of their own specialist responsi- from private reports.
bilities, by pushing uncertainty outside the (8) Outside laboratory information from
borders. It is this situation which many tech- prototype.
nologists see as the natural state of affairs (9) Deduction from first principles.
(Figure 16.2). (10) Estimate of possibility by expert.
What constitutes certainty is largely a The status will vary from company to company
matter of practice and attitude. What is and will depend upon the state of the art. It will
reasonable certainty in one design office is also depend upon the stage in the design
not reasonable in another design office. Ofthis process and upon the magnitude of the operation
Input Output as a whole.
boundary boundary The datum is office practice in the
Input r - - - - - - - - - - - , Certaon company. This will differ from department to
~ ootput
department, depending very much upon group
R~ec!-- L-r-------~_r~~~.
uncertaonty experience and leadership and the class of work
Pa•ss
t
Pass done. Given the datum then any design which
uncert aont y uncertaonty
to research to test house
departs from this requires to be operated upon
in order to make it acceptable. The greater the
Figure 16.2. Elimination of uncertainty departure from the datum the more the work
required to be done. The work might consist of
I am very conscious as a result of practical intensive technical discussion accompanied by
experience. I believe that the difference in nagging and banter; it might involve sending
attitude may be attributed in part to the someone to visit a distant works; it might mean
organization as a whole (a top management permitting a conservative dissenter to spend a
responsibility) and in part to key individuals substantial period of time exploring some
in the design field. alternative suggestions; or it might mean
From an analysis of design behaviour, laboratory or works tests. Practice showed that
not previously reported but noted in connection unless something of this kind was executed
with the studies of decision cited earlier tension would develop leading to results such
(GREGORY, 1964a), I venture to suggest that as 'working to rule', going sick, having a direct
it is possible to classify the relative status of argument, or causing a superior to summon an
information in terms of its content of uncertainty investigation. Under certain circumstances the
relative to a given organization as follows. situation might be such that a proposal would
not be accepted, leading thereby to a loss of
Certainty Status of Information opportuni ty. What was most remarkable was the
(1) Successful office precedent design observed need to repeat work already presented
reported in detail. in report form from another organization where
(2) Licensor successful precedent design the work has been carried through under the
reported in detail. supervision of the person proffering the
(3) Own laboratory information from proto- suggestion. His information was treated, in
type. fact, as if it had been subjective.
(4) Own laboratory information from pilot Within such a perspective it is easy to see
plant. that test procedures themselves can have
(5) Well-established textbook information. different degrees of uncertainty. The same
THE ELEMENTS OF DE~GN 137
holds true for the less expensive and less need further consideration. Even with technical
information-rich model situations. Working matters there are other problems; for example,
models are less expensive to set up and test it is difficult to determine adequately whether
than prototypes and they may be limited in a given technique or production will persist.
information. Calculations take less time and
Seller's market
money than working models. Each of these
procedures involves some kind of model but may
only be used for the basis of decision provided
that the level of scrutiny and the quantity of
information associated with it are adequate for Buyers market
the purpose, i.e. to reduce the uncertainty of Input CNerall organization
the situation.
These observations suggest that it might General flow of uncerta1nty
be possi ble to characterize the atti tude of
design offices by their reactions to different Figure .16.3. Uncertainty flow and
types of information material. Examples might dependence on market conditions
consist of reports of experimental work carried For reasons such as this economists have
out in their own laboratories (note possible developed methods of discounting the future.
differences of value between researchers); The discounting effect must be greater in fields
reports of other people's experimental work where development or change are likely to be
published in the literature; word-of-mouth rapid, as opposed to the si tuation in a stable
reports given by a competitor's employee; state.
reports paid for either from a consultant or from Certainty is impossible in the production
a licensing firm. Another series of reactions of a radically new design, particularly where
might be sought dealing with the amount of many alternatives offer themselves. It is not
reliance placed upon tests on a wide range of possible technically to explore all the
models, from prototypes to the most abstract, alternatives in depth. Some strategy must be
complete with simplifying assumptions. introduced to cut down the number of
alternatives considered. The introduction of
Where Certainty is Impossible this strategy has to be done in such a way as
Certainty, i. e. substantial certainty, is not to carry conviction in any decision.
possible in a very large number of practical Every creative design which deserves the
cases. When market demand is high and offtake name introduces problems of decision. Every
is relatively certain then difficulties are likely fresh attempt to produce a creative design in
to arise with raw materials and production an existing field means the questioning of
methods. (Such a situation exists in the rational decision.
Midlands at this moment.) When demand is low
competition will be particularly active (Figure Decision and the Individual
16.3). These market situations hold even when Some remarks have been made earlier on
the product itself is technically well subjective probabilities and the effect of
established. experience, unconscious cues and involvement.
Estimating the market's pattern for some It has also been suggested that practice
years into the future has many difficulties. In shows a difference between individual and
spite of good trend data, any one year may be individual and also between organizations in
much out of line. Where relevant the respect of the amount of uncertainty which may
complication of fashion and subjective values be accepted in the making of a decision.
This difference between individuals may assigns to the maximum payoff in respect of
be discussed in terms of utilities and also in its chances against the minimum payoff. This,
terms of general personality make-up. Utility of course, is back to the field of subjective
is concerned with the value of outcomes to an estimates.
individual and is greatly influenced by that Another approach which attempts to provide
individual's existing state. Thus the gain or some method of valuing the possible outcomes
loss of !10,000 is likely to have much greater is that offered by Savage. This is the approach
significance to the normal designer than to a of the bad loser. It is based upon the attempt
millionaire. It is possible, with some careful to minimize the maximum regret feit should the
work, to establish utility curves for individuals. outcome prove unsatisfactory.
This has been done by GRA YSON (1960) in
sturlies of oil-drilling decisions. It is possible, Decision and the Organization
provided one has such values, to use them in Some people advance design rules which
a sensitivity analysis. are, in effect, decision rules. Of these the most
In addition to the individual's employment obvious concerns the use of precedents. This
of a given level of uncertainty as a jumping-off is implicit in Booker's paper. EDER and
point for his decision, and his appreciation of GosLIN G (1965) make it explici tly. This is a
the consequences to him in terms of reward or 'play-safe' rule which, although creditable
punishment (] AQUES, 1956), it is possible to under many circumstances, in other situations
adopt different decision rules which, in their· cannot be satisfactory. For example, in
turn, may be related to the characteristics of attempting to make radical changes to prepare
the individual, although formally they may be for decades ahead it would be useless to
the expressed policy of an organization. operate so. Decision rules must fit the occasion.
The range of these rules is now weil
Eder and Gosling also list some criteria of
descri bed in the li terature ( e.g. Starr). The
decision for the aircraft, automobile and marine
earliest is the Bayes-Laplace criterion. This
engineering industries. These criteria they
states that, since one has no information about
characterize, according to industry, as the
the states of nature and the situation is quite
relevant design philosophies. This would
uncertain, it is reasonable to assume that each
suggest that differences in decision criteria
state of nature has the same probabili ty. In this
might be found according to the industry,
Nature is taken as neutral in its dealings with
according to the company in the industry, and
the individual.
according to the design group within a given
The minimax criterion of Wald assumes that
company.
Nature acts as an enemy. Under these
circumstances the rule is to choose that Since most present-day design takes place
outcome which provides the minimum loss. The within organizations it is useful to have some
maximin criterion is also pessimistic but rather idea of the way in which an organization may
more positive: it relies upon the choice of the influence design. The most obvious effect of
outcome which gives the largest minimum the organization is the part played by the
payoff. structure. More points at which decisions have
To accommodate the optimist, the man who to be taken or questioned mean more time taken.
believes that Nature is on his side, the rule is More points of question mean, generally, more
to select that course of action which provides reasons why not. The !arger the organization
the maximum of the maxima. The Hurwicz the less the concentration of adequate
approach is to introduce a coefficient of information, unless great care is taken. It is
optimism which permits the decision-maker to probably to combat difficulties of this kind that
make some rating, between 0 and 1, which he people tend to adopt a 'project' or 'organic'
type of grouping for the introduction of new system. The designers may co-operate with one
schemes. Only when the bases of decision are another or be in competition; their behaviour
clear does it become possible to use the more depends upon the management policy. Apart
common 'matrix' or functional structure. from the interdesigner attitude, which will tend
Structure is not all, since attitude determines to influence decision-making, there will be the
whether structure will work adequately.Structure inevitable effect of the management attitude.
can be seen as a facilitating device. This will be influenced by the manager's state
Decision-making is a manage m e n t of optimism or pessimism and the way in which
operation. Big decisions need close contact this is reinforced or discouraged by overall
with the relevant level of management. company attitude and what the market suggests.
Innovation and, particularly, creati ve design Together with this must be taken the way in
tend to bring about decisions in their own which the manager chooses his approach to
right, which may lead to tensions between the decision. Is his approach relevant to the
designer and management. situation in hand?
In an attempt to bring together some of the
factors which have been considered in the Conclusion
development of attitudes to decision within a What this chapter has endeavoured to
design organization Table 16.1 is offered. This bring out is the complexity of decision. Some
table allows for designers to be of any of the people, for instance HITCH (1957), see all
possible types: of the first kind - rational, techniques such as operational research, as
rational/intuitive, or intuitive; of the second the art of su!roptimizing, i.e. of breaking
kind - pessimistic, steady, or optimistic; of problems down into pieces which may be made
the third kind - experienced, experienced with the subject of some kind of rational decision.
knowledge, having knowledge but inexperienced. Hall remarks: 'There is no comprehensive
For the purpose of the table the third cate- procedure for ... the 11 grand decision".' There
gory is not indicated although it may influence is, in fact, no royal road to design. In this
atti tude. These designers work within some chapter I have presented grounds for asserting
kind of organization having a management that designers need to look at the act of
Table 16.1, Some Passihle Patterns Relating Market, Management, Decisions and Design
(the effects of competition and collaboration are omitted)
Market situation Mass/seller's Mass/buyer's 1-off/seller's 1-off/buyer's
----~-~--- -~
r-- ---
Management Trade-off ßargain Judge lmpose or fail

approach to decide
Decision type Computa tional Risky Uncertain Subjective
N eeded designer Pessimistic Steady Steady Optimistic

attitude
Associated Rational Rational/Intuitive Intuitive

design approach
decision in design in much more detail and to individuals for the satisfaction of human
study more carefully the way in which indi- individuals: that decision is, perhaps, the most
viduals behave in relevant surroundings, having critical step in this activity and, with creativity,
regard to the type of design being undertaken; places the greatest challenge upon human
that with careful selection they may improve performance. We live in a world which, for us,
design performance; that design organizations is neither timeless nor limitless in resources.
might derive advantage from a review of their SHACKLE (1961) sees decision as the ultimate
concepts of certainty and their expendi ture to
effort of imagination. For the designer decision
gain certainty.
is an imaginative act which balances out the
People must come to terms with the idea intangibles and uncertainties, and assumes
that design is an activity carried out by human responsibility for the outcome.
PART IV
DESIGN TECHNIQUES
Chapter 17
MODELS IN PRACTICAL DESIGN

S. A. Gregory
lntroduction a general view of the subject of models is

Without a very full use of models of worth while.
various kinds design is hardly possible, except
for the Operation of chance. Models
In copying an object and in remembering The discussion of models and what is
one uses a model; in transmitting information meant by them is rendered difficult at the outset
regarding a product one uses a model; in by the absence of any comprehensive theory of
managing the whole operation one uses one or models, although there are several well-
more models. In checking a design, one uses developed regions of thought and discussion
all possible methods of simulation, as noted by concerning models of which account must be
BOOKER (1964). taken in any approach. Included are those
The deployment of models is at the heart branches of engineering where people are
of design. Much of recent effort in the practical concemed about conditions of similarity
development of design has been devoted to between some small or partial system and a
specifying the details of this deployment in !arge system (this is considered again later);
order that computers may be programmed to physics and the philosophy of science as
fulfil some part or the whole of a design discussed by HESSE (1961, 1963); applied
operation. Thereby the computer becomes a mathematics and the associated branches of
model of the designer. But, in order that this problem-solving such as operational research;
might be achieved, a very substantial effort is pure mathematics where T AR SKI (1954) sees
required in the collection of models for the model-making as a new branch of meta-
details of each programme. mathematics. The stage has been reached where
No design effort is likely to have practical no academic subject with pretensions to
success unless, in his work, the designer has connection with science can avoid some
a satisfactory model in his mind of the discussion of models and their presentation,
behaviour of nature, and unless he has a satis- usually in symbolic form. In biology, which
factory mental model of the market. deals with systems somewhat more complex
Within a given technology the essence of than chemical systems, it is usually
skill and experience lies in mastering the necessary to consider önly analogue or
relevant practical and theoretical models of the mathematical models, in view of the difficulty
subject and the strategies by which they are of obtaining similarity in any close material
interrelated in their deployment. Each sense. The limitations and complexities are
technology has its own set of such knowledge discussed by CANGUILHEM (1963) and the
and techniques. But among the technologies range of possibilities illustrated in the
there is such a wide resemblance, already compilation of BEAMENT (1960). When
noted in respect of certain types of model, that behavioural situations are considered, as in
143
Table 17.1. Some Words Containing Aspects of 'Model' (Roget may have more)
abridgrnent eidolon illustration pattem sketch

abstract eidos image picture sort
algorithm epitome Imitation pllot species
altemative essence plan specification
analogue example kin principle stamp
archetype exemplar kind programme stock
arrangement prototype structure
family mannequin style
basis fashion manner quallty substitute
bulld form map supposition
formula metaphor reference surrogate
cast frame microcosm referent symbol
category mirror reflection system
character gauge mode replacement
dass genus mould representation template
code group resemblance theory
computer guide nature role type
concept rule
conception hypothesis opinion variety
conjecture original schema version
copy icon outline scheme view
idea set vision
description ideal paradigrn shape
design idol parallel similltude way
draft ilk parody simulacrum
Table 17.2. Uncategorized Functions of Models (relate to Table 17.1 aspects)
adjust encode increase perceive shape

amuse enumerate indicate persuade simulate
evaluate in form plan sort
calculate experiment Interpret please specify
check explain investigate predict state
classify explore produce store
command judge suggest
communicate find
compare forecast manage teach
compose measure reduce test
control guide refer train
operate relate transform
depict help optimize replace translate
describe order reproduce
design illustrate resolve
determine imitate pattem reveal visualize
DESIGN TECHNIQUES 145
Table 17.3. Suggested Glassilication
Communication (X to Y) Problem-solving and decision (operation by X)

amuse order adjust find predict
forecast produce
depict persuade calculate
describe check increase
reproduce classify investigate reduce
enumerate compare relate
explain specify compose judge replace
state control resolve
guide store manage
suggest measure shape
illustrate determine sort
imitate teach operate
indicate encode optimize test
in form experiment transform
interpret visualize explore plan translate
psychology and sociology, the need begins to pictures become linked with others to form
arise for hypothetical constructs such as the considerable structures. According to a
concept of 'role'. Discussion about the range person's inbuilt facilities and his training and
of models employed in scientific acti vi ty has experience the mental process is able to deploy
been collected by FREUDENTHAL (1963). a range of preconscious and conscious problem-
The range of functions of models is not solving methods. The individual then carries
limited by the list gi ven and an attempt to out some action directly or communicates with
sketch out a fuller picture is provided by others.
Table 17.1. The information is then analysed Communication with others requires some
in Tables 17.2 and 17.3. The result of the extemal model. External models, just as
analysis is to suggest that models are used internal models, are representational. They
generally in problem-so! ving and communication. either represent to some degree things or
The future development of model theory is events which exist or are believed to exist or
likely, therefore, to involve information theory. have taken place, or they represent things or
events under consideration for the future.
External models may be made of any kind of
'Interna!' and 'Extemal' Models information-bearing or information-yielding
Models may be divided into the classes of 'substance', for example three-dimensional
'internal' and 'extemal'. Interna! models are solid objects, graphic models such as plans,
patterns in the head, either preconscious or word pictures, computer programmes and
conscious. These are presumably provided with symbolic statements. Provided they deliver the
raw material through the modalities of necessary amount of relational and other
perception as described by BARTLEY (1958) information they may work through any of the
and developed upon neural nets by some modalities of human perception. They may be
process which is still unclear, but is now more abstract or more detailed than the original
believed to have a chemical basis. By some or the final copy or representation. They may
mechanisms, possibly like those hinted at by be distorted. Provided some specified trans-
Newman in Chapter 13, small and insignificant formation procedure exists by which the
necessary information may be extracted or computerized optimization procedure, might be

applied they are satisfactory. adopted.
Models may convey information according For emoti vely-charged situations, the
to a spectrum of Situations (rather like the models tend to be made of, or found in, the
situations in which artefacts are used, as same substance as the product is to be made.
described in Chapter 5), ranging from the Sketches serve as models for painting; clay
technical to the human, or from the 'neutral' models precede sculpture. For writing, trial
to the psychophysical. For strictly technical passages may be dictated or drafts made, or
reasons concerned with the inborn abilities of the ideas of other authors may be borrowed.
an individual, visual or concrete modes of The design problems of such media tend to be
communication may have to be chosen rather solved in those media and in no other way,
than semantic or symbolic. On the other hand, except it be in the head. They communicate in
in order to provide moti vation, conviction, or their own right.
deep-seated response, some form within a Technological design problems may be
given mode or group of modes may have to be solved in a variety of media and communicated
used. in others. Therein lies much of their difficulty
More generally, it is preferable to choose and complexity. To study a technology to
from among the models available, according to effect requires an adequate grasp of the facts
the task which is to be attempted. Where a of the subject and a working facility with the
'neutral' situation exists with not much problem-solving tools and the communication
information, it is desirable to use something procedures.
which closely resembles the 'real thing'. Thus,
with a complex metallurgical process,
experiments would be carried out on the actual Models in Technology
plant if it existed; if it did not exist an attempt Three principal classes of models appear
to build a pilot plant should be made. Where it in technology: material models for problem-
is desired to carry through many transformations so! ving; abstract models for problem-so! ving;
in an attempt to sol ve a problem, mathematical communication models.
symbolism might be used in view of the
richness of mathematical procedures. Where Material Models
shapes or positions are to be defined to human These range from the actual plant to
beings, some optical, or optical plus tactile, various smaller, distorted, or simplified
mode is found to be suitable. In communicating versions. Treatments of the theory have been
such information to a machine, punched tape is provided by LANGHAAR (1951), jOHNSTONE
offered; in carrying 'charged' information to and THRING (1957), and PANKHURST (1964).
human beings, visual or sound channels are Analogue computers represent a stage some-
used and several layers of information, both where between material models and abstract
immediate and emotionally symbolic are models. ROGERS and CONNELLY (1960)
employed. In such selection, the effecti veness review aspects of the use of analogue
of the choice must be considered in terms of computers in engineering design. Clearly,
resources available, whether technical, money, analogue computers can only operate in certain
time, etc. Thus to solve a non-repetitive types of situation, i.e. those for which
geometrical problern as in plant layout, a analogues exist; their use eliminates much
strictly geometrical model method is likely to original information and may tend to introduce
be used as the first choice. With repetitive irrelevant information. On practical grounds
situations such as pipework layout, three- they tend to be restricted to situations of few
dimensional model kits, or more recently a stages. Material models give factual data.
Abstract Models made smaller certain features may have to be

These are normally symbolic and employ eliminated. What are the rules of elimination?
mathematical or propositional operations. They How can the information content be reduced?
may often be transferred to computers provided What is meant by information in this context?
that suitable procedures may be found. The It is also known that if the magnitude of the
whole range of applied mathematics may be model is reduced, some kind of distortion
used. Within the field of design the most of geometry, or time, for example, may need
general attention has been directed to to be introduced in order to make the model
optimization procedures. Introductory treatments operationally comparable with the original.
are given by ASIMOW (1962) and STARR (1963). Whether concrete models or any other
More extended treatments of economic substance is being dealt with, it must be
analysis for manual operation are referred to in appreciated that, in design, model situations
Chapter 18. Abstract models manipulate data are met which do not involve reduction in
but do not provide new factual data. magnitude. There are situations in which
full-scale models are used. More importantly,
Communication Models situations must be dealt with in which the
These have recently been treated by model - the product - contains more detail
ROSENSTEIN, RATHBONE and SCHNEERER and is more concrete than the original. The
(1964). Some communication models form part designer starts with a relatively unformed
of the problem-solving process: thus in and primitive conception and builds this up
mechanical engineering, drawings often serve to provide a specification for manufacture.
a dual function; GosLIN G (1962) refers to In arriving at this kind of result the designer
diagrams in electronic system design. The has to make transformations which introduce
history of engineering drawing by BOOKER detail not previously abstracted from his
(1963) summarizes the development of this art, original. What are the rules for this kind of
an art now likely to be on the decline. transformation?
The complex task of designing and Summarizing the discussion so far, it
communicating in a system of the class man- may be seen that there are three classes
computer-machine is still undergoing develop- of model relationship:
ment. According to present findings some
visual channel from computer to man seems the (1) The 1 to 1 relationship
most fruitful, with digital communication from (2) The n to 1 relationship
computer to machine. The man to computer (3) The 1 to n relationship
channel appears to be rate-limiting, although
this may well depend upon the kind of material fhe 1 to 1 relationship is at the back of most
being handled. thinking conceming models in the
representational sense: people tend to like
Transformations Relating Referent and Model their model to represent the original. Even
Unless the referent, which may be the when the system is transposed from some
original thing or the projected product or a concrete substance to mathematical symbols
mental scheme, resembles the model in all or to digital code, people still tend to think
respects (and this raises problems of definition in 1 to 1 terms. In many practical situations
of the method of comparison) some kind of they are forced to accept an n to 1 relationship
transformation must be accepted. since every item of the original cannot be
Commonly, model transformation is thought taken in or appreciated; nor, for many
of in terms of size: the model may have a purposes, is it worth while to reproduce every
smaller mass than the original. If the model is detail of the original. The human bodily
6
system, including perceptual channels, has the information content having some relation-
built-in devices to protect it from excessive ship to the transformation ratio.
stimuli which might be termed information. In the development of information theory
People are naturally predisposed to n to 1 Shannon (see BRILLOUIN, 1962) preferred to
operation. Under such circumstances it becomes use uncertainty as a measure of information
possible for different originals to become content: in a situation of much information
related to a single model. This is usually there is little uncertainty and vice versa.
fortunate and time-savin g. The difference in previous history between
Where there is a sequence of reputedly 1 indi viduals makes it extremely difficult to
to 1 transformations performed, errors may have much assurance about the correct transfer
enter into the overall transaction for one of information between them by any· model.
reason or another. This situation is significant Practical experience indicates that the more
in the communication of information and, realistic the model the greater the likelihood
tagether with other features, has led to some of satisfactory transfer. The material model
of the developments of information theory. One is likely to be more succ~ssful than the
conclusion that may be drawn from the theory abstract model. The material model is likely
is that, in order to give an adequate to reduce the number of degrees of freedom and
representation at the end of the series of hence the uncertainty. A model which contains
transformations, it is necessary to provide more information than the referent may, in fact,
redundancy of information at the input. The in troduce uncertain ty.
amount of redundancy necessary is related to With certain classes of system it is
the uncertainty likely to be experienced by possible to define rules of operation in such a
the receiver regarding the intention of the way that models in those classes may be
communication. For example, if one has a comprehended by people of different kinds and
workshop next door to the office and if one backgrounds. lt would seem that the essential
knows the foreman well and company rules do feature of any such operation is to exclude
not forbid it, one may say: 'Please make me a uncertainty by elimination of unnecessary
11 deral" 26 inches long from this stuff .' lf the
degrees of freedom. Much of the development
company system has to be used, drawings of the physical sciences since the time of
become necessary and a job number as well; Galileo has been concerned with either
but it may still be possible to rely very much simplification for the purpose of experiment or
on shop experience and practice. lf, however, the development of theory. At the extremes of
the work must be done by tender outside, then modern physics we are concerned with rules of
a considerable amount of written information information: Einstein has provided rules of
and drawin gs have to be produced, which is measurement to provide observers with the
particularly complicated in this case because same information; Heisenberg's principle
there is no British Standard Specification for implies that the amount of information in a
'derals '. system must exceed the quantity which it is
hoped to withdraw.
Information and Uncertainty In engineering work, when an n to 1
It must not be taken for granted that the transformation has to be carried out, a level of
essence of representation is that the model relevance of information must be established.
should contain the same amount of information This is by no means easy. It may be convenient
as the original: this only holds for the 1 to 1 to think in terms of some kind of system level,
relationship. For the n to 1 or 1 to n for example the Boulding degree, or the number
transformations changes should be expected in of physical dimensions. Usually one tries to
use the simplest system possible and then add variations do not upset overall production
extra information from the stock available. As beyond some acceptable limit. These changes
it were, one overshoots in the initial reduction. can be related to small material models of
In constructi ve situations, where 1 to n the system or to models set up on suitable
transformations are to be carried through, analogues (probably an analogue computer
'filler' information must be provided. This may today) or to models on a digital computer. In
come from standard procedures or may have to this kind of situation, only the details which
be prepared for the occasion. Within such appear to be practically significant need to be
constructi ve si tuations a number of different put in, gradually building up the models to
models are employed: account for more details.
Where operating plants or systems do not
(1) The scientist's model of the relevant yet exist and information for design is
parts of the universe; the scientist may be a required, a decision must be made on the best
physicist, chemist, biologist, sociologist, etc. means of providing the information which will
(2) Mathematical models and logical models diminish the possibili ty of financial loss to
by which the scientist's information, and any some acceptable figure in the given situation.
concepts which may have become associated, Since there is no operating system, some
can be manipulated. model system must be established which
(3) Experimental models which may be set adequately resembles the large-scale system
up to give extra information at minimum cost. in mind. This raises the problern of similarity
( 4) Pilot plants and prototypes. and its cost, which is discussed by KLINE
(1965). The analysis of similarity is aided by
For the engineer, the uncertainty reigning the use of dimensionless numbers and a
in each of the model situations leads to the substantial list of these has been provided by
need for some risk to be taken in practical BOUCHER and ALVES (1959, 1963).
work. The taking of risk requires decisions by To obtain similarity becomes increasingly
the person responsible. The lack of difficult as the levels of greater complexity in
correspondence between the model and the a hierarchy are reached, for example, in
original may therefore be seen as a large passing from static mechanical structures to
contributor to the development of the decision dynamic systems.
function in engineering. At bottom it derives Mechanical systems have been investigated
from the Iimitation of material resources and by physical models for a long time. Structures
from the human requirement to simplify in order ~nd stress analysis are readily treated with
to be able to deal with situations. such models. Today digital computer
programmes are available for many practical
Material Models and Design situations of this type. The inception of
Material models, i.e. models which arenot vibration .increases the difficulty but it is
abstract or symbolic and bear some close still possible to undertake useful work with
relationship to the referent in terms of cheap models. MORLAND, ATKIN and
substance, tend to be used either where they GANULES (1965) have given some practical
are very cheap, or where the mathematical examples, particularly relevant to an area just
models have not been adequately developed, out of reach of easy digital computer practice.
or where information has not been isolated. Flow systems have many complexities.
Where a plant or system is in being, the This is outstandingly so for situations which
effect of variations of conditions on the system involve turbulence, erosion, waves, or
can be determined, provided that these roughness. Physical models are actively used
to deal with such situations, for instance dams Mathematical Models

and barrages, waterways and harbours, and for For certain primitive and frequently
wind effects on structures of all kinds. This, occurring Situations it is possible to use
indeed, is the classical area of testing by abstract symbolic models. These are situations
models. in which the relationships between the
Heat transfer studies by models introduce parameters may be expressed by equations or
problems which may concern time where other symbolic expressions. Manipulation,
conduction is concemed, or incompatibility within the limits of mathematical possibility,
of mechanism where convection is concemed. may then take place.
For this reason there is a tendency to approach Frequently, in order to achieve a model,
heat transfer studies by way of analogue serious simplification has to becarried through.
devices. A fairly general idea of the use of mathematical
Chemical reactions often involve models is given by BELLMAN and BROCK
complicated systems, not readily treated by (1960). They say: 'Problems may be divided
models. Clearly, people dealing with the roughly into three types. These are the natural
design of chemical and allied process plants problems arising in the world about us; the
have had to devote much thought to their model problems, which we immediately
difficult task - made more difficult by the high recognize as an element of a dass of similar
rate of development of new processes which problems; and, finally, the symbolic problems
differ markedly among themselves. Good where an analytic formulation already exists.
discussions in this field are provided by: To solve a problern by mathematical techniques,
Johnstone and Thring, who look at the problems we transform it step-by-step from a natural
of similarity in general and in terms of problern to a model problern and from a model
combustion and c h e m i c a 1 engineering; problern to a symbolic problem. Obtaining the
contributors to the Institution of Chemical solution of the symbolic problern in analytic
Engineers publication edited by PIRIE (1957); terms, we must interpret it suitably to obtain
contributors to the Iron and Steel; Institute a solution of the model problern and then go
symposium (1965) which dealt with problems from here to the natural problem. None of these
of combustion, metallurgical and chemical steps is trivial and each requires experience
engineering. The aim of any pilot plant needs and skill.'
to be clearly established. It is a step in an It is important to note the part played by
overall design operation. judgment in the handling of the requisite
Biological systems might, in principle, models. In principle, provided the necessary
be more complicated than chemical systems, differential equations can be set down and the
but in practice this is not so because of the boundary conditions su'pplied, any problern
relatively narrow range of life conditions. can be sol ved to any level of accuracy
However, with human systems there come the required. In practice there are various
psychological effects discussed in Chapter 5. restrictions: unsol vability, inadequacy of
PAGE (1964) considers the investigation of information, and inability to compute either for
human environments by models. The technical or economic considerations.
contribution of Sheffield to model studies An undue enthusiasm for mathematical
cannot go by without acknowledgment. Electri- models and a disregard for data-giving
cal systems do not lend themselves to material situations is seen as one of the current faults
model investigations although electrical analo- of academic technology. FIRTH (1965) says:
gies are frequently used for non-electrical 'At the present time .... the designer sees an
systems. ever-growing mass of allegedly scientific
models, usually described in mathematical of use includes the following five steps:
terms, purporting. to represent the designer's (a) initial broad planning; (b) derivation of
reality, but in fact embodying no confirmation descripti ve equations; ( c) choice of model
of such either in a rig or in an actual machine. form; ( d) determination of model parametric
Designers are very familiar with papers values; (e) evaluation of model validity. In
carrying titles which lead the designer to their comments upon the deri vation of
suppose that what is contained therein has a descriptive equations they allow for (i) the
direct bearing on his reality, only to find that classical or conventional equation system,
the title bears little relation to the contents .... (ii) the network theory approach which they
A paper, for instance, on the design of hydro- see as particularly useful for handling complex
stalle bearings which contains a mathematical problems in a systematic manner, and (iii) the
model of what purports to happen on a slideway assumed mode approach.
or in a bearing, containing no indication that
experiments have confirmed that the model Optimization Models
accords sufficiently weil with reality for the A great deal of design is concerned with
designer to use with impunity. Again, a paper optimization. For this reason the new texts on
on the characteristics of a variable speed design place considerable emphasis on the
epicyclic gearbox might turn out to be a series range of normal mathematical methods. In some
of equations which have solutions only in the cases it is possible to arrive at the optimum by
sense that the real characteristics of both percei vin g a discontinuity. The normal approach
gears and hydrostaUe transmissions are is through differential equations to obtain a
ignored.' maximum or a minimum. If need be Lagrangian
This tendency to overplay mathematics multipliers are used. An alternative approach
may come from a widespread belief in the may be by network theory employing fixed
'teductionist' approach to technology and the values, inequalities, or probabilities. Both
sciences. According to this view any Asimow and Starr provide elementary
technology is but the application of some approaches.
science. Therefore it is only necessary to A great deal of modern effort in the
look at the science to which the technology development of optimization models has been
may be 'reduced'. But among the sciences concemed with the attack upon adaptive control
biology may be reduced to chemistry; chemistry systems centred upon digital computers. A
may be reduced to physics; and physics may be typical source of current information is
reduced to mathematics. Hence, and without provided by the 1965 American Institute of
any qualms, mathematics is at the basis of Chemical Engineers - Institution of Chemical
everything. An associated view may be traced Engineers symposium.
back to the Pythagorean Brotherhood, and to
Plato who has been principally responsible The Relationship between Classes of Models
for the anti-empirical approach imposed in Tagether with the two classes so far
uni versi ties. mentioned (the fact-giving material models and
For mechanical engineers an introductory the information-manipulating mathematical
note upon the use of mathematical models in models) practical design uses other classes of
design is provided by TRA YSER and models. These are (i) analogue models,
CRESWICK (1962) and MITCHELL and (ii) digital computer models, and (iii) general
CRESWICK (1963). The latter contribution problem-solving models.
describes the two major types of models as A simple way of showing some of the
either simulating or tracking. Their methodology aspects of the relationships is provided by
Figure 17.1. In this figure, which shows a this point that the general problem-solving
pyramid in plan view, material models are models need to be recognized. The greatest
set at the side opposite to symbolic models. range of analogues is likely to be provided by
general systems theory. The best developed
Syrrbolic techniques of problem-solving are likely to be
those of mathematics. DESCARTES, in his
first known publication Rules for the Direction
of the Mind provided some hints which are
still valuable. A more recent approach comes
from PoLY A (1945, 1954).
Computers and Design

Analogue computers, particularly those
based on modular electronic systems, have a
facili ty for dealing with relationships of a
continuous character and show considerable
Material advantages in the case of non-linear
behaviour. They provide good visual display
Figure 17.1. and easy modification of parameter values.
Plan view of modelrelationship pyramid Most practical development is however devoted
to digital computers am:l the models which may
Analogue models lie between these two sides; be built upon them. Digital computers have
so do digital models. At the apex of the great in trinsic flexibili ty and possess the
pyramid are the simplest systems, such as advantage of substantial memory, the magnitude
static structures. These may be investigated of which depends largely upon economics.
geometrically (the simplest kind of analogue) Digital computers have intrinsic rules of
on paper, or by the equivalent symbolic operation which demand the presentation of
systems. Given a little more complexity, it data in discrete units. Generally, differential
pays to go to experimental stress analysis on equation models require special handling
a suitable model or to make network whereas network models based on discrete
calculations which are easily transferred to points lend themselves to digital computer
digital computation should the opportunity Operation.
offer. The most straighttorward application in
At the base of the pyramid are the complex design is for items such as rigid structures of
systems of high Boulding degree. Here the an elementary character for which there is a
distance between model systems may be frequent demand for calculation. Many
greater. programmes tend to be readily available in this
The Watts model of the autonomic designer, dass of design problem. The interchange of
as in Chapter 11, may be conceived as passing others will help to develop this region and
along the axis of the pyramid and tending to its neighbours.
use the material nearest to hand. This will The more complicated the system and the
always be simpler than the model represented higher the Boulding degree the greater the
by the surface of the pyramid. difficulty in computer application; but it is
In order to deploy his efforts to the with systems ofthis character that considerable
greatest effect, the iconic designer will need advantage is to be expected. MORTLOCK
to carry the most general strategies. It is at (1964) has described how it becomes possible
to investigate the behaviour of large electrical computer optimization it is possible to have

networks by digital model. In process the computer issue direct signals to mechanical
industries, such as oil refining and chemical or electrical controls on the plant. In such a
manufacture, the application of the digital situation the computer is actually making
computer makes it possible to model reactors modifications in the process. This is
and systems of reactors. YOULE (1963) has tantamount to a marginal redesign, which is
described the hierarchical nature of such in the process itself, but not in the hardware
models. He says: 'Individual molecules collide which encloses the process.
and react in the chemical reactor which was
The greater participation of computers in
part of the chemical reactor described above,
design depends upon the considerable improve-
and that plant itself is part of a production
ment of the means of communication between
complex which is part of the acti vities of a
designer and computer, and much greater
large chemical concern, whose proliferating
immediate flexibility of the Computer
ramifications form part of a possible model
accompanied by suitable visual displays. These
of the whole national economy.
needs have been appreciated for some time and
'Out of this serial succession of possible the most attractive development in this area
models, only a few at present repay is 'Sketchpad' as described by SUTHERLAND
mathematical and computer study, the others (1963) at the Spring Joint Computer
being over-complicated. We usually cannot Conference, at which several other allied
attempt to calculate the effects of molecular developments were discussed.
collisions, and the simplest useful manageable
What would be most valuable for many
model is that of the reactor. For this, with the
practical parts of design appears to be the
aid of physical chemistry and chemical
genuine three-dimensional display with
analytical techniques, and with the help of a
facilities, rather like those of the electronic
statistlcian in planning the experiments, we
analogue computer, for feeding in and modifying
are able to set up a reasonable mathematical
lines and similar features. This is something
model. This can then be embedded in the
to look forward to.
!arger model of the whole plant.'
The combined model of the reactor and the It would seem that, although manipulatory
whole plant may be used in setting up a control procedures of this kind might be obtainable,
system for the plant, or if incorporated into a great difficulty will be found in attempts to set
model of the firm as a whole it lends itself to up programmes for creative behaviour or general
the study of new venture propositions. design problem-so! ving behaviour. MIN SKY
(1 961) notes that: 'It is very difficult to take
BYRNE and V AN KOOTEN (1965) discuss apparently common-sense procedures, such as
two ways in which models of reactors have those described in Polya's books on practical
been obtained. One was developed from heuristics in mathematical problem-so! ving,
theoretical considerations; the other was and put them into machine-usable form.'
obtained by working back from the practical Against this should be set the work of SIMON
data from suitably conducted experiments on et al. (1958, 1962) and NEWELL (1963, 1964)
the plant concerned. Optimum behaviour is in relatively restricted fields. SHUBIK (1964)
calculated by using linear programming based has more recently stressed the difficulty of
on considerable approximation of the reactor codifying very general problem-so! ving
models. strategies. Beyond this MACKWORTH (1964)
Instead of modification of the plant has emphasized the even greater difficulty of
conditions by human operators on the basis of problem-finding.
Models and Creativity programmes, the models of procedure. Ev ANS

Central to a creative perfonnance in design and NEWMAN (1964) have produced a
would appear to be the possession of skill and speculative model of the process of dreaming,
facility in the manipulation of the most based upon analogy with computer operation,
important models within the specific branch of which suggests that dreaming is a process
technology. It might also be thought that skill which is necessary for the elimination of
with very general patterns of problem-so! ving 'junk' from the mental computer system. In
would be of prime significance. This, indeed, Chapter 13, Newman speculates about the
gains support from the work of GUILFORD nature of the aesthetic experience and suggests
(1959); but one might also expect that such that it comes from the satisfactory match
skill could be largely wasted unless adequate between the thought processes and the object.
models of the situation prevailed. It has been noted, however, that much creative
Since most design work is evolutionary in work comes through difficult and troubled
character the first level of creati ve skill might thinking and some from 'incubation'. This
be seen in the flexibility and fluency in recalls the earlier speculation on dreaming.
handling first-level models, z.e. models Might not incubation play the same part in
immediately related to existing products. The creative work as dreaming plays in normal
'schema-scrambling' discussed by Broadbent in life? Incubation, according to this view, is
Chapter 14 would apply here. What is needed is concerned with the deaning up and regeneration
facili ty in devising, hunting down, or applying of the mental programmes and the essential
the less obvious variations applicable. simplification of basic concepts and the
The Synectics group seem to rely on preparation of new interrelationships.
biological analogies for many of their If this model has any validity it may have
approaches. Perhaps this dass of models is considerable practical repercussions. As far
chosen since it comes from a range of systems as is known, present approaches to the
having a greater degree of variation than is· development of creative potential and to the
possible with the simpler mechanical systems. exercise of schema-scrambling depend upon
But this is a variation, or dass of variations, short-term operations, and upon the ability to
which tends to be more relevant to fields which produce answers in restricted periods of time.
have been considerably explored, particularly Perhaps, because of this, their success is only
in terms of mechanical gadgets. Other people in terms of trivialities. For major creative
have found metaphors a good natural source operations it might be that the suggested deep-
of distortion of schema. down and radical process is necessary.
It is the attempts to escape from prevailing The working models of creative behaviour
major concepts, usually system concepts may be wrong. To achieve greater success in
which carry their own apparatus of organization the art of producing radical new ideas - the
and obedience, which seem to need rather field of 'antithetics' - new models may be
different models. These are the simplified needed.
models which pennit the generation of com-
pletely new system structures; structures now Strategies of Model Deployment
possible because the old linkages are dissol ved The practical skill of the designer lies
away by the process of thought. This may be in his deployment of the different kinds of
helperl by a change of circumstance or a lapse model. ARCHER (1965) notes that: 'Engineers
of time. may be weak in the systematic construction of
However, in the process of thought there the brief and in searching for original design
has to be some reorganization of mental ideas, but they are strong in the techniques of
the development of detail '. These techniques points are raised in the Iron and Steel Institute
are, of course, model techniques. Areher Symposium on Pilot Plants ( 1965).
recommends that the most general models
should be used in the opening stages of design Practical Outcomes
for the development of concepts, and that the There is now substantial agreement about
most detailed models should be used at the end, the significance of the different classes of
both for the design activity as such and for models in design, both for problem-so! ving and
communication. This is shown by the Watts communication. What is still to be decided is
model of the designer. whether the models and the methods of using
But although one may stipulate a very them can be improved. What about the poorly
general approach, as discussed by GREGORY developed areas of search and selection? Are
(1966), which has the merit of including more the best techniques being employed, or may
potential solutions within its 'catch' and they be improved? By how much should
going to the limits of the bounding system, disciplines come before data? What, in fact,
there is also the need to be intensely practical. are the important disciplines? These are the
There is, indeed, a continual requirement to kinds of problems which may be tackled by
reinforce the broad concept with practice. suitable research.
There is a dialectical interplay between the There are also the larger models, the
manipulative model and the fact-giving model. general models of problem-so! ving and the
models of the designer; and models of creative
The kind of generality possible depends behaviour. It is against such models that
upon the state of development of the system, practical achievement may be set, and from
its complexity, and its Soulding degree. It
validated models that new educational
also depends upon the more usual components
procedures may be developed.
of the design situation. In addition there are
In Chapters 18-21 the opportunity is gi ven
the obligations which help to determine of looking at the use in design of two technical
strategy: (a) the need to complete the design- procedures, the use of economic analysis, the
tk.e work must proceed; (b) the need to design use of computers, and two complete design
with limited resources - every avenue cannot sequences. The first of these is a typical
be explored and those most likely to yield industrial design operation carried through
results according to experience must be chosen; without self-consciousness and dedicated to a
( c) the need for the product to work. practical outcome. Its value lies in the way it
Since paper studies tend to be eheaper shows a varied deployment of models. Then an
than fact-giving studies, particularly in the example is given of a design carried through in
case of the more complex systems, the critical a special educational situation. This was
discussions centre on the reasons for con- intended not only to give a practical design
structing pilot plants or prototypes. This but also to aid the designer to restructure his
merits a discussion of its own, but most of the mentallllodel of design as a whole.
Chapter 18
THE IMPLICATIONS OF ECONOMICS IN

ENGINEERING DESIGN
A.P. Shahbenderian
lntroduction and manufacture of a plant or product; but in

The commercial success of an engineering most industrial environments techno-economic
design (whether of a large-scale plant or design considerations are often paramount,
individual product) depends largely on the with the aesthetic aspects influencing in a less
technical, economic and aesthetic tneri ts of quantifiable way the eventual design. The
the design, and although these merits can to 'information flow diagram', shown in Figure
some exten t be judged independently, the 18.1, indica~es the interaction between these
design process itself involves a close study three aspects and their place in the overall
of the complex and dynamic interrelationship design and production scheme.
between them. Even a superficial examination The arrows in Figure 18.1 indicate the
exposes this interrelationship between the direction of information (or influence) flow and
technical, economic and aesthetic aspects of feedback. Company policy influences design
design. For example, a technically efficient motivation which, in turn, influences a train
design often leads to profitable manufacture of. design activities (the design process)
and the lack of aesthetic merit can cause involving a study of the interrelated technical,
commercial failure, particularly in the case of economic and aesthetic aspects of the design.
consumer goods. Conversely, the desire for The result is a designed product or plant. The
high technical merit in a design can raise product or plant is then normally manufactured,
subsequent production costs to a point where and eventually either the product is sold, the
financial success is endangered. The success- plant is sold, or the plant manufactures a
ful design is therefore a compromise involving product which is sold. A market analysis of
an optimum combination of technical, economic the projected sales can then give rise to
and aesthetic merit. This optimum combination feedback information which, in turn, affects
can be arrived at qualitatively or quantitatively manufacture, design process and company
by the individual designer or by a large team of policy, as indicated in Figure 18.1. The
engineers, 'designers' and economists. lt is influence on company policy is such that
undoubtedly successful optimization that has fresh motives may be created for evolutionary
led to the design of such a plant as the Scottish (improved) or radical (novel) design. The design
Agricultural lndustries fertilizer factory at process therefore depends on the dynamic
Leith, or such a product as the 'Mini' motor interaction of those factors indicated in
car, or such buildings as those designed and Figure 18.1.
constructed by Nervi in ltaly. Quite clearly In many industrial design schemes,
there are other factors, such as social and aesthetic aspects are quite frequently severely
political needs, which can affect the design curtailed by technical (i.e. functional) demands
157
158 THE DESIGN METROD
- - - - -- Des1gn
-------- process
I
I
I •
I
I I
L---------- -- - -'
Figure 18.1. Information flow diagram
and, in a purely industrial environment, it is design depends strongly on the chosen

often permissible to consider them as implicit economic objecti ve function. In general, the
in the technical design. Since this chapter is problern also arises of whether overall or
essentially concerned with engineering design sub-optimization is to be used.
in an industrial environment, the discussion
is henceforth limited to design problems of a The Economic Environment
purely techno-economic character. The 'tech- The economic environment in which a
nical aspects' indicated in Figure 18.1 are company operates conditions the policy of the
taken to include: company, and company policy in turn in-
fluences design motivation. For example, the
(1) Conventional design techniques. design of a food processing plant or pharma-
(2) Evolutionary design techniques. ceutical factory for an underdeveloped country
(3) Radical design techniques. is likely to be different from the design lbf a
plant producing similar products in a highly
These technical aspects are discussed in industrialized country. In underdeveloped
conjunction with economic aspects involving countries having a planned and labour-intensive
a consideration of: economy, the incen ti ve will be to design
factories that will operate with a large labour
(i) The economic environment. force (for social as well as economic reasons).
(ii) The type of economic objecti ve function. Batch and manually operated processes are
(iii) Overall economic optimization. likely to be favoured rather than continuous
(iv) Economic sub-optimization involving: fully-automated processes. Furthermore, the
( a) Fixed costs alone. optimum life and size of plants will differ in
(b) Fixed and variable costs. underdeveloped and industrialized countries
(SHAHBENDERIAN, 1963) due to the differing
Conventi onal Design Techn iques cost of maintenance and repair and differing
By conventional design is meant the design restrictions on capital expenditure. The nature
of a plant or product using existing knowledge of the manufactured product is also likely to
and established practices. Most industrial differ due to different social attitudes and
design processes fall into this category and economic needs.
the successful design depends on a proper The so-called planned obsolescence of
study of the technical factors and of the consumer durables also extends to the design
particular economic environment; the final of entire plants - particularly chemical, in
which the rate of technological improvement is optimum capital investment for a particular
so rapid that the process (rather than the plant, and this in turn defines the equipment
equipment) used becomes obsolete in a design capable of operating economically at
relatively short time. the optimum capacity level.
The Type of Economic Objective Function Overall and Sub-optimization

The optimum conventional design is The above discussion on the influence
affected not only by the economic environment of the economic objective function presupposes
but also by the type of economic objective that at each plant capacity level an
function used. For example, if the design of a economically optimum design has been made
complete plant rather than of a particular of every component of the plant, with each
engineering product is considered, the optimum component linked tagether to form an inte-
size of a plant(i.e. capital investment involved) grated, functional and profitable complex.
will depend on the objective function chosen. lt is very often possible to design individual
As shown in Figure 18.2, the optimum invest- components for optimum techno-economic
ment for a typical sales and cost pattern performance (i.e. to carry out a sub-optimi-
depends on whether profit, percentage return zation) without reference to other components
on investment, or venture profit is chosen as in the complex design; but quite often the
the o bj ecti ve function. linking tagether of sub-optimized components
There is a tendency nowadays to use inore produces an overall design that is less profit-
sophisticated economic objective functions, able than a plant that has been designed on
the basis of an overall economic optimization.
In general, overall optimization techniques
I
~ Manufactunng
should be used whenever possible and a point
1 costs on the manufacturing cost curve shown in
C/year I (O)
Maxtmum Figure 18.2(a) should be minimal at any
apactty annual I
1 for maxtmum proftt 1 particular investment. The following example
annuat prof 1
1
of a reactor-separator system design (based on
0 1Cap1tal mvestment (plant capac1ty) C -
I
I
I
I
that of KRAMERS and WESTERTERP, 1963,
1 Prof1t I
/ I
and a 1 ready p u b li s h e d in full by
J: /year (1:1) SHAHBENDERIAN, 1964) will illustrate this
point and show how the objective function
influences the design of individual plant
0 'Cap1tal Investment (plant capac1ty) C -
I I 1
components.
I I
Prof1t 1 /Siope proport1onal to
'- m1mmum acceptable
percentage return on
Overall Optimization: Reactor-Separator System
J: /year ~nvestment A first order reversible and isothermal
(C) reaction A ~ B takes place in a tu bular reactor
0 Capital mvestment (plant capacity) J:. - as shown in Figure 18.3. <Pm lb./h of pure A
are mixed with <P.mr lb./h of recycle A before
Figure 18.2 entering the reactor (in which plug flow
occurs). The products of reaction pass to a
such as present worth, venture worth (HAPPEL, separater where cj:;m lb./h of pure B leave as
1958), interest rate of return, etc., using bottoms product, and cj:;mr lb. /h of pure A are
discounted cash flow techniques. In general, recycled. The nomenclature and example is
the use of each function leads to a different based on that of Kramers and Westerterp and
for the first order reaction considered the </>mr lb /h recycle

---
reactor volum e Vr is gi ven by the equation
-~--
<f>m l b /h feed
'!m_ ~c
V _ _
r - t--
K
k (K + 1) ~AL In I 1 -
( K :o AL
) A ~=1
where k is the reaction velocity constant, K is W8 =1

an equilibrium constan t, p is the density of A
and B (assumed equal) and ~AL 1 - WAL = Figure 18.3. Reactor-separator system
= WBV where WAL and WBL are the mass
fractions of A and B respecti vely leaving the Capital and operating costs were specified
reactor. A numerical calculation has been
as afunction of V,, cpmand 4mr·
published by Shahbenderian (1964) in which
the following values were used: The capital cost 1R of the reactor was
assumed to be gi ven by 1R =.t:900 V, 0·7
k = 0 ·09 h·•
K ~ 10 The total annual manufacturing cost was
;:; = 60 lb./ft., compounded and, assuming that B could be
The production rate c/Jm was fixed at 100 lb./h. sold for 3 d./lb., Table 18.1 was drawn up
Table 18.1
Yield Reactor Total Sales Armual

c volume manufacturing revenue pro fit
:oAL
p
vr cost s
c
(ft?) (i/year) (.f:/year) (.E/year)
0•4 24 ·55 7,952 9,900 1,948

o·s 26 ·88 7,672 9,900 2,228
0•6 30 "34 7,555 9,900 2,345
0•7 35 "40 7,552 9,900 2,348
o·s 44 •89 7,724 9,900 2,176
0•9 86 •10 8,811 9,900 1,089
Yield Total capital investment Percentage retum R Venture

c of reactor and separator on investment pro fit
:,AL
VP
IR + 1s =1
(.E) (o/o per year) (.E/year)
-·- --------- -~----- 1 - - - - - - - - - - - - - --
0"4 10,314 18 ·8 398
o·5 10,636 20•9 632
o·6 11,292 20"8 652
0•7 12,216 19"2 514
o·8 14,050 15"4 68
0"9 21,414 5 "1 -2,127
showing the reactor volume, total manufacturing design of plant components can be reached
cost and various profitability criteria as a without reference to the other plant components.
function of yield .; AL' The simplest type of sub-optimization
problern is one in which annual operating
The last column in Table 18.1 shows the
costs can be ignored and only capital costs
venture profit VP, calculated according to the
=
equation VP P - Rml! 100, where R mis the
(i.e. fixed annual costs) need be considered.
A typical case is pressure vessel design as
m1mmum acceptable percentage return on given by Happel. The following simple example
investment, taken in this example as 15 per on the optimal design of tin cans also
cent/year. The venture profit is clearly the illustrates this point.
excess of actual profit over the minimum
acceptable profit. Fixed Costs Alone: The Optimal Design
Table 18.1 shows that if total cost C or of Tin Cans - In the manufacture of closed
profit P are chosen as objective functions then cylindrical tin cans using tin plate of a given
the same optimum design results (,;AL in the thickness, it is necessary to minimize the area
of sheet meta! used. (i.e. the capital cost of
region of 0 •7, Vr = 35 • 40ft'), but that if
meta!).
percentage retum R, or venture profit VP, are If the diameter and height of the can are
chosen as economic criteria then the optimum D and H respectively then, for a can of fixed
designs will differ (,;AL in the region of 0·5, volume V
V r = 26 ·88 ft. '; and ,; AL in the region of 0-6, 77 D 2 H
V (18.1)
vr =30·34 ft.', respectively). Thus the Optimum 4
design (i.e. reactor volume) of reactor-separator
system clearly depends on the choice of the The area of sheet meta! required is given by
profitability criterion. A where
Sub-optimization A 71 DH + 71 D2 (18.2)
In the previous example an overall 2
economic optimization was carried out on the
reactor-separator system. It would have been or, from equation 18.1,
invalid to carry out an economic sub-optimi-
zation on the reactor alone: for instance it 4V
A 77D - -2 + (18·3)
would be invalid to find the reactor volume 71 D
that gave minimum total operating costs of
reactor alone, since this would indicate the Differentiating with respect to D and equating
smallest reactor possible as the optimum. to zero in order to find the minimum area
(This would mean high recycle pumping costs required, gi ves
due to incomplete conversion of reactant A and
high separation costs in the separator). Thus, dA 4V
+ 71 D
in this example, an economic balance between dD D2
reactor, separator and recycle costs must be 0
calculated on an overall basis; sub-optimization Thus, if D denotes the optimum value of D,
0
of reactor or separator design does not then
correspond to the design attained on an overall
basis.
There are, however, instances where
sub-optimization is valid and the optimum Note that d 2A/dD2 8V!D' + 71 which is
positive and thus corresponds to a minimum (1) The cost of inserting and removing
value of A. Therefore, using H 0 to denote the components from the machine.
optimum value of H (2) The cost of the cutting time.
(3) The cost of replacing worn tools.
(4) The cost of reconditioning worn tools.
He then proceeds to minimize total cost for
(4~r various conditions of tool speed and feed,

assuming that the machine tool, the work
Therefore material, the depth of material to be removed,
and the type and angles of the cutting tool are
{ ;:} = 1 initially selected. lf none of these latter
variables had been fixed then the problern
Thus, if the volume of a can is specified, the would have been more complex to take into
area of tin plate required is a minimum when account the interaction between these vari-
the can diameter equals the can height. ables, together with any other factors affecting
MALLY A, KING and EILON (1964), in the machining rate. It would, in fact, have
discussing the optimal design of tin cans, have been more a problern of overall rather than
shown that if seam and trim losses are taken sub-optimization.
into account in forming the can body and
punching out the metal caps from metal plate Evolutionary Design Techniques
then Using conventional design techniques,
i.e. existing knowledge and established prac-
{ ::} = 1·24 to 1-42
Table 18.2
Yield Reactor Total
the Variation from 1·24 to 1·42 being dependent volume manufacturing
~AL coat
on the effectiveness of nesting in punching vr
out the end caps from metal plate. c
The optimum design is again realized as a (ft. 1) (E/year)
result of a sub-optimization problern in which
only the capital cost of the meta! plate is 0•4 22.21 7,865
involved. o·5 24.55 7,581
0•6 27 '14 7,431
Fixed and Variable Costs- There are many 0•7 32 '14 7,426
designs resulting from sub-optimization o·8 40 ·5s 7,580
0•9 78•40 8,611
techniques in which both fixed and variable
costs are involved. For example, the optimum
Yield Reduction /::, C
design of heat exchangers (Happel), the
in C (c.f. Table 18.1)
optimum thickness of pipe lagging (Happel) and ~AL
(i.! year)
the optimum design of nuclear reactors
(MARGEN, 1960) depend on a balance of fixed 0•4 87
against variable costs. Another typical example o·s 91
of sub-optimization is the selection of economi- 0•6 124
cal machining rates. BROWN (1962) shows that 0•7 126
the cost of a machining operation can be o·8 144
0•9 200
divided into four main items:
tice, it is possible to arrive at an optimum Table 18.3

design based on an appropriate economic objec-
tive function. It is then interesting, and often Yield Total Reduction !:::. C
advantageous, to investigate the direction in ~AL manufacturing in C
which the design is likely to evolve as a result cost C
of changes in the technical or economic para- (.f/year) (.t/year)
meters. This exploration into a techno-economic
region that is at present unattainable may be 0'4 7,817 135
called evolutionary design. o·5 7,528 144
Consider, for example, the reactor-separ-
0'6 7,397 158
ator system discussed previously. It is
interesting to consider how the overall optimum 0'7 7,377 175
design would change as a result of changes in o·8 7,518 206
the technical or economic parameters. Such 0'9 8,485 326
parameter changes might involve expenditure
on research or development, on improved
management or sales techniques, or on superior The new reactor volume V r required and
raw materials. Consider, for example, the the resulting total manufacturing cost would
implications of a 10 per cent increase in the then be shown as in Table 18.2 (compare this
value of the reaction velocity constant k, from with Table 18.1).
0 ·09 to 0 ·099 h-1, due to (say) the discovery of The final column in Table 18.2 shows the
a new catalyst or the purchase of a superior reduction in total manufacturing cost that would
catalyst - without any increase in the cost be brought about by the improved catalyst.
parameters.
Table 18.4
Yield Affect of 10 per cent Affect of 10 per cent Affect of 10 per cent
~AL increase in k. increase in sales price. decrease in~.
Reduction OC in C Increase in revenue Reduction in C
(.f/ year) (.t:/ year) (.f/ year)
0'4 87 990 135
o·5 91 990 144
0'6 124 990 158
0'7 126 990 175
o·8 144 990 206
0'9 200 990 326
Yield Affect of 10 per cent Affect of 10 per cent

~AL decrease in raw material decrease in labour and
prices. Reduction OC in C overhead costs.
Reduction !:::. C in C
(.f/ year) (.t:/ year)
0'4 165 300

o·5 165 300
0'6 165 300
0'7 165 300
o·8 165 300
0'9 165 300
Thus, restricting the discussion to changes in benefits likely to accrue from process improve-
total manufacturing cost C, a 10 per cent in- ments. A subjective assessment is clearly
crease in k has the effect of reducing the necessary concerning the possibility and prob-
optimum total manufacturing cost (at ,;AL = ability of such changes.
0 • 7 as before) by H26/year.
Changes in economic parameters can also Radical Design
be investigated assuming the value of the There are stages in the design history of
technical parameters (k only in this case) to most industrial products where evolutionary
remain constant. Thus, for example, the reduc- design reaches a technical or economic limit;
tion in capital cost IR of a reactor of volume for if the designer chooses an inherently poor
system in the first place, techno-economic
~, from IR= !900 V,. 0 " ' to IR= !810 ~ 0 " 7 (10
optimization only helps him make the best of
per cent reduction) would lead to the cost
a bad situation. The need for radical (i.e.
changes shown in Table 18.3 (again compare
completely novel) design usually arises in a
this with Table 18.1).
forward-looking organization, but engineering
The total cost is again a minimum at ,;AL = designers are not always aware of the propi-
0 • 7 and the reduction in C is !175 for a 10 per tious time to discontinue evolutionary design.
cent change in the capital cost function. The A paper by STARKEY (1964) touches on this
economic implications of independent changes aspect. He discusses the influence of economic
in k or IR have been demonstrated in Tables factors on the commercial design strategy of
cathode ray tubes. He considers the case where
18.2 and 18.3. It is conceivable that improved an increase in technical merit of a product of
values of k can only be obtained at the expense conventional design may reduce the cost of
of an increase in values of IR (i.e. increased promoting sales but also increase the manufac-
capital costs); but a study of the effects of turing cost. If the sales price of the product
these independent variations is a very useful remains constant as technical merit is improved,
preliminaty guide to the amount of money that then clearly an optimum (i.e. minimum) total
can be spent - either on evolutionary design cost is likely to exist at some level of the tech-
and development or on improved materials. The nical merit - assuming that technical merit has
financial implications of improved technical a quantitive significance. There is also likely
design can also be compared with, for example, to be a lower and upper limit to the technical
the economic advantages accruing from reduced merit, below and above which the manufacture
raw material costs as a result of better purchas- of the product will result in a financial loss.
ing efforts. The influence of a reduction in This is simply an expression of the fact that
labour and overhead costs can also be assessed. cheap but shoddy goods are unlikely to sell
It can be seen from Table 18.4 that even these profitably and that it does not always pay to
limited independent variations furnish a greater improve too much the design of a particular
understanding of the feasibility of evolutionary product. Starkey then considers the economic
design and the nature and extent of any finan- implications of a product of new design (i.e.
cial advantages that may arise out of such radical design in the context of the present
improved design or cost reduction. It is clear, chapter) and shows that, if technical merit can
for example, that a 10 per cent reduction in be improved by effective co-ordination of de-
labour and overhead costs has a greater effect sign effort and if the cost of sales promotion
on overall manufacturing cost than a 10 per and manufacturing cost can both be reduced,
cent increase in the reaction velocity constant then an optimum technical merit will exist for
k. Furthermore a 10 per cent increase in sales the novel design, but at a higher merit level.
price per lb. of product would fumish an in- Although the unit profit arising from the
crease in annual revenue far exceeding any cost sale of the radical product may be no greater
than the unit profi t of the conventional product al product design and manufacture, and if the
(due to the possibility of lowering the unit economic limitations of improvement by evolu-
sales price of the radical product) annual sales tionary design are recognized.
and hence annual profit are likely to be greater.
Furthennore a larger share of the market may, in
due course, lower even more the cost of sales Conclusions
promotion. An attempt has been made in this chapter
Naturally, a radical design can only be pro- to show that there is a dynamic relationship
duced as a result of co-ordinated efforts within between the technical and economic aspeets of
a particular organization and any economic engineering design and that the successful
advantages that may arise will depend largely commercial design of an engineering plant or
on technical innovations. However, the need product depends not only on technical skill but
for radical design is always apparent if a proper also on a proper understanding of the economic
economic assessment is made of the convention- implications.
Chapter 19
THE USE OF A DIGITAL COMPUTER

IN DESIGN OFFICES
K.C. Parton
lntroduction be analysis programmes giving a specific

The digital computer has now been accept- answer to a specific calculation, or any amount
ed extensively as a tool for use in design of elaboration to synthesis programmes where
offices. This acceptance, although earlier fas- the computer may, by using logic, rapidly run
tered by the obvious value of the computer for through the whole series of designs in order to
Straightforward long arithmetical calculations, obtain an optimum. Figures 19.1 and 19.2 illus-
has now spread to cover a far wider range of trate the major differences between an analysis
interests. The various functions associated programme for the design of salient pole syn-
with design offices can be listed broadly as chronaus machines. The points that emerge
follows: from comparison of these two is that the analy-
sis programme does not require anything other
(1) Designing apparatus to meet customers' than a chosen set of known physical quantities
requirements. (i.e. standard stampings, pole and slot tools
(2) Detailing the design for manufacture. and stock copper sizes for existing standard
(3) Examining manufacturing problems. frames). The designer with his own experience
( 4) Checkin g test results. of machine design behind him can rapidly
(5) Considering the physical and perform- choose what he thinks would be the answer for
ance effects of new materials and techniques. a particular enquiry and feed these into the
(6) Constant appraisals of all new methods computer for an accurate performance pre-
to improve design from the point of view of diction calculation. Usually several alternative
improving cost, ease of manufacturing, designs are demanded giving results of minor
efficiency, noise, and general appearance. modifications that the designer would like to
examine. As this type of programme is extremely
In all of these the digital computer can be of fast, easily written and cheap in overall use,
assistance to some degree, but the economical it has been found to be extremely popular in a
limits of computers in these various functions variety of design offices for stock or standard
are now discussed. machines and in !arger and special machine
design groups. Day-to-day usage is consistent.
Designing Apparatus to Meet Customers' The extension of analysis programmes into
Requirements pure synthesis programmes, however, results in
The main problern facing any particular very comprehensi ve programmes which from the
group beginning to use the computer will event- management point of view may appear the ideal.
ually resolve itself into planning the type of Unfortunately, they have in many instances
programme to be tackled. In general, these can proved not particularly practical in the long run
167
Giv~ all ~chanical

1 Give all rTWchanical
Cust~rs' and
dirneonslons tor dm.nsions for
manutacturers'
stator core roter pot•s
spec•frcation
-------t---------- ----------~------- -------- ---+-----

and wrndrngs and w•ndings
Evaluate tundamentals, Merge all constants Evatuate fundameontals,

matenal werghts. with ~rmanent data, mat.,-ral wf'i ght5.
drmensions, magnetrzatron curves, dlfT'Ktnsrons.
conf•gurat•ons lasses .. confrgurat•ons ..
f f ' _.l_ _I_

Magnpt1c Los~s and Heahng and Mechanrcal
I I I
c•rcurt ~frctenc••s at ventJlatron requrreoments.
catcutallons vanous Ioads catc\_jatfons shafts, bearings ...
l
----------------------- ------1---------------- -----------
IPnnt relevant output reoquired tor design records
r------------ ----1
.I
I
I
'
Wh•n r"'luor•d 1ssu. compl~• sngp
instructions by opt•onal pnntrng
J
~------ --- -~
Figure 19.1. General analysis flow diagram for salient pole synchronaus machine
and are not used as much as was anticipated. non-standard designs are produced, (such as
One important technical reason for this has large altemators and transformers) but in gen-
been that even though written with the full Co- eral such programmes will be more used in
operation of the designers concemed, fixed obtaining a quick overall picture than for the
tolerances have to be set on many dimensions final complete design. Thus, one might for this
throughout the programme. However, in practice type of machine use a synthesis programme to
a designer working individually can at times obtain first sets of likely dimensions and then
relax these tolerances slightly when he feels obtain the best result with a really skilful de-
circumstances justify it. This is an insuperable signer starting from these and working with
problern as there is always a certain amount of analysis programmes backed by the organiz-
art in design which at times justifies flexible ation's experience. For this reason, some of
boundaries that cannot possibly be reproduced the synthesis programmes used in my organiz-
in automatic synthesis programmes. It thus ation have provision for rerunning with the
follows that a good designer working with previous logical decisions of the computer re-
analysis programmes will in many cases pro- placed by data specified by the designer. The
duce a slightly better design than a fully- problern requires a lot of serious heart search-
automatic synthesis programme, and with mod- ing, as the obvious publicity value of synthesis
ern commercial competition this difference may programmes and their apparent easy under-
be all important. Synthesis programmes, how- standing by senior management tend to make
ever, can be extremely valuable in the field of their advocation more popular than is justified.
very large machine design where completely The time, effort and cost necessary to produce
a really error-proof synthesis programme can of a very general synthesis programme could
become very large and must be compared with have a serious effect on the eventual design
the likely extra value that would be obtained by ability of the office, particularly in developing
spending the same effort in improving analy- junior desigpers.
tical techniques. In my opinion, if the objtct of the design
It must finally be appreciated that what- office is to produce the best possible designs
ever decision is reached will have a profound always consistent with up-to-date techniques
effect on the eventual design methods, overall and materials, there is no doubt that initially
engineering efficiency and organization of the the main emphasis must be on the development
design office. Furthermore, implicit acceptance of comprehensi ve and really good analysis
f j_ _l
Gtve preltm.nary start1ng
Customers' and G•ve prel1m1nary starting
values of destgn parameters.
manufacturers' value5 of destgn parameters,
flwc denstties. current
spectftcations toadmgs. for roter
toadtngs tor s.tator
l
- - -- - - - --- -- - - -- -- -------
Use tog1c to des1gn complpte stator
cores, s.tots, wtnding~ using
I IUse log1c to deS~gn complete roter
poles, w.ndmgs . ustng spectfled
destgn hmtts.
specified design Iimits
Preltmtnary destgn checksl Sat1sfactory Sat1sfactory JPrel,mmary des1gn checks J
Not satts.factory Not sattsfactory

J
~
_J
Use tog1c to mod1fy start1ng ,
dtmens!Ons.loadmqs wtthm the
spectfled ltmtts
I l Use tagte to determtne tf stator
redestgn tS preferabte
I
Ves ~No
J
I Use tagte to determtne tf roter
redestgn ts posstble
J c
0
I ~:J
No .e=:_ves
J
I I
a.
Merge all constants w1th permanent E
data, magnettzatton curve5,l055e5 Use tog1c to mod1fy start1ng , 0
u
parameters wtthtn 5pectfted \tmtts
\
f
Magnet1c
I f
I
i
• l __l
J
I
Losses and Heat1ng and Mechamcal
ctrcutt eff1Ctencte5 at venttlatton requirement s,
calcutattons vanous load5 calcutatlon5 shafts, bearings ..
Yes
l Use log1c to determ1ne 1f overall performance
meets the vanous spectftcatlons and design ltmtt5
;I No
- - - - - - - - - - -- ±_- - - - - - - - - - - -
l Pnnt relevant output reqUired for des1gn records 1
I -----1
r-
I Where req01red tssue complete shop J :J
'I...__
a.
mstruct10ns by opttonal printing ;
--- -~ 0
Figure 19.2. General synthesis flow diagram for a salient pole

synchronaus machine
1
~
I l ll
I
1_:_
Gtve complete stator, Gove only a no. of rotO< slot s
no of slots, conductors, and a lost of all possible
dlmenSIOOS rotor bars tn stock
----------------------.1-------------------------
[ Utll•z•ng tog•c select the more llkely bars to be cons1dered from stock I
' to proceed Wlth the destgn I

rBasiC evaluat•on of fundamentals reqUired
r Oetermmatton ot a!r gap compatlble Wllh deftectton, pulsation. noise I
tn1ttate des.gn procedure w1th f1rst selectE>d bar
~ _}_
Torques and
.
Mechan•cat
Magneue Losses and Heattng and drive requirements.
circuit eff•ctenc•es at v~nlttation
appllcat•on shatts.
catcutations vanous Ioads. calculatrons
checks beanngs.
•
_J
Istare au tntormat•on w•th•n the computer J
r '
Select next preterred bar 1f any
l•st completed L1st not exhausted
dl
l
ICampare sallent features and setect best design
-----------------------1-----------------------------
I
~--- ---------------- ___ ,
I Pnnt relevant output requtred for destgn records J
., __________ ------
1 l~here requtred issue complete shop instructtons by opttonal prontingJ
-----~
Figure 19.3. A mixed synthesis-analysis flow diagram for squirrel

cage induction machines
techniques and programmes. A limited amount designer is actively Co-operating in the project
of fairly Straightforward synthesis can be added throughout.
as required and is likely to prove the best
solution for many items of design. Figure 19.3 Detailed Design for Manufacture
illustrates a typical induction motor design Detailed design is a mixture of design
flow diagram of a heavily used programme em- office and drawing office responsibilities and
ploying this principal. involves a mass of routine detailed work that
In general, a balance must be maintained has to be done. For example, in an electrical
by which the designer receives the maximum machine complete details must be prepared of
benefit for the minimum computer cost. A rapt the total length of copper of a certain section
absorption in the art of programming to include that must be requisitioned from the stores, to-
all the logical intricacies of machine design gether with careful details and drawings of how
can be extremely unprofitable. This, however, to bend it to form the windings, and then the
will naturally be discouraged if a senior length and type of insulating tape to be bound
round it, etc. Many details like these appropri- Examin ing Manufacturing Problems
ate to each design are sent to the shops, re- During the course of manufacture, or in
quiring the continual preparation of copious test afterwards, there are occasions when the
detailed specification sheets and drawings. design cannot proceed exactly as originally
Once prepared, they have to be rigorously planned. When this occurs urgent decisions
checked since any mistakes, particularly on may be called for from the design office and
large machines, can lead to very costly scrap- hence the speed at which any modifications
ping of material, and hold-ups in the production can be calculated is most important.
chain if they are not discovered until manufac- The design analysis programmes, particu-
ture has started. The computer is, therefore, larly individual detailed items such as
being pressed into service to help in several mechanical stress programmes, are then heavily
facets of this work: used and quite indispensable under these
conditions.
(1) For various fairly standerd items the
computer can automatically list all the various
sections, component numbers and lengths, etc. Checking Test Results
and these can be arranged to be directly Because of the pressure of work on
punched on to the side of standard unscaled design offices, the checking of test results can
drawings with lettered dimensions. at times prove a bottleneck. This is naturally
(2) Where routine explosion of detail is a situation to be avoided wherever possible
required for a standerd assembly this can since it directly slows down the flow of work
similarly be broken down and scheduled by through the production departments. The test
the machine. calculations required are however often quite
(3) For the very important overall prepara- complicated; a lot of factors from the basic
tion of detailed main assembly diagrams etc., test figures recorded by the test department
programmes have been prepared to carry out (such as the calibration curve ofthe instruments,
much of the earlier checking work done by the efficiency of the test driving machine for
senior draughtsmen. These check programmes the speed and load, etc.) must be properly
can carry out rr:uch of the detailed basic taken into account in order to derive the exact
checking previously done by hand, since the performance of the new machine under test.
more junior draughtsman can directly fill in This often requires several different types of
the detailed input form for the computer. calculation, including references to various
Thus, providing the result is satisfactory, he calibration curves, nests of efficiency curves,
can then present his final drawing and speci- etc. These complications make it impossible
fications to his senior together with the output to burden the test department with the work and
from the computer checking programme. This the application of Computers to the problern
can substantially reduce the amount of time is ideal. Consequently programmes are often
needed for careful checking particularly as the prepared which will carry out this analysis and
senior will be fully conversant with the check- print out the required data both in numerical
ing programme and know exactly what he must and graphic form.
check hirnself and what can be left. Apart from Thus work is organized by rev1s1ng the
the saving of time the high accuracy of the layout of the test sheets filled in by the test
details prepared by the computer is a valuable department into a form suitable for direct
extra bonus that is continually being more feeding to the digital computer. The processed
trusted and appreciated. results are then returned directly to the test
CUSTOM[R, 5lAC~STChE5 AWr CC !~EHS JOe I TEST 1 HACHINE •D, 5TZ58Z6/I
• tESCRIPTION, J P~ASE eRUSMlESS ALTERNATDR 6/1/65
• o. KVA 230. KW 400, VOLTS 414, AHPS 50, CYClO 750, RP~ Pf o,a
• OPEN C!RCUIT CHARACTER!STIC

• . --- . --- .. ----- ..... ---- .... ----
GfNFUTOR D,C, DRIVJNG ~CTOR EXCITATION RP~ OPEN CCT, r I ELC IR ON
VOL TAGE CURRENT LOSS
• vnLTS AHP5
------------------
V"L TS VOLTS
----------AHPS Kw,
-,J~~,~O 409.00 . 5.55 3.70 lolO 750,00 58,50 3,30 oOI
• 110.00 4IO,JO 6,10 4,90 4,70 no,oo ao,oo 4,70 ,2]
I ?4 ,50 •IC.OO 6.~0 1.50 7o20 750.00 124,50 7.20
!46,00 410,00 IO,H 14,70 l4ol0 750,00 24~,00 14,30 2 0 01
• 1z~.oo 411,00 14.20 20,00 20,00 750,00 320,00 zo,oo
'"
;s,ee
J~z.•o •15.00 17.00 2], zo 2Jo20 7~o.co J6Z,•O 2lo20 4,&,
·H~~.oo 414,00 19.75 n.oo 27o00 BQ,QO 006,00 27,1)0 5,12
• ••r..oo 416.0~ 22.f0 31.20 30.90 150.00 ... o,oo ]0,90 6,88
4•z.oo 414,00 21,45 31,50 37 ,?.o 150.00 4SZ, 00 31.20 e,tz
5?2. 00 416.00 32.75 47,00 46o00 750,00 52Z. 00 46o00 10,e2
• Hf,OO 414,00 oo,8o 61,00 59.20 750,00 556.00 59,20 u,eo
5~0.00 ot3,00 47,00 73.50 TOo20 750,00 580,00 70.20 16,06
• D,C, MOTOR INPUT POWER WITH GENERATOR UNEXCITED 2o26 KW
RE5trUAL V~L T5 r,oo A~e. AIR TE•P, 20,0 DRIVING HCTOR WOG, TE~P, lOoO DR!VING HOTOR HOT RfS!ST, ,06200 OH"S
•
• SHORT CIRCUIT CHAOACTERISTIC
• ~ENFOATOR STATOR r,c, DRIVINr, HOTO~ fiCITAT ION STATOR SHCRT CCT, rtELn 5TRAY
----------------· ---------- tOPPE~ RP,.., CURREHT CU .. EHT LOSS
A~•s 111 AHPSIZI ------------------
VCL TS ·~•s YOL T5 AHPS T(MP, KW,
• 6?1.60 630,00 400.00 32.5J ]5,00 35o80 Z2.00 750. 625.80 H.•o z,,,
,,z,oo '12. 40 40o,o~ 24,00 29,00 Z9o20 Z2,60 750, 29.20 1,91
H3,20 433,20 400,on 19,00 24,70
'u, •o
• 2o.eo 23,50 750, 433,20 2•.•o 1,47
3~•.oo 354.00 400.00 14,70 20.00 20.10 Zlo50 750. 354,00 20 .t 0 1,04
21e,4o 218,00 •oo.on 9,00 12.20 12.50 23,00 750, 2te, oo IZ,50 ,l4
• ~~.20 !5,20 •oo.on 6.40 •.eo ••• o ZZ,50 750. 85.20 •. ,o tll
O,C, ~OTCR t•PuT POWER WITH GE•ERATOR UNEICITED 2·30 K,l,oj
• HS !rUAL AMP$ ,oo A•a, AlP TEMP, ZO,O D~IVING ~CTOR wn&, TE~P, JO,Q DRIVIHG MOTOR HOT RESISTo ,46ZOO 01"1~'~5
• GENFRATOR STATOR RES!S~AUCE ,00640 0HM5 AT 1q,5

GENF'ATCR STAT~R RES(STANCE ,oo6oe Orl~S ' ' 22.e
GENr•ATo• ROTOR RESISTA~CE ,96250 0H~5 AT ' ' · '
• DRIVING •cTOR REFEPENCE •c, 15!1126 Co•PUTER J05 HO 3450 MACMINE HO, 5T25826/I
•
•
Figure 19.4. A typical output result
department within two or three hours of the economical to consider quickly a wide range of
tests having been completed, and the testers new materials, dimensions, stress limits, etc.,
can themselves see whether the results are on a range of machines in order to fully
acceptable to the design office. If they are ascertain the optimum. The only prerequisite is
satisfactory, the results can be passed on to a clear head by the designer. Usually the
the design office and the machine released for number of practical possibilities is small
despatch at the same time. Only when the enough to make the need for any logical
results are clearly unacceptable is there any programmin g of the survey unnecessary.
need to consult the design office and ask for Further interesting large-scale appli-
their comments. When this happens, of course, cation of computers have been in the careful
the design office know that there is trouble evaluation of all component sizes and ratings
with the machine and can immediately give it a in order to construct a complete range of
priority that is impossible to give to a steady machines from the minimum of stock parts.
stream of ordinary test results that just need Further, the programmes carefully space out the
routine analysis. A particular programme for incremental steps between sizes to be used in
the open and short circuit test of a range of order to optimize on the widest possible range
synchronous machines carries out the following of machines for the lowest overall cost. Thus
procedure. most of the programmes are highly detailed
Winding resistance at specified tempera- economic exercises using advanced statistical
tures of the test plant are held, in an easily methods. Not only have surprisingly high
updateable form, by the computer as permanent savings been achieved but all the variations
data. True voltage, current and resistance of range can be tabulated and the breakdown
values are calculated from the given test of their parts sorted out at a moments notice.
figures, instrument constants and temperatures. The programme thus forms an integral part of
The various known losses are computed from the stock and production control arrangements
the above and hence the wire loss and stray for the factory. When necessary the programme
loss found. The output will consist of: is rerun in order to evaluate the cost of revising
the optimum designs being produced in the
(1) Revised input details. light of trading results and forecasts. Such a
(2) Open circuit curve. programme would be highly complex and need
(3) Iron loss curve. several man years of writing to complete, but
( 4) Short circuit curve. could of course quickly pay for itself in a
(5) Stray loss curve. factory which mass-manufactured a standard
range of small motors, for example.
A typical output result is shown in Figure 19.4.
Conclus Ions
General Design lmprovement In my experience, computers have proved
Aceurate and reliable analysis programmes most beneficial where they have been used as
are in continual use in the constant endeavour aids to designers, and not as substitutes. The
to improve equipment performance. In the reasons for this are described for the specific
precomputer era, for example, it was only case of e 1 e c tri c a I machine design by
possible to check on the likely improvement CONCEICAD and PAR TON (1963), and the
resul ting from a new material on only a few subject is discussed fully by PAR TON (1964).
specific designs; it is now practical and The broad issue however is that design is
basically an art more that a science, and possible of the routine aspects of design in
although computers can be programmed to order to leave designers free to concentrate
prepare regularly an adequate design quite their abilities on the fundamental design,
automatically, this design will hardly ever be work where the computer is very much inferior.
the best.
It is this little bit extra that the human
designer's intuition and experience can Acknowledgments
contribute that is vital in modern markets and I would like to thank the General Electric
must always be encouraged. The role of the Company Limited for permission to publish
computer must be to take over as much as this chapter.
Chapter 20
A PRACTICAL DESIGN: AN Oll BURNER FOR

LARGE WATERTUBE BOILERS
A.M. Needham
lntroduction outstanding problern is the corrosion of various

This chapter concerns the practical design parts of the boiler system which is caused by
of a critical component for an existing plant burning fuel oils with a high sulphur content.
which had to operate on a different fuel from When these fuel oils are bumt sulphur dioxide
that originally specified. It shows the need to is formed, and a small percentage of this is
study the interrelationship between the com- further oxidized to sulphur trioxide. This con-
ponents of the system and how the analysis of denses in the cold parts of the boiler, notably
the effects resulting from the change of fuel, in the air heater and combines with the water
tagether with an analysis of the system opera- vapour present in the flue gases to form sul-
tion, enabled an outline specification for the phuric acid.
component to be defined. From this specifica- The quantity of sulphur trioxide formed
tion, a design was developed on the basis of can be limited by reducing the amount of
existing knowledge, employing rig testing oxygen available in the combustion zone and
where it was not possible to predict per- considerable success has been achieved by
formance. The selection of materials and the this method. However, it has been shown by
detail design of the component were based on several laboratory investigations and a few
ensuring operational reliability under all boiler Operations in Germany that the excess
phases of the system operation. oxygen in the combustion zone must not be
As a result of these studies, tagether with more than 0. 2 per cent greater than the stoichio-
personal knowledge of plant conditions, the metric system requirement for complete elimin-
new component design has fulfilled expecta- ation of sulphur trioxide. This corresponds to
tions completely. a quantity of 1 per cent air in excess of the
stoichiometric or theoretical amount required to
burn the fuel completely.
Plant
A number of Central Electricity Generating
Board Power Stations were converted from coal The Problem
firing to oil firing in the late 1950's. These The boilers at a typical power station
power stations were mainly coastal stations (Figure 20.1) could not be operated at this very
with adequate berthing facilities for the tankers low value of excess air because carbon mon-
delivering the heavy fuel oil from various oil oxide and smoke were produced resulting in a
refineries. Many problems were encountered drop in boiler efficiency and infringement of the
resulting from buming heavy fuel oil in boilers Clean Air Act. Also the final steam tempera-
originally designed for buming coal, and not ture was reduced because of the change in the
all have been successfully solved. The major heat transfer within the boiler.
175
One bumer was tested on a combustion loads of 40 MW and occasional peaks of emer-
test rig and it was found that smoke was pro- gency generation of 66 MW. The boiler and
duced at an excess air level of 5 per cent and turbine had been designed for maximum effici-
that it was not possible to obtain the required ency at an output of 60 MW. The variation in
steam output only varied from 66 per cent to
Superheaters 110 per cent of the normal load.
The burner fuel system had been designed
to have a 10 : 1 variation in fuel flow and to
give a good fuel atomization over this range
a spill system of atomization was used. Twel ve
oil burners had been installed on the boiler and
Windbox variations in output of the boiler were achieved
by altering the fuel flow to all of them. The air
supply was also common to each and could be
adjusted in proportion to the fuel flow. As the
maximum air pressure possible was 6 in. w. g.,
and since flow is proportional to the square
root of the pressure, the windbox pressure at
10 per cent fuel flow would have to be 0.06 in.
Main oil
burner w. g., that is 1/100th of the full pressure.
Obviously, this could not be achieved in prac-
tice and the minimum windbox pressure which
Figure 20.1. Boiler plant could be used was 1.5 in.w.g. Below this
pressure combustion would become too poor for
air flow through the bumer itself when operat- normal boiler operation. The variation in boiler
ing at the normal fuel flow. Even at the max- output was 2·1, fixed by the air flow and not
imum air pressure which could be obtained on the fuel flow.
the boiler there was insufficient air flow It was therefore decided to dispense with
through the oil bumer to achieve the stoichio- the spill system of oil atomization and use
metric air-fuel ratio. lt was found that the simpler pressure jet atomizers, thus reducing
boiler was operated at above stoichiometric the nurober of valves on the fuel system by half.
air-fuel ratio only because of inleakage of air Boiler output could be controlled in steps of
either through the boiler casing, or through the 5 MW by shutting off burners, and in stages of
small burners used at start-up only. less than 5 MW by adjusting the fuel pressure.
As a result of these tests a decision had The variation in output required on each burner
to be marle on whether to try to develop the would then only be down to 87~ per cent of full
existing bumer or to design and develop a com- load, taking the case of a 40 MW load with
plete new burner. The power station staff were eight bumers in operation. The corresponding
consulted and decided to adopt the latter change in fuel and air pressure would be down
course. This decision was influenced by a to 72 per cent of the maximum.
nurober of factors both practical and theoretical.
Superheat
Boi Ier Operation The steam supplied from the boiler to the
A study of the operational records of the turbine must be at a constant temperature at
boiler and turbine showed that the normal out- all loads as the efficiency of the turbine is
put was 58-60 MW with infrequent continuous very sensitive to changes in steam temperature.
It has been found on all oil fired boilers that flow conditions. The final size could only be
as the load decreases the final steam tempera- determined after testing the burner under simu-
ture falls and the boiler has to be operated lated contitions.
with higher excess air to maintain steam
temperature. Stabilizer
It was reasoned that this is because the The stabilizer provides a stable recircu-
flame shape and total radiation change little as lating zone of fuel and air local to the atomizer.
the air velocity is reduced. To maintain a con- When this zone is ignited it provides an igni-
stant final steam temperature at a constant tion source for the remaining fuel-air mixture.
air-fuel ratio, the ratio of the heat transferred A swirl vane type of stabilizer which had been
by radiation to that transferred by convection previously developed for a smaller oil bumer
must be approximately constant. However, was known to be effective and so this was
flame size and emissivity is a reciprocal func- scaled up to suit the !arger air nozzle, the
tion of air velocity, since this governs the rate ratio of the cross-sectional areas being made
of mixing of fuel and air. similar.
Observations made at normal and reduced The air nozzle and stabilizer size govern
load conditions with the original burners the flame length and combustion performance to
showed that there was little change in the total a !arge extent. The maximum flame length
flame envelope. This would indicate that the which could be tolerated was fixed by the
amount of heat radiated from the flames did not dimensions of the boiler, and as there was no
change significantly as load was reduced and method of predicting burner performance, rig
thus less heat was available for transfer to the testin g and development were necessary to
steam by convection in the superheater. ensure that the oil burner would be satisfactory
If the system of shutting off burners to when installed on the boiler. This development
reduce load and operating the other burners work was entirely confined to the air nozzle
under nearly constant conditions were adopted, and stabilizer and so the other bumer parts
the total flame envelope would be reduced pro- could be designed to meet the specification
portionally to the load. It appeared hopeful that implied by the operational requirements.
the amount of heat radiated would be reduced,
enabling final steam temperature to be main- Atomizer
tained at all loads at a constant air-fuel ratio. An atomizer size was chosen which gave
the required fuel flow at the normal fuel press-
ure used on the boiler. The spray angle has an
Burner Design
Once the broad requirements of the bumers important effect on combustion performance and
necessary for improved operation of the boiler the most suitable angle would be determined
were decided, the more detailed design re- during the rig testing of the oil burner. How-
quirements could be listed. ever, the extemal size of the atomizers of
different spray angles are the same and so
Air Nozzle whichever one was finally selected, the mech-
The air nozzle diameter fixes the velocity anical design of the bumer would not be
of the air through the bumer for a given fuel affected.
flow and given air-fuel ratio.
It was known that for good combustion, a Air Valve
velocity of the order of 100-200 ft./sec, would The air val ve or register directs the air
be required and so a diameter was fixed to gi ve from the windbox into the air nozzle. The air
a velocity of 150 ft./sec under normal fuel is accelerated from a low velocity and is also
turned through 90°; this should be clone with give low excess air rates at all loads would
the minimum loss in total energy as losses at be lost.
this point serve no useful purpose. The only
pressure loss which is necessary is that Materials
caused by the stabilizer in creating the re- The burners which are shut off have no
circulation zone for flame stability. cooling air passing through them and so the
air nozzle, stabilizer and atomizer soon reach
The air valve was therefore designed aero-
very high temperatures. The gases in the com-
dynamically, and based on the principles used
bustion chamber contain products of vanadium
in the intakes for static testing of jet engines
and sulphur which can reach these burner
where similar flow conditions apply. The
parts when the burner is shut.
pressure loss was expected to be less than
The best material for these conditions was
6 per cent of the total energy.
known to be a high chrome low nickel steel,
The overall pressure drop of the burner to AISI 309 specification, and so the stabilizer
would be mainly determined by the air nozzle and air nozzle were designed to be made from
area and the stabilizer blockage. Calculations this material. The atomizer was made from a
based on the sizes initially selected and an hard stainless steel which softens with excess
estimated blockage for the stabilizer showed heating, resulting in rapid wear. This part thus
that the quantity of air required for stoichio- required protecting from heat and so was de-
metric air-fuel ratio at the normal fuel flow signed to retract inside the air nozzle, together
would pass through the burner at a windbox with the stabilizer, when the burner was shut.
pressure of 3 in.w.g. Examination of the This would shield it from direct radiation from
forced draught fan and the system pressure the flames of the other burners.
loss characteristics showed that at this wind- A further point which had to be considered
box pressure the required air flow could be was that, although the oil pipe and atomizer
obtained at 75 per cent of full fan speed. Air are purged of oil when the burner is shut down,
control on this boiler was regulated by means some oil remains inside. When this is heated
of fan speed and so fan power required would excessively, cracking of the oil takes place,
be less with the new burners. leaving hard carbon deposits which can block
the small orifices of the atomizer. A passage
Sealing was therefore designed so that cold air from
As the boiler was to be controlled by the boiler house would be drawn down the tube
shutting off burners to reduce boiler output it supporting the oil pipe and atomizer. A radia-
was essential to design the burners so that, tion shield in front of the atomizer was also
when they were shut, all air would be sealed designed to direct this air over the atomizer
off from the windbox. If this was not clone face. Cool air would always be drawn down the
then, since the boiler was to be controlled to a tube because of the low pressure zone down-
low overall excess air rate, the burners which stream of the stabilizer when the bumer was
were firing would be starved of air; the air operating. When the burner was shut off, air
leaking through shut off burners would provide would be drawn in since the combustion cham-
the required amount. This would affect the ber is operated at below ambient pressure for
flame shape and emissivity of the burners safety reasons.
firing and final steam temperatures would fall,
making it necessary to increase the excess air Air Distribution
rate to maintain final steam temperature. Thus In any type of boiler installation with a
the whole effect of modifying the oil burners to nurober of bumers taking air from a common
duct or windbox, there is usually a maldistribu- to enable the tests to be carried out in the
tion of air to each burner caused by static minimum time.
pressure Variations both in the windbox and in It was anticipated that the air nozzle
the combustion chamber. Thus it is desirable diameter would not vary as a result of combus-
tobe able to adjust the air flow on each bumer, tion tests, by more than ± 2 in. from the dia-
and if a boiler is to be operated at nearly meter estimated, so making it possible to finish
stoichiometric overall air-fuel ratio it is the detail design before complete test results
essential. Otherwise, some burners will operate were available.
with high excess air rates and the others with A number of important factors were con-
a deficiency of air, resulting in the former pro- sidered when designing:
ducing corrosive SO, and the latter producing
carbon and possibly smoke. (1) Ease of maintenance,
The adjustment of the air flow should not (2) Expected life,
affect the flow pattern through the air nozzle, (3) Weight,
otherwise combustion perfcrmance would be ( 4) Method of manufacture,
altered. The obvious way of achieving this was (5) Cost.
to alter the amount by which the air valve
could open. If the loads which could be expected in the
normal operation of the bumer were considered
Detail Design alone, the whole bumer could be manufactured
The basic framework of ideas for the oil in 16 s. w. g. materials. However, the air nozzle
bumer were now formulated and detail design could be subject to corrosion and high thermal
of the hardware was commenced (Figure 20.2). stresses and, aiming at a minimum life of
3 years, it was decided to increase the gauge
Operating rods Air val ve of this part. Consequently, the rest of the
I !
bumer had to be strengthened not just to carry
the loads but to allow for the type of handling
which the bumer would receive when being
fitted to the boiler.
To reduce maintenance to the minimum,
~ I
I
I
' ,;' I
Pneumatic Seals Carrier tube
cylinders
Stabilizer simplicity of design was aimed at, with the
least possible number of moving parts. The
pneumatic cylinders necessary to actuate the
B~rner backplate
retracting of the stabilizer and atomizer were
Figure 20.2. Oil bumer also used to operate the air val ve. In order to
reduce their working temperature, these cylin-
Sizes of certain parts were fixed by the dimen- ders were mounted as far as possible away
sions of the existing windbox and so the out- from the burner backplate.
line of the complete burner was laid out. The igniter, a propane gas-electric torch,
At the same time a very simple burner was was fitted just above the tube carrying the
drawn and manufactured for testing on the com- stabilizer and a larger tube carried both
bustion rig. This was simply a fixed air val ve through the two supporting bearings. This
and nozzle combination, which could be easily design eliminated a separate mounting for the
changed, and a central tube carrying the stabil- gas torch and its actuator.
izer and atomizer. A number of different size Since only twel ve bumers were required
air nozzles and stabilizers were manufactured and because short delivery was essential, it
7
was decided that it would be eheaper to use Prototype Burner

fabricated parts. With this in mind, all efforts The results of the rig combustion tests
were made to reduce costs by making parts showed that the size of air nozzle and stabil-
from stock sizes of material where possible. izer first estimated were the most suitable for
Problem- Corrosion
I
Reason -503 product1on
I
Solutions
Rvmovot sulphur Combustion Combustion Antt·corros1on

trom tu~tl additives control materials
I I I I
Cost ap.rational Control requirements Cost
difficulties
I Not posLbte w 1 t h \
Cost exist1ng burner
I
Rig lest burner
I
Test results
Develop ex1sting burner

I
Extensive mod1fications
requ1n~d
I
Present reliability pocr
Good combustion Tight air Capable ot a.r Rell~bil1ty Replace I
with minimum air shut ·otf flow adjustment eXISiing Not possible to meet all
S•mpticity burner
I I I I I
requirements
Low pressure drop Valve Des1gn ot vatve Minimum numb~r Size ;ond
of mov1ng parts cdst
across intake seals control length
I I fixed
Similar to static Mater~ als
jet engine intake
I
Figure 20.3. The application of information in the design process
Stock lists of local suppliers were found to be the boilers. This was fortunate and enabled a
far more useful than reference to British prototype burner, for installation on the boiler,
Standards to obtain tube sizes. Also, since it to be manufactured and tested before convert-
was known that the !arger diameter rings would ing all the burners.
be cut from stock size sheets, the external The manufacture and installation were per-
diameters were made so that two rings could sonally supervised by myself and a number of
be cut from one sheet, thus reducing wastage. changes in the design became obvious. Be-
cause of the wide tolerances which had origin- Conclusions

ally been allowed, it was necessary to increase This case history of the design of an oil
the clearances on parts which had to mate with burner is intended to show the need to consider
existing fittings on the boiler. These modifica- not only the part being designed but its inter-
tions were incorporated in the design of the relationship with other parts. First-hand know-
production burners. ledge of the operation of the plant, as well as
adequate consultation with the plant engineers
Boiler Conversion and operators, enabled a satisfactory design
Following the successful three month trial outline to be formulated. Close association
of the prototype burner, the other eleven burn- with the combustion test rig team resulted in
ers were similarly modified and installed. On many possible difficulties of burner operation
recommissioning the boiler, no difficulties being eliminated in the design stage. The
were experienced with the operation of the personal supervision of the installation, as
burners. The boiler performance came up to well as discussions with maintenance staff,
the expected standard, and superheat tempera- greatly influenced the detail design of the
tures could be achieved easily at part load. burner.
The control of the boiler during start-up also A flow diagram which indicated how all
improved, because of the elimination of inleak- this information was used in the design process
age air through the burners not in use. How- is shown in Figure 20.3. This illustrates how
ever, the improvement of the oil bumers has complex the design of a relatively simple com-
shown up the poor sealing of the boiler casing. ponent can be and the important part that good
Although the burners are operating at nearly communications play in the design process.
stoichiometric air-fuel ratio, the overall excess I believe that a designer must be not only
air rate is still too high to prevent the forma- a person who can produce elegant designs but
tion of sulphur trioxide. This excess air must also one capable of explaining designs to the
be entering the combustion chamber through the people who will use them, and of obtaining
boiler casing, since special attention was paid their Co-operation and experience to assist him
to the small burners used for start-up during in producing the best design to meet the re-
the overhaul period and these appear to seal quirements at the least cost.
well when shut. The inleakage has increased
since the boiler was recommissioned following Acknowledgment
replacement of some boiler tubes and is much I wish to thank Mr. R.H. Coates, Regional
worse than was originally expected. Steps are Director of the South Western Region of the
now in hand to improve the casing and so bring Central Electricity Generating Board, for
the excess air under complete control. permission to publish this chapter.
Chapter 21
PRACTICAL DESIGN BASED ON METHOD

(FUNDAMENTAL DESIGN METHOD)
E. Matchett and A.H. Briggs
lntroduction the method content of fundamental design

This chapter has three primary objectives: method lies at a deeper level than the applica-
tion of a systematic approach analogous to a
(1) To introduce a method of developing
computer programme, in that it intimately
further the design skills possessed by senior
involves a major part of the designer's total
practising designers, laying particular em-
mental skills, attitudes, knowledge and person-
phasis on the kind of learning environment
ality. There is little similarity with the kind
which has been created for this purpose.
of method that is a mere manipulation of
(2) To outline some of the kinds of thought
formulae.
and feelings that a designer experiences
during this development process in which
The Fundamental Design Method Course
formalized design projects are integrated with The course is based on the concept that
more traditional methods of designing.
the extent to which a designer is able to im-
(3) To invite discussion on possible weak-
prove further his design ability, is closely
nesses in such learning processes and how
allied to the extent to which he can become
they might be overcome, and on how their aware of his mental skills and attitudes em-
strengths might be further exploited - both in
ployed in designing.
connection with designers of considerable
The emphasis in all of the course work is
experience, and those just entering the pro-
on the further development of the individual's
fession, whose level of skill will influence
mental skills and on how the individual can
tremendously the pattern and quality of future
make better use of existing and future know-
ci vilization.
ledge in the designs of his company's products.
Many of the remarks made are illustrated In order to achieve this it is necessary for
by a practical example studied on the Funda- each course member to reflect deeply upon the
mental Design Method Course at Engineers' methods and objectives embodied in his daily
House, Bristol. Though this example is repro- work, particularly within that part of the work
duced in some detail it has not been possible which is the most creati ve and important. It is
to cover all aspects of the study, as this would assumed that within this part of his work there
have required far too lengthy and tedious an are inadequacies and redundancies of thought
exposition. It should also be noted in connec- of which he is unaware, yet which will open up
tion with this example that it employs only that possibilities for improved design once they are
part of the formal design approach covered on discovered.
the course which appeared best suited to the The course is intended for senior practis-
needs of the particular design project and of ing designers holding positions of responsi-
the particular individual concemed. Much of bility. Each individual designer is expected to
183
explore, under guidance, his own mental skills achievement and provides a deadline which
and attitudes and determine for hirnself how he adds reality to the exercise. The opportunity to
might more fully exploit their strengths and describe and discuss the methods and results
remedy their weaknesses, and work more delib- helps to consolidate what has been leamt.
erately and thoroughly towards clearly defined The tempo of the course is as rapid as
and carefully analysed objectives. possible, consistent with the achievement of
As a result of the course, and the Subse- an adequate depth of thought. Apart from lec-
quent train of events it is calculated to pro- tures, group discussions, and the individual
duce, a designer should obtain a progressively project, each course member takes part in
increasing degree of conscious control and several exercises and group projects, being
systematic working, with a heightened aware- responsible for one of the latter as a group
ness of what mental actions are necessary at leader. Care is taken to arrange the groups so
any point in the design process. that each man is likely to benefit from working
Each course member is visited before the with particular individuals.
course to ensure that, as far as is practicable, The bulk of the formal tuition is concerned
the tuition he will receive will suit his own with building up model concepts of the design
and his company's needs. A design project is process and objectives, which become basic
selected which is typical in scope and com- standards, used to direct and evaluate Subse-
plexity of the work which forms his most quent design thought. This involves the study
difficult assignments. The choice is usually of skills and attitudes which all successful
difficult owing to the very short period of time designers probably possess to a considerable
allocated to the project on the course, and the degree, though usually without realizing either
desirability that a satisfactory solution be their presence or their importance. Particular
found dudng the three-week course. The visit attention is given to describing the most funda-
also serves to prepare the designer for the kind mental and most commonly used elements of
of experience he is to undergo and to moti vate design thoughts which, to date, it has been
and condition his thinking, so that maximum possible to identify. The course member is not
benefit may be derived from the course tuition invited to accept such descriptions but rather
and practical work. Lack of preparation prior to to grapple with them, try to discover flaws in
the course and lack of factual data would them, and make them personally more meaning-
require assumptions to be made on the project; ful. Once this is clone, however, he is expected
this would then become more of an academic to attempt to make use of the knowledge in all
exercise than a determination of how to satisfy his design work. The revised model concepts
the demands of a real and exacting set of become the yardstick by which he decides
circumstances. whether the thoughts with which he is currently
The course is residential and assign- occupied are timely and in line with what is
ments are arranged for every evening and for theoretically required. In effect he uses mental
the first and second week-ends. This provides signposts and landmarks which he hirnself
a means of employing fully what is admittedly erected during the process described above.
an extremely short period of time in which to It is very important in this kind of training
help bring about significant changes in a de- that the course member absorbs and digests
signer's capabilities. The results of all the material given in lectures, so that it becomes
projects tackled and other practical work, are second nature to apply it at speed. Unless the
exhibited on a course Open Day to which bulk of material is so digested through thought
members of the companies represented are in- and practice during the course, it is unlikely
vited. This acts as an added incentive for that the intended improvements in mental skills
will be acquired as a result of reading course course members do persevere in trying to do so,
notes afterwards. This remark would also apply even though the process involves a good deal
to a great number of methods of charting and of strain, is indicative of the interest such a
analysing thoughts, which are discussed during form of study generates.
the course. The solving of any problern is aided It is sometimes suggested that this kind of
by the fundamental basis underlying all such leaming places a person under, too much strain
methods, rather than by attempts to employ a and makes him too susceptible to suggestion.
rigid technique. It is true that the course is designed to change
To cover the ground in depth is not possi- people - the development of one's abilities is
ble without the whole-hearted co-operation of not possible with a change occurring. Al-
each individual. Much of the material which though there is good evidence to suggest that
repays the deepest study is superficially change does take place, and often appreciable
simple and obvious, yet it is essential that change, there has been none to date to suggest
from the outset one is prepared to grapple with any sign of change for the worse. People 's
this. Other vital aspects of one's thought skills are developed, their horizons are widen-
processes involve factors which are not readily ed and they obtain greater satisfaction from
identified in lectures, etc., and which must be the deeper insight into, and control over, their
approached by use of analogy. For example, work. There is frequently evidence of a new
physical activities and methods of analysing dignity and enthusiasm which emerges from
these, present a useful analogy to mental knowing more · about the tremendous achieve-
actions and their analysis, provided that a ment that any design work of quality repre-
person is prepared to make use of such aids to sents - whatever the current status of a
understanding. Significant progress on the designer in the eyes of the world.
course is not possible without such endeavour.
Each individual is required to consider object- An Example of the Application of Fundamental
ively aspects of his own thinking which may Design Method
not be similar to that of other people. He can The project arose out of the need for the
only be assisted up to a certain point, beyond company concemed (English Steel Corporation
which a concentration of thought and energy is Limited, Sheffield) for a mine car coupling
required that can only be his own.
Several of the course exercises are con-
cerned with simply writing reports on what one
believes to be the stages, and criteria of
judgment, employed in reaching design decis-
ions of different kinds. The early exercises
are almost entirely unstructured, but later ones
invite comments on specific aspects concerned
with the application of the imagination and
j udgment, and particular difficul ties one is
likely to experience. Such exercises are not
easy to construct in that they must, with as
little biasing effect as possible, draw out from
the individual those details of the approach he
has been usin~ which can be most beneficial
to him once they are exposed. This is normally
not at all easy to achieve and the fact that the Figure 21.1. Typical tippler installation
which would dispense with manual operation at dispensed with the need for manual operation,
the uncoupling and coupling stages. Such oper- and thereby from the customer's point of view
ations had hecome a standard practice in many reduced manpower.
coal mines, where the mine cars are discharged The project was chosen for the Funda-
by rotary dumping which necessit ates tempor- mental Design Method Course, where the major
ary separation of the train (see Figure 21.1). features of the new design were established.
The Willison mine car coupler couples Calculations, detail drawings and subsequent
automatically but must be uncoupled manually, prototypes have proved the effectiveness of
and in some cases must be neutralized so that the design which is shortly to be manufactured
the cars may be shunted tagether wi thout coup- in large quantities.
ling. Previous methods of neutralizing Willison As with all projects studied on the course,
couplers are shown in Figure 21.2. Reports an attempt was made to record those parts of
were received from the service engineers which the design decision process, and relevant
made it obvious that as coal production in- factors, which were judged to be most likely to
creased, haulage efficiency must also increase justify detailed study.
to keep pace; one way of doing this would be Such a judgment is largely a matter of
to produce a fully automatic coupler which employing one's past experience, though this
proved far less valuable in the early stages
than the idealized model concepts of the design
process and fundamental elements referred to
previously. In order to expose the decisions
involved, a number of charts based on those
given on the course, but modified to suit the
particular project, were drawn. It should be
noted that these charts to some extent took
the place of sketches which would normally
have been produced in such a design study.
They also recorded thoughts which would
Figure 21.2. Previous me thods of neutralizing Willison couplers: (a) coupler with lock retraction
pm; (b) coupler with operating gear on car end; (c) integral coupler with neutralizing plunger
traditionally be the non-recorded links betweeo. Related to this last comment is another
sketches. Such normally non-recorded thoughts vitally important one. Although the charts are
and factors include the reasons for making a produced individually and are initially analysed
decision, reasons for deciding against a in that sequence felt to be sensible, the inten-
particular course of action, relationships of sity of thought invol ved is such that the salient
items in time and sequence, the number of points of all previous charts tend to be kept
factors influencing a particular characteristic, in mind ,_;,hen analysing any one of them. There
and the patterns and trends of decision is a frequent back-checking of the significance
clusters. The value of formally recording such of a proposed decision or new factor revealed.
thoughts and factors can be judged from a The fact that all previous decisions and data
detailed study of the full series of charts have been recorded is of immense value in
(Figures 21.3 to 21.12) produced whilst rapid thorough cross -checking. This value is,
evolving the design. One should particularly however, minor compared with the effect that
observe in this connection an additional the initial exposure of thoughts has had in
benefit of using charts in preference to bringing them and their various relationships
sketches in the early stages: this is the into the centre of one's mental focus.
greater freedom from the biasing and restricting Another important general point arising is
effect of anything pictorially defined. Its that the mental breakthrough which has been
pictorial form almost inevitably triggers off sought by means of the charts frequently
only such ideas as are closely allied to it. appears to have no connection with them on
superficial examination. It might appear to
The particular series of charting methods
be a new factor or aspect not recorded on the
· chosen proved to be a useful one. lnitially,
charts, but is actually a projection of the
the ones shown in Figure 21.3 to 21.5 were
recorded factors and aspects which has been
most helpful towards obtaining a rapid feel
created by concentrated thinking around the
for the problems involved, and for the most
information charted. One does not know what
likely area in which a practical solution would
one is looking for until it appears, yet it
occur. Once this was sufficiently determined
appears because one has been looking for it
a sketch design was produced to act as a
and it is readily recognized as a solution.
catalyst for further charting. Every subsequent
chart was produced as a result of a deficiency For those who would object that such an
in the preceding one, which concentrated approach is neither scientific nor very system-
thought around it had revealed. This progres- atic, we would be only too happy to agree. We
sion of the charts is important, and is closely would point out, however, · that it works, and
interrelated with the striving for some decisive works well. What is actually happening is that
clue as to which path to take. Hence both the 'human computers' are being given at least
nature and sequence of charting are not a more positive guidance than before, whilst
matter of simple logic which can easily be being freed from much bias often found in de-
explained and copied for other projects, but signing. It is an approach which permits a good
arise out of the needs of the project in band deal of systematic design without having to
and the needs of the individual concerned. resort to the use of electronic computers. It is
The Iead obtained on how to proceed is also one which is highly practicable until such
probably more related to the intensity and a time as the design process is so understood
concentration of the individual's own thinking that satisfactory computer programmes can be
than his detailed knowledge of charting written. The design considered, though re-
methods available, or the clarity of his con- latively simple, involves many thousands of
cepts of design fundamentals. factors and their interrelationships. No method
is yet known which is capable of producing the (15) Histogram prepared to show relative
design by computer (certainly not in the thirty usefulness of each scheme as an aid to
hours actually taken to produce this design); implementation (Figure 21. 21).
neither is any design method so systematic yet (16) Release of designs for trial production
so simple as to dispense with the need for the (Figure 21. 22).
concentration of thought applied to this project.
Detailed Description of Charting and Analysis
Major Phases in the Design Study on the Project
(1) Investigation of primary functional The design process was commenced by
need - the one need which if not satisfied in- drawing up a chart of the primary functional
validates all other achievement (see Figures needs of the proposed design (Figures 21.3,
21.3, 21.4 and 21.5). 21.4 and 21.5). Although not shownon any of
(2) Design sketches to provide partial or the charts, one of the primary functional needs
complete solutions to the primary functional of the company was that the device must be
need (Figure 21.6). contained within the head of the coupler, and
(3) Preparation of a !ist of items and the that the coupling contour was not to be modi-
functional means they provide (Figure 21. 7). fied because of the necessity of interchange-
( 4) Study of the possibility of eliminating, ability with existing couplings. The primary
combining or transferring etc., items or whole functional need chart listed the various basic
sections of design (Figure 21.8). methods of operating a coupler. After consider-
(5) Use of the functional process chart to ation it became apparent that only a mechanical
show the sequence of operation of design device could be used on the coupler, although a
(Figure 21.9). hydraulic or mechanical means, or combination
(6) Charting of operational Variations of of both, could be used to operate the track
coupling and uncoupling devices (Figure device. It was later established that a mechan-
21.10). ical track device would give the simplest
(7) Use of the functional process chart for solution.
new design (Figure 21.11). To initiate the thought process, a sketch
(8) Study of motion of Operating cams was made of an extremely inelegant device
(Figure 21.13). which nevertheless satisfied every aspect of
(9) Preparation of wooden models to check the primary functional needs (Figure 21.6). In
operation of 90 ° and 180 ° cam mechanisms this sketch, the finger A makes contact with
(Figure 21.14). the neutralizing stop on the track device and
(1 0) Preparation of preliminary sketches of is rotated through approximately 70 °. The
parts of mechanism and preliminary stress spindie B and peg F also rotate against the
calculations. spring I in torsion and the collar G is raised
(11) Preparation of prototype coupler includ- against the spring until the peg F positions
ing modifications to parts to ensure correct itself in the V-notch. As the finger A rotates
form of operation (Figures 21.15, 21.16 and the double cam C inside the coupler also
21.17). moves through the same angle and retracts the
(12) Preparation of final drawings and detail lock D against the lock spring E. (The lock
stressing (Figure 21.18). and lock spring are not illustrated.) To reset
(13) Use of chart to check functional effec- the mechanism, a ramp in the centre of the
tiveness of the design (Figure 21.19). track raises the plunger H which in turn lifts
(14) Investigation of field application the collar via the connecting pin /, thereby
(Figure 21.20). releasing the peg F which permits the spring to
Prtmar ~ f u nct I anal need
PAM ol retr~cttnga lock aga1n!.t a s.pnng~

retam•ng and releas.tng as reqUired
Rea~on -Uncoupltng two couplers

Ttme --When tratn ts to be partf'd
Place -Entertng and teavmg ltppler or CiiQf'
Means. -Non ·manual opl!'rat•on
Method -Molton of car to operate device
j ··-~
_T_r_~~ ~- d_~l!:l~~ Cou~ter de V!Ce
N"ed -Convertmg car motton Need --Trans.fer mot 1on to lock

Reason- Oper ate coupler Reason-Retract and r es.et lock
Ttme -Every t•me tn any dtrectton Ttme -Every ltme in sequence
Place -Where tratn ha~ to part Place -lns.tde exts.ltn g couPter head
Means. -U-:.e car motton Means -Postttve ac:.ho n, fa1l s.afe
Method- Etectncdl. mechantcal. Method -Etectncal, me chan,cat.
magne-ttt, hydraul•c. pneumalte rnagnet•c, hyd raul1c, pneumatic
II
I Magnet.c Hydraul1c I I Magnet1c I H
PAM PAM PAM
IEt.,clr~cal]
P.A M
rechanlcai I PneumafiCI
PAM PAM
I Electr.<.aiiiMechaniC
PA M f'!A.M
Figure 21. 3. lnvestigation of primary functional need
Track ~1ce
Prov1de a means of c~ting car mot10n 1nto

form to operate coupter m..chanism
Reason-P.A.M. ofoperat1ng coupler mechan1sm
on mov1 ng car
Time - Every t1me '" e1ther d~rKt,on
Ptace - BPtwe-en or at S1de a
ralls;
'"front or behmd cage or t1ppter.
MPans - Convert lorward or backward mot1on
of car 1nto o"" acceptabte by th"
coupler ~~c•
I
M•thod
I I
Magn.,tic Hydrau\1c
Prov1dl> a m"ans of operat1ng PA.M. of us1ng mov1ng car
coupter devlc" by pl'rmanent signal to create or r@'lt-ase
magnetic field or a f1eld s.t ~ssurP l'nergy to operate
up by car" mot1on. coupt•r dO"Vice.
E•t•rnal power needl>d
EIKtncat Mechan1cal Pn~umat1c
Provlde a means of g1v1ng PA M.of making mot1on of P. A.M. of chang1ng mov1ng

an 1mpulse to couple-r dev1ce car acceptable by coupter car signal to ~leoase of
activall>d by car•s motion. devic" to opl'rate or tngger pre-ssure energy to activate
t he lock mechan1sm. COlJpt~tr mechanism.
Sens1ttVf' to damage, fire nsk Air suppt y needed: precisi on
Figure 21.4. lnvestigation of primary functional need of a track device

190 THE DESIGN ME THOD
return the spindie B and finger A to their hililies for combinations of minor items became
original positions. In doing so the lock is re- apparent but no revolutionary changes suggest-
tumed to the reset position by the lock spring ed themsel ves immediately.
as the cam rotates to its original position. Then it was decided that a different kind
This mechanism could operate in both of charting would be necessary to produce the
directions of travel by the use of two lock required breakthrough. Each p art of the mechan-
stops, (one on each side of the track centre ism was listed as in Figure 21.8 and compared
Provideo a meoans of transfeorrtng r·~r malten to

lock mP.chantc;m
Reason-PA M ot retracttng and n.•sf"fttng lock
Ttmf' -WhPn tratn must be parteod or JOineod

Placeo -8ehmd coupler contour and wtlhtn
Plusfing heoad shape
Means -Postl•ve neutrahzmg and reset wllh
normal couplmg act•on; fad safe
I
M~thod
I
Magnette Hyd!aulrc
Provtde a means of recel'.,.rng PA M of changmg lrack
a stgnal to open and closeo devtce s1gnal •nto pressure
lock a nd prov tde a tonergy to neutraltze and
neutral•ze postlton reset coupler
I nverswn of coupler, leak s,

E•ternal power needed
•no rtoser vot r'
Mechan•cal Pneumalle
Prov•de a meoa ns of ustng a PA M of convertrng track PA M of changtng stgnal

current or stgnal from track dl"vtce s1gnal tnlo backward recetved rnto reiE>ase of
deovtce to provtde or forw~rd motton of lock tn pre~sure E>nergy to operate
neutrahztng dnd reseot sequence. coupler mechamsm
achon
Sensitive to damage, fm~· nsk Atr supply needf'd; prectston
Figure 21.5. lnvestigation of primary functional need of a coup/er device
line) and two ramps (one in front and one with every other part in turn to ascertain all
behind the lock set stops.) However, it would possibilities of elimination, combination,
be essential that the leading ramp be lowered transference or standardization.
into the track so that the lock set stops operat- This chart emphasized what had become
ing first. Some method of sensing by the motion apparent in the first, namely that certain areas
of the cars would be required to present the of the mechanism were likely to contain major
ramps in the correct sequence. redundancies, although the precise location of
The next step in the process was to list these redundancies was not clear. It was con-
the parts of the mechanism and against these cluded that the key to the solution probably lay
record the functions which each part carried in a fundamental dimension not exposed on
out (Figure 21. 7). From this it became evident either of these charts, possibly the time
that certain parts of the tentative proposal dimension. (Fundamental dimensions available
were of minor importance while others were comprise need, reason, time, place, means
carrying out several essential functions. Possi- and method.)
I For example, the plunger H and connecting pin

! J could be seen to move without doing any
\
useful work. Similarly the twist of the spindie
D•rechon of F intended to be resisted by the torque spring
motron
I, was seen to be an entirely unnecessary
action that could quite easily be replaced by
exploiting the reaction of the lock spring E on
®
Reduced
sectron
\
the cam C via the lock D. At last there was a
really positive indication of both the location
and the nature of the redundancies only dimly
sensed before. How to remove the redundancies
I(
whilst maintaining essential functions became
8 the major problem.
It is sometimes useful to be able to stand
away from a piece of work in order to survey
ramp
X lncompat1ble
Figure 21.6. Design sketch of a partial 1 Eilmonate
solution to the primary functional need
A functional process chart (Figure 21. 9)

was produced to reveal the sequence of opera-
tion of the various parts of the coupler mechan-
ism during the neutralizing and reset operation.
This chart to some extent indicates areas or
items having unnecessary or non-useful motion.
Plunger
Neutral•z•ng
collar
F"unctronat means provlded
Sp1n<"lle peg
---+-~
Coupler Connect1ng p1n
P-'M. of retractmg and holdmg lock
non-manually '" s~nct
Lock Operatrng ftnger· Figure 21.8. Study of the possibility of

PA.M. of Couplong ht!ads PA.M of Convertrng linear to
Transm•tttng forcK angular matten eliminating, combining or transferring
Retracting l<>ck
Locating spnng Neutralrzrng coHar items or whole sections of the design
PA M of Holdong spindie
Lock spnng Gurdrng itseU
PA M of Re-turmng lock Releasmg plunger
P.,m•tt ing coupling
Keeprng tock coupted Torque spring: the whole vista and possibly by doing so to
PA.M of Returntng Iever
Cam
PA.M of Retractmg lock
Convertrng honzontal
to. vertrcal mohon
to see a new line of action. To this end, after
Conv.,-ting motion
Cottar gu•de
careful consideration of the first three charts
Spondle·
PA.M of Transf•rnng motron
PA M. of Guodmg collar
in relation to one another, a further chart was
Guid•ng Plunger Connectmg pm
Neutral!zrng PA.M of Connect•ng plunger prepared to indicate in broad outline the opera-
Applyong Ioad to to coltar
torque sprrng tion of the coupler and track device, and the
Locating cam and fing.r Ptunger peg
Access for plunger ~g PA M of Locating plunger compatibility of various partial solutions. It
must be emphasized at this point that no de-
Figure 21.7. Functional means provided by the sign sketches had yet been produced for a
partial design solution coupler or a track device, all work having
Neutralizing
Leckstop
exerts
torce
Lock retractron INeut~alrzrng

WWW'\f!WWW
, I I :Ramp'
WW'\f
I : I
: : 1exerts 1 I I 1
r 1 ·force I I
I I I
I I
I
I
Figure 21. 9. Functional process chart to show sequence of operation of design
been clone with the aid of a single 'Heath immediate thought triggered by this discovery
Robinson' sketch and four extremely simple was that the mechanism should be able to lock
charts. and unlock itself by the track device if only
The fourth chart (Figure 21.10) helperl to the operating finger A and plunger H could be
establish that the track device could be single combined. The redundancies that had been
action (one device to lock and one to unlock) troublesome would then have largely dis-
or could be double acting (thereby locking and appeared.
unlocking successively on the same device). This last thought was the breakthrough
The coupler 'branch' of the chart revealed an which had been looked for. It was confirmed
important but hitherto unnoticed 'obvious' point when the functional process chart (Figure
that while a coupler was locked by operating a 21.9) was again studied. The closer scrutiny
finger, an entirely different part of the mechan- of this chart from the viewpoint of the new
ism was operated to reset the coupler. The knowledge, revealed that items A to E were
concemed with retraction of the lock and items

F to L with the retaining of the lock in its
retracted position. Using a functional process
chart as shown in Figure 21.11, it was there-
fore argued that if the first five items were Lockstop
exe-rts
made self-locking the mechanism and head force
construction would be of a ·far simpler form.

Only at this instant was it realized where
the design thought-process had been most
clouded and wrongly influenced by previous
designs. In all previous coupler designs the
retraction operation had beEm locked by an ~$'0.
additional mechanism be it a simple pin or a ~· .,q
S' !!" cf:.. <t
plunger device. Illustrations of these can be '?q c,~ " -.,0
seen in Figure 21.2; these show the various ~v\7W

stages in the development of methods of neu- Lockstop
eoxeorts
I
I
I
1
I
I
I
I
I I I I
tralizing previous couplers, i.e. temporarily force I I
I
holding the lock in the retracted position. In I
I
Figure 21.2 (a) the original coupler had a lock- otat 1

I
1
tomin.l~ ~
ing pin to drop into a recess in the lock. In
Figure 21.2 (b) the operating Iever is held in r.Advance
. \L~f'-"!'!1!,
releasect
\_,____/ "-.___./
the retracted position by a Iever mounted on
the end of the mine car. The integral coupler Figure 21.11. Functional processchart
illustrated in Figure 21.2 (c) uses a spring for new design
loaded plunger to drop in front of a projection
on the lock after either of the two side Ievers whether or not they are subsequently rejected.
has been pulled to retract the lock. In the usual design process, decisions are
This realization brings out the true advan- made and ideas rejected at every stage, and in
tage of charting one's thoughts and ideas many cases good designs are lost because one
cannot call back ideas and decisions which
have been made days or weeks earlier, and
committed to the mind's waste paper basket.
In this case it was possible to refer to the
previous charts to confirm the decision that
the coupler mechanism could be self-locking
for simplicity. Without such charting one might
weil have been satisfied with operaUng only a
halfway solution, namely a combination of
plunger and finger instead of the complete
elimination of fifty per cent of the entire
mechanism. The ways and means were listed
of providing a self-locking mechanism to be
operated by the track device and employing
the energy of the moving car. Thoughts on the
Figure 21.10. Operational variation of requirements of the haulage circuit of the mine
coupling and uncoupling device produced the idea that the track device should
be capable of operating the coupler to lock or

unlock when the coupler passed over it in
either direction, pennitting standard single
symmetrical track devices.
Reverting to the self-locking coupler
mechanism, consideration was given to various
non-return devices resulting in a short list of
four devices:
(1) Toggle- Would have to provide l'i4" in.

movement at the lock and would occupy more
space thah was available within the coupler
head and shank. The toggle would also re-
quire three pivots, possibly machined.
(2) Cam - Would be compact, may not re-
quire machining, and would fit in coupler head
space. Could be 90 ° or 180 ° operation cam,
i.e. double or single operation of lift per rev- Figure 21.12. Proposed new design
olution. Could be affected by wear. of coupler mechanism
(3) Worm and Wheel- Would be precise but subject to two periods per cycle of accelera-
possibly expensive, and could be affected by tion and deceleration. This was undesirable
dirty working conditions. from the point of view of increased ~ear. The
(4) Friction - The use of friction, which is 90° cam mechanism provided a smooth operation
variable and unpredictable, may not give a and was therefore adopted for further develop-
satisfactory solution. Additional effort would ment. This stage of the process is illustrated
be needed to overcome friction before it could in Figures 21.13 and 21.14. Figure 21.14(b)
be used to lock the mechanism. shows the much simpler replaceable stop
of the 90 ° cam mechanism on the track device.
From this short list, the cam appeared to
Motion
satisfy the greatest nurober of operational
factors in the life cycle.
Considerable thought was now focussed
on this possible solution and resulted in the
proposed design for a 90 ° operation cam,
although the possibility of using a 180 ° cam (a)
had not yet been entirely eliminated. This
was the stage reached at the end of the course
at Engineers' House when the proposed new
mechanism was sketched out as in Figure
-
Motion
Track
21.12.
Following theoretical studies of cam pro-
file and motions, wooden models of the two
cam operations were made and tried out for
(b)
functional effectiveness. It was found that
the 180 ° cam mechanism had a dwell in m.id- Figure 21.13. Study of motion of operating
motion and therefore the mechanism would be cams: (a) 180 ° cam, (b) 90 ° cam
(a) (b)
Figure 21 .14. Wooden models to check operation of 90° and 180° cam mechanisms:
(a) underside of coup/er mechanisms; (b) inside of coup/er mechanisms
To ensure the correct form had been used cranked lever during the neutralizing and
for the various parts of the mechanism a proto- reset conditions can be seen in Figure 21.18,
type coupler was produced from an existing which is a sectional plan view of two mated
coupler, which was split so that the side couplers.
could be removed to observe the motion of the To assist in the dimensional checkings of
mechanism and to facilitate modification of the the production drawings for functional effect-
parts. This is illustrated in Figures 21.15, iveness, the chart in Figure 21.19 was
21.16 and 21 .17. produced. This ensured that all the various
The relative positions of the cam and interrelationships in the mechanism were
Figure 21.16. Prototype mechanical coup/er

with side removed to facilitate inspection and
Figure 21.15. Prototype mechanical coup/er modification of the intemal parts
196 THE DESIGN METROD
investigated for clearance, effect of production

tolerances, etc. In addition the possible effect Reset Neutralized
of a change in any one item on other associated
items was immediately apparent. As each
relationship was investigated the appropriate
code number was entered into the corresponding
square, and where there was no direct or
indirect relationship the square was blanked
off. (The chart shown in Figure 21.19 was
also used to check a hand operated integral
coupler design which was developed to replace Figure 21.18. Sectional view of two
mechanical couplers in neutralized and
an existing integral design.) reset condi tion
shown in Figure 21.20. Each scheme was

checked against the operational factor to
determine (i) the effect of the scheme upon the
operation, and (ii) the effect of the operation
on the scheme. This brought into the open
many points which had previously been hidden.
The chart shows only a few of the many
questions asked and answered during the
investigation of several different schemes.
It will be seen that the idea of symboliz-
ing one 's thoughts in the form of simple charts
1 Nominal clearances
2 Elfect of production
toleranc..s
3 Fleid trials to dMI!Nlline
shaP" and size
S Standard flts
Figure 21.17. Bottom view of prototype

mechanical coup/er with a display of the
parts of the mechanism
This completed the design for the coupler

device.
Attention was next focused on the instal-
lation and operational aspects of the intro-
duction of the mechanical coupler to a mine
circuit. This was a complex problern in that
the factors influencing the siting of the track
devices varied from mine to mine. To facilitate Figure 21.19. Chart to check functional
this part of the investigation, several schemes effectiveness of the mechanical coup/er
were drawn up and each scheme was charted as and track device
to aid the particular stage of working was ex- correct height by a spring contained in a tele-
ploited right up to the completion of the design scopic casing. After operating the coupler
study. mechanism the track device is depressed
In Figure 21.21 the various factors affect- below the level of the car underframe by the
ing each scheme have been given a rating, and car axles and retums to normal height when
the total sum of these has been used to pro- the car has passed forward. This is required
duce a histogram. From this the overall value because the coupler operating gear is located
Eff•ct of sch•m• I Op~rat1on lactors Eff•ct of o~ration

on op•ra tion on sch•m• I
( 1) LocomotJve- moveome-nt
( il Exot and •ntry to ClfCUil
0 k Loco shed ok
o.k. Workshops ok
Use track occup.Pd by 1•1 train Mater•al storeos Sw•tches 1 and,,stop • r•tract1ng
0 k. Ma.n-r•d•ng teorm1nus Reotract switch j; restle t•rminus
o.k. M1neo term1nus W1tl m•ch"t coupler go 1nto m1ne?
(2) Input ol cars to clfcuit

0 k. ( il Workshops ,) ,(ii),(,i,) must cross J
Rf'POSition, re-tay track ( il) Matproal storPS (,), ( 11 ;), musl b• 32 car train
o.k. (liil Man-nd1ng teorminus
? (IV) Mme- terminus ?
(3) 9t"moval of cars from circu•t

None Points pOSSible Or1ft,loco shed, stor..s. terminus
" Points Hser,taal tor correct op. o k eoxc•pt cem•nt store- exit
(4) o.raolmpnt r•ducong traon t•ngth

Not l•ss than 32 c.ors haul•d Ett•ct on sch•m•s Train will n<?t span crHP«S
?
Jncr.as. lo'"gth ot traons?
Figure 21:20. lnvestigation of field application to determine siting of

track devices and method of Operation
of each scheme may be compared with above the axle level as a means of protection
each other and with the ideal. From Figure against accidental operation by debris on the
21.21 (a) it can be seen that scheme V in this track.
instance was preferred although it was still Throughout the project an attempt was
not ideal. Reference to Figure 21.21 (b) en- marle to consciously steer towards good design,
ables one to evaluate which part of the scheme i. e. the optimum solution to the sum of the true
could be improved or further refined. needs of the particular set of circumstances.
Figure 21.22 is of a mechanical coupler Besides the very thorough consideration of
installation showing the relationship of the functional needs, far more detailed attention
track device to the coupler which is mounted was given to customer and company needs than
on the mine car underframe. In this installation in normal design practice, partly again by the
the track device is held horizontal at the use of appropriate charting.
Schemel
Ideal
Rating of L..ocomot1ve movement ..

Input of cars
each scheme
Removal of cars
compared w1th Deraliments
ideal solut1on Failures
Load er
Tippler
Creeper
Cost of equipment
Cost of installat1on
Cost of ma1ntenance··
Personnet number
Personnet quality
(a) (b)
Figure 21.21 . (a) Histogram showing the relative values of various schemes for the application
of rnechanical couplers; (b) details of one scheme forming part of the histogram
The new coupler is now being produced the essential nature of good design, and that
and shows every sign of being very successful. this study con tinues in parallel with the work
It has not been possible in this brief on the actual design project should not, how-
account to describe adequately how the ab- ever, be overlooked in evaluating the results
stract concepts of the nature of the design of this project. Similarly those who wish to
aims, processes and elements assisted either improve their own design approach by private
the choice of charts used to symbolize thoughts, study, may well benefit from paying as much
or the methods of analysis and synthesis. The attention to the various characteristics of the
fact that much of the fundamental design course as to those of the formal 'techniques'
method is concerned with a study in depth of revealed.
Possible Misunderstandings of the Nature and

Practical Value of Abstract Concepts used in
Fundamental Design Method
The process involved in designing is
basically one of applying knowledge to
visualize .and judge. The ability to produce
designs which satisfy an amalgam of needs in
an optimum way is as much related to the
methods of applying knowledge itself. Unless
and until knowledge is organized araund the
basic needs of a design situation, and the
designer becomes at least partially aware of
the fundamental characteristics which all
design problems and processes have in com-
mon, he cannot begin to apply knowledge in a
Figure 21.22. View of track device really effective way.
installation and a mechanical Willison Fundamental design method is not primar-
coup/er ily a simple problem-so! ving procedure. It is
more a way of working which involves a high awareness of his working method or abstract
degree of awareness of one's own mental concepts of the type mentioned, and its aims
moves, coupled with an awareness of abstract are more limited. While P ABLA is intended to
concepts of the essential nature of good design achieve a degree of improvement throughout an
and the design decision process. These con- organization by simple prescribed procedures
cepts which a designer can leam to refer to related to certain areas of design work, funda-
rapidly at will and use as 'absolute' standards mental design method relies on a heightened
are constructed initially by a process of deep awareness on the part of the individual, as
reflection around partial concepts proposed on well as his continual use of abstract concepts,
the course. Before they can become meaningful to determine the most ~ppropriate actions
and therefore truly useful, the designer con. throughout the entire design process. In other
cerned must be prepared to attempt to express words it requires a kind of thinking which is
his own experiences in an abstract form, and significantly different from traditional design
this is by no means easy. thought.
Improved methods which will arise from The concepts used in fundamental design
such examination will necessarily differ in method may be said to be derived from digested
many respects depending upon the persons and experience, whereas the designer's intuitive
design project involved. The PABLA system mental moves will stem from his undigested
used at the United Kingdom Atomic Energy experience. Ar first the concepts will almost
Authority, Aldermaston, has received wide certainly suggest changes in approach that
publicity and can easily be misread as a uni- 'feel' incorrect. Although the designer has to
versal system for designing, which it was be very careful in the leaming stages to check
certainly not intended to be. PABLA began as that the concepts are not indeed in error, he
a project on the fundamental design method must be prepared to use them until they are
course and has been developed in close proved wrong, and to go on correcting and
collaboration with. Engineers' House, who refining them until they can be relied on
help to train all UKAEA designers in its use. completely. Hence theory and practice continu-
Although derived from fundamental design ally interact on each other and become a most
method, P ABLA does not necessarily require advantageaus living partnership.
the individual to possess a high degree of
Chapter 22
DESIGN REALIZATION
F. L. Ellis
I ntroducti on factories. In the case of such assernblies

The aim of this chapter is to discover there may be a clear interest in avoiding the
generalizations of value conceming the process direct problems of fabrication proper of parts
of design realization so that the tasks of or components. In spite of this the overall
designers may be better comprehended and design has to be accomplished in such a way
studied. In approaching this it is only too as to facilitate assembly, although correct
obvious that experience in restricted areas of function is the final aim.
specialization will tend to be drawn upon. In addition to integration of the design
Clearly, builders and ci vil engineers will idea in terms of materials, fabrication and
tend to specialize in one way; production assembly, thought has to be devoted to the
engineers will specialize in another way. effect of these operations on the function of
Design realization is concemed with the the product in order to provide the necessary
practical tasks of converting a design idea function. This again is increasingly concemed
into a working entity. In some form or another with system behaviour, with reliability.
this working entity is expressed in materials. To obtain realization in a material embodi-
The materials are chosen to provide the basis ment, certain instructions must be passed.
of the functions that the product has to carry The provision of the technical information
out. These materials have to be provided with for these instructions is commonly a principal
a practical shape and this is done during the function of the drawing office, although,
fabrication process. The selection of materials depending upon the industry and the company,
is therefore not solely a question of product the drawing office may have other functions
function but of the ability to provide the shape earlier in the design sequence.
required. Clearly, fabrication method may The relationship between materials, fabri-
influence design. Consider, for example, the cation, and assembly is now considered.
difference in approach needed for tradi tional
building and industrialized building. Materials and Systems
An important feature of modern products Various ways may be used for categorizing
is that many of them are assernblies of parts materials but, in the present context, it is
of various kinds. This is implicit in most convenient to use an approach based on system
engineering system products. The procedure function. By system function is understood
for assembly may constitute an imposing part the main action of the system: the kind of
of the total realization operation. This is transformation brought about by the system
particularly true for the two extreme kinds of whereby the input emerges as an output.
system commonly manufactured: flow products It has been said that systems may handle
such as consumer durables, e.g. motor cars; energy, mechanical motion, chemical opera-
complex one-off products, such as oil refineries, tions, or information. In practice, systems
nuclear and other power plants, chemical tend to be complex and usually hybrid. For
201
example, an electrical generating station may expressed in the form of a merit rating, as the
contain a chemical system (the combustion ratio of the number of points awarded to the
sub-system), a thermodynamic system (the sub- number obtainable.
system in which heat is converted into rotary A small difference in a !arge total would
motion), and an electromechanical system not be considered significant and a further
(the sub-system in which rotary motion is evaluation, allowing for the relative importance
converted into electrical energy). There are of the original conditions, would normally
also other parts of the major system, each enable a final choice to be made. If they still
with its immediate Containment and supports, prove to be of almost equal merit it would be
and the totality of these is interlinked and advisable to proceed with detailed design of
contained as necessary. both until it became clear that one of them
As a mechanical engineer my interest was markedly superior.
tends to lie in the field of kinematic systems, When the kinematic scheme has been
but I have interests in thermodynamic systems, decided upon, the designer is ready to begin
particularly refrigerators, and in the Contain- the task of accurately determining the material,
ment of pressure systems, as involved in the manufacturing processes, and the dimen-
steam-raising or in many types of chemical sions (with tolerances) for each component.
process. This is achieved only after a considerable
In machine design the materials of con- amount of trial and error work, since the
struction play a great part directly in the components are interdependent and the deci-
system transformation process, as well as sions made with regard to one will affect
contributing to enclosure and support. In many those made concerning another. (This highly
prime movers there are temperature and detailed interdependence, demanding many
corrosion problems to be solved, as well as small decisions, may well occur in other
the pressure situation to be dealt with. Aircraft branches of technology, e.g. electronics.)
and space propulsion units bring further Characteristically a few decisions are tenta-
problems: the need for lightness, and possibly tively made, the effect of these on others is
the handling of exotic fuels. checked, and necessary adjustments are made.
This procedure is repeated until the design
Mach ine Design eventually reaches completion.
In the case of machine design a methodical The design of the contact members
approach will include: (a) searching for an provides suitable starting points, since in
existing solution to a similar problem; (b) look- many instances they control the design of the
ing at books which include examples, preferably other components. Contact members include
in c o 11 e c t i o n s, of already developed gears, cams, chain and belt drives, brakes,
mechanisms; ( c) attempting to obtain solutions clutches, and springs. The shafts for supporting
by combining known constructional elements. the contact members may then be designed.
It is likely that a number of solutions This is frequently done on the basis of shaft
will emerge, some of which may be eliminated deflections, since these will be limited by the
immediately. If a number appear to have equal engagement requirements of the contact mem-
merit a methodical evaluation process is bers (in the case of gears, for example, it is
advisable. A method which enables a choice desirable to have good load distribution across
to be made quickly is given by MATOUSEK the face width), or by clearance requirements
(1963). This consists of allocating points for between moving members (e.g. between the
all the conditions which the problern requires rotors and stators of electrical machines and
to be fulfilled. The best solution is the one turbines). The shafts cannot be designed
that has the largest total number of points; without, at the same time, decisions being
its closeness to an ideal solution may be made regarding the bearings and the bearing
lubrication system. When these have been properties of the material, and the design
taken into account the proposed shaft must be details involved in the process. The shape of a
checked for strength, including the possi- component is often determined by the properties
bility of failure due to fatigue. of the material and the manufacturing process
None of the decisions in this sequence (TWEEDDALE, 1962). For example, ordinary
can be taken without knowledge of, or assump- grade cast iron has a poor tensile strength
tions concernirtg, the materials to be used. but high· compressive strength. If it is used
The frame of the machine and the links for a member which is subject to bending,
which transmit the forces and motions from the such as a cantilever, this would be designed
with a cross-section which utilizes the high
contad members to the other elements may
then b~ designed. The frame has to be strong compressive strength but keeps the tensile
stress to a low value. The cross-sectional
enough to transmit the reaction forces to the
shape would also be influenced by the necessity
foundation and rigid enough to provide reason-
to provide draught angles so that the casting
able alignment between the moving parts. The
could easily be removed from the mould.
absorption of vibration by damping may be
Generous fillets or radii would also be required.
necessary. Again, in each of these points,
materials must be considered. Where no specific requirements for
materials are stated in the problern given to
In addition to obtaining a solution which
will not fail while providing the service the designer, he has the respönsibility of
listing all the factors involved and making
expected from the machine, the designer tries
to obtain the best solution. This is usually choices based on them. The most suitable
materials may be chosen by using an evaluation
achieved by making decisions which result in
plan in conjunction with an extensive list of
a compromise between a number of conflicting
factors (Matousek) or by employing a system
but desirable attributes.
involving optical coincidence feature cards
Successful designs in different fields of (SELWYN, 1965).
activity are brought about by the application
In some cases the number of materials
of different design philosophies, i.e. attitudes
that will satisfy the function requirements with
which arise from the relative importance
reasonable economy will be small. For example,
attached to the factors which must be con-
there are relatively few material combinations
sidered. Thus the attitude of the aircraft
suitable for worm gearing. Where Ioads and
designerwill differ from that of the car designer
ruhhing velocities are high it is usual to
and from that of the ship designer (EDER and
employ centrifugally cast phosphor bronze for
GOSLING, 1965; FAIR ES, 1955). These
attitudes are reflected not only in the basic the wheel and nickel-chromium alloy or nickel-
chromium-molybdenum alloy for the worm. As
design constraints but also in the approach
to the use of materials, to fabrication methods, another example, the tubes in heat exchangers
and even to assembly methods. are normally required to have good thermal
behaviour. If mild corrosion is present as well
the choice of material may be restricted to
Choice of Materials and Fabrication Methods aluminium, brass of some kind, or copper. In
in Design the case of highly loaded spur, helical or
When a machine component is being bevel gears, cams and followers, and sprockets
designed, the processes by which the part is and chains, the Controlling factor is likely
to be made must at the sametime be considered. to be surface failure and this would point to
The designer has to be familiar with such the use of case-hardened steel.
interdependent factors as the suitability of With electrical functions the choice is
materials for various fabrication processes, usually heavily restricted, and in complex
the effect of the processes on the subsequent systems it is made more difficult. For instance,
the squirrel-cage rotors of fractional horse- previously feasible. Designers who act system-
power cartridge-type motors, used in hermeti- atically in a conservative way have a responsi-
cally sealed refrigerator systems containing bility to bring forward and exploit new materials
dichlorofluoromethane as refrigerant, may have in the same methodical manner.
either copper or aluminium conductors in the
slots; the former leads to the use of brazed-on Fabrication, Assembly and Design
end rings, and the latter to the use of end The selection of materials and associated
rings cast integral with the conductors. processes is, of course, influenced by the
Aluminium would need to be free from mag- quantity of components or nurober of systems
nesium to avoid possible chemical reactions to be produced. It is unlikely for example that
invol ving the refrigerant. pattems would be made for the production of,
In some cases the selection of a material say, four cast iron gears if, as is likely,
is rendered difficult on account of the large suitable forged steel blanks could be easily
nurober of possibilities. These are often cases obtained. There are, however, cases where
where the forming, welding, or machining pattems are made for one or two castings, a
processes demand sizes which are in excess typical one being the drawing dies for producing
of those required for adequate strength and panels of motor car bodies. In this case fabri-
stiffness, either for function or other purpose. cation does not provide a suitable product as
If the material value is high and there are pattem. The panels of mass-produced motor
other advantages to be gained, there will be car borlies are butt-welded in a flash-welding
much pressure to change manufacturing machine. Where quantities are small, instead
processes. This may be seen in the recent of this machine and the design associated
drive to microminiaturization. with its use, it is common to use a spot-
There are certain rigidities involved in welded lap joint (joggle joint) which is filled
materials selection. Apart from the general with solder and then finished off to present a
pressures in industry to provide a safe solution smooth outer surface.
the designer has usually had special experience The shape of a steel cabinet is often
with some kinds of material. He has designed box-like because the quantity to be made does
components from them and the results have not warrant the production of the special tools
been satisfactory. He is therefore naturally needed to make a more interesting shape.
disposed towards using them whenever he can. Not only does the designer need to
Further, he will be influenced by the avail- consider the cost of material and the cost of
ability of fabrication facilities, particularly fabrication and assembly (machines, conveyor
in his own company. He will normally be systems, and labour) but he has also to
reluctant to incorporate new materials into consider the cost of design. The amount of
his designs until they have been satisfactorily time that may be spent to detail work on a
demonstrated to be superior to the ones which design, broadly depends upon the type of
he normally uses, or to those found doing project, i. e. whether it is a simple one-off
similar work on other equipment. He will tend job, a complex one-off job, a batch production,
to stick to established practice and resist or a flow production. This applies to the
change whenever in doubt. calculations connected with the design of the
Fortunately there is a development product in addition to the information issued
potentiality in most designs. Somewhere it is with regard to its manufacture.
usually possible to try out new materials in a It is in this area that advantage may be
practical way on current products. This gives drawn from the use of standard components,
the opportunity for trying new materials under by reliance upon drawing office practice, and
known conditions of operation before exploiting by the exploitation of a close knowledge of
their characteristics to provide designs not particular workshop practice.
Flow Production Flow production tends to modify the

In many cases flow production involves fabrication processes and auxiliary operations.
large capital investment in connection with Thus, instead of brazing a number of bushes
the manufacture of a low or moderate cost into a casing separately with a hand-torch
consumer product. In order to protect the followed by cleaning by shot-blast, a preferred
investment and make profits it is important method would be to employ furnace-brazing
that the product should give reliable service under a suitable protective atmosphere to
within at least the warranty period and be avoid subsequent cleaning. This would not
available at a competitive price. The effective only permit the use of less skill but would
achievement of these ends requires painstaking provide a more consistent Ievel of quality.
attention to detail design in connection with Similar results follow from the use of auto-
function, appearance, transportation and matic arc-welding, replacing manual methods.
production. In addition there must be adequate Special auxiliary plant may have to be
quality control during production. developed to keep the flow moving. In the
Quantities are so large that the product manufacture of domestic refrigerators, for
designer, in order to achieve an optimum example, equipment is needed for dehydrating
design, does not concern hirnself too much the units to avoid decomposition of refrigerant,
about incorporating standard components of for de-aeration of the oil to be used as lubri-
the type which are usually available at cant, and for charging oil and refrigerant into
engineering stockists. He will, however, think the systems.
in terms of standard pieces such as bolts, Prototype models should be made to match
nuts, rivets, washers and split pins, and of as closely as possible to the product that will
standard intermediate materials such as wire, eventually be mass produced, because the
tube, bar, sheet and plate, which are them- time available between the making of the
selves the results of large-batch or flow first products from tools and the beginning of
production. mass production is frequently inadequate for
One may view flow production as a rigorous testing. This means that the parts for
particular method of solving an assembly prototypes should be made from drawings which
problem. This method is then supported by closely resemble those to ·be eventually issued
some method of obtaining the necessary parts for tooling up the product. Even when this is
and components. These may be purchased from clone, there is difficulty in obtaining proto-
an outside supplier or made directly. In turn, types which correspond in quality to the
whether in an outside factory or at the main products made in bulk since the former tend
works, the parts may themsel ves be produced to be made to high standards by skilled
by some selection of flow processes. In order craftsmen. There would seem to be some
to secure fabrication that is highly economic scope for laying down specifications which
it is sometimes necessary to instaU special give prototypes with behaviour more akin to
purpose machine-tools, often with automatic the products of the flow line.
transfer equipment between them. There is a
trend now towards multiple-operation machine- Batch Production
tools in order to diminish transfer requirements. This classification covers products
With increasing emphasis upon economy ranging from large system units such as
of operation in flow processes attention is aeroplanes, to simple machine components
being directed to methods of economizing in for which there are unlikely to be sufficient
the use of raw material stock. This may Iead orders to warrant continuous production.
to rethinking about the method of fabrication In a large number of cases batch production
employed, and thence reflection on the original is characterized by the use of standard
component design. machine-tools and general purpose processing
machinery. Different parts tend to be made Simple One-off Jobs

by retooling automatic lathes, changing the Press-tools, jigs and fixtures are typical
cutters and fixtures on milling machines, examples of one-off jobs. They are usually
changing the drills and drill-jigs on drilling made to detailed drawings and exploit stocked
machines, changing the dies in presses, etc. materials and standard parts (some of which
This type of operation can be particularly may be internal company products from batch
difficult. Not only are many detail drawings production). Parts might include drill bushes,
needed, but the complexities of planning to screws, dowels, clamps, and springs. Stocked
get the best utilization of machine-tools and materials would include a selection of com-
delivery of the components and parts in the mercially available rods, bars and plates.
right sequence, at the right time and to the These parts and products are then used for
standard specified, produce many problems of making items for batch or flow production.
an intractable nature. Such one-off items may be made in a
Since much manufacture falls within this simple jobhing shop or, in the case of tooling
class there is a sound case for exploring and items, in the tool room. Where possible,
exploiting the various possibilities for re- advantage is taken of a knowledge of shop
ducing difficulties. The use of standard practice in the specific case. Typically tool
materials, standard parts and standard compo- designers may be expected to have served an
nents available to industry at large, helps apprenticeship in the tool room and in other
with this kind of work. Thus, when designing aspects of tool-making.
compressors for use on commercial refrigeration Tooling items tend to be designed on the
plants, it might be desirable to try to use as basis of experience, the greatest requirement
many components as possible from models apart from low cost being sufficient rigidity.
mass-produced for domestic purposes. The A job may consist of little more than the
limiting design then might well be a three- manufacture of a frame to support a number of
cylinder radial compressor (two cylinders for pieces of commercially available equipment.
some applications) incorporating the pistons, Although the job itself may seem simple the
grudgeon pins, connecting rods, valve plates, system design may be complex.
cylinder heads and gaskets designed for the Many one-off jobs come from people who
domestic model. It may also be possible to specialize in a particular kind of fabrication
use other common items such as terminals process. The items manufactured are often to
for connection to the leads of the electric be used in more complex assernblies brought
motors. In a !arger unit it may be possible to together by other organizations. For example,
use connecting rods mass-produced for motor heavy work such as the fabrication of pressure
car engines. vessels, or the production of items in special
materials such as stainless steel, aluminium,
It is in this class of manufacture that the
or certain plastics, tends to be the field of a
maximum advantage may come from numerical
restricted number of manufacturers who keep
control of machine-tools. There is the possi-
their position by the possession of special
bility of saving in setting time and skill, and
skill or skill with necessary equipment.
from the re-use of prepared programmes.
In general, there is a line of development
Anyone who has spent time in the works in which such specialist manufacturers attempt
is only too conscious of the need to find some to take advantage of their skill and to widen
way of dealing with the problern of planning their outlets by introducing the design of
and progressing. Research in this area should products which particularly require their
be strengthened with a view to reducing the facilities. Pressure vessel manufacturers, for
large commitment of human resources and the instance, move towards the manufacture of
frustration so typical of this kind of work. steam-generating equipment, heat exchangers,
and certain items of chemical plant. In their Conclusions

normal one-off trade they are required to assist The tasks of design realization lend
in the optimum design of their customers' themselves to analysis and the development
items. As an almost natural consequence the of suitable systematic approaches. In some
build-up in design skill leads to self-generated regions of the subject this has already been
design. Here is a case of fabrication causing done, largely in an informal way; but the
the development of design. possibility exists of presenting the field in
its totality, showing the range of approaches
Complex One-off Jobs in several directions. This will enable a more
These are essentially system projects rational development of the subject and will
with high capital cost of the type mentioned reveal new possibilities in the course of the
at the beginning of this chapter. Although the enterprise, stimulating some people by the
more expensive heavy plants are largely mere mention of techniques employed in a
constructed from one-off or similar purpose- different class of business.
designed units there is always the pressure to As with all systematic approaches the
use standard units. This pressure increases designer will still have to deploy his funda-
as the capital cost of the system is forced mental skills. The number of cases where
down or the number of units in a system design may be made entirely by system are
increases. certainly few, but they are useful and their
It is usually an obligation that the system number will increase (British Standards 537
works efficiently and promptly, with the and 436; BROWN, 1953). An example is given
minimum of modifications after erection. For in the Appendix.
this reason the design is usually worked out
in great detail, although, for practical and APPENDIX: Spur Gear Design
other reasons, this detail may arise from A 5 h.p. electric motor, with a shaft speed
previous experience in one way or another. of 1,450 rev/min at rated power, is to drive a
In operations of this type which involve shaft at approximately 290 rev/min. Design a
the building up of substantial teams for the pair of spur gears wi th a momen tary overload
execution of each stage of the project, a high Capacity of three times the continuous rating
degree of organization is required in order to of the motor, and a centre distance of approxi-
make the best use of skills and the time mately 3·5 in.
available. From companies employed in this
kind of operation comes much of the pressure For the required momentary overload
for standardization in design procedure, for capacity:
the acceptance of standards and specifications Strength rating
= 1·5 (see Clause 67 in B. S. 436)
of wide industrial currency, and for the appli- Wear rating
cation of Computer methods of design. Pinion speed n = 1,450 rev/min
Characteristic of this class of engineering
Wheel speed N 290 rev/min
is the exploitation of critical path methods in
the solution of the assembly problem. Although Gear ratio R =!!..
N
by definition each assembly is different,
=5
investigation is beginning to show rational
patterns with the possibility of building up Also,
Computer programmes for each occasion from
existing modules. In such a procedure the
design of the management technique begins and
xcw
to show a parallel to the design method for the a ---
product itself. w- xbw
where X.cp and Xcw are the speed factors for The nurober of p1mon teeth required will
wear for the pinion and wheel, respectively; therefore have to be read from Forged Steels
Xbp and X bw are the speed factors for stren gth Chart No. 2 (shown in Figure 22.2).
for the pinion and wheel, respectively. From
the Forged Steels Chart No. 1 (shown in Figure
ß : St~gth rat•ng Xr1• R]" X
«w
22.1), a;p = 1•0 and a w = 1·0. Therefore 6
2 W•ar rat1ng [ 2C
C 'Centre distance (in )
R 'Gear ratio
I 'Number of teeth on plnion
a
~ =1•0
ap
0
1
=Strength rating x
Wear rating
[1 +~
2C
l
j
0'2 x
ap
where C is the centre distance. Therefore
~ = 1•5 [ tJ 0 2
• X 1•0
= 1•454
Using Figure 22.1, the point for R = 5 and = o,

1·454 lies to the left of the awlap = 1·0 line.
14
.;:. _ Strength rat•ng x [-1 +R_]o 2
Ot- Wear ra!1nq J"[ XUp
C 'Centre d1stance (m) 13
R:: Gear rat10
I 'Number of teeth
on ptnton
0~,~-3~~5~~7~~9~~,~,~-,~3~-,~s~-,~7~-,~9~
5 Gear ratio,R
~~-= 100 50 Figure 22.2. Forged Steel Chart No. 2

\ P~nton or wheel speed. rev/mm
4
[1
Using Figure 22.2 the point for R = 5 and
'' ~ _ Strength rating + R~ o·2
a
' ''
02-
Wear rating
X --
2C
X
w
\
I = 1·454
I
I I leads to the adoption of 32 pinion teeth (t = 32).
1 I / 1•20
1-4 54
----"1 I':----1_'
I
I
-r--------~----~~
II
Nurober of wheel teeth T = Rt
:~ 1•30 =5 X 32
I'--- 1•40
1~1·60 1•50 = 160
T +t
Diametral pitch P = ~
Gear rat1o. R 160 + 32
2 X 3•5
Figure 22.1. Forged Steel Chart No. 1 =27·43
Adopting a standard diametral pitch of 26, 585 X po·e

gives: Scw =X. x z
cw
160 + 32 585 X 13•55
Centre distance C
2 X 26 0•334 X 3•3
= 3•7 in. = 7,200
for wheels
Pinion torque rating required S _ 877 X p
bw- .Xb w Yw
f or wear -_ 63,000 X 5 X
1, 450
877 X 26
= 217 Lb. in. 0•34 X 0•705
for strength = 1•5 x217 = 96,000
= 326 Lb. in.
Pitch circle radius of pinion = .jp There are no forged steels with permissible
factors as high as these. Therefore case-
32. hardened steels must be tried.
=52 ln.
Circular pitch p =; Using Case-hardened steels Chart No. 1

(Figure 22.3), the point for R = 5 and = 1•454 o,
Maximum face width of pinion = 5p lies to the right of the aw I ap = 1•0 line and
571
=p- R _ Str•ngth r•tmg [1 • 02 R]
"'1- W..r r•ltnQ X [ 1C X er"
571 C = Centre distance (in)
=26 R = Gear ratio
I =Num!»r of teeth on pinion
= 0·604 in. 22
Tangential force per inch of facewidth 20

217 x 52 _........~= 1·00 on R= 1 line
for wear
32 X 0-604
=585
Lb./in.
326 X 52
for strength = 32 x 0 •604 I= 10
' ' ... ... I= 12
:= 877 Lb./in. '
'
'
Basic surface and bending stress factors .ö ' I
I =14
required I
I
I
585 X po·e I
I =20
xcpxz
585 X 13•55 I= 30
0•24 X 3•3 1=40
1:50
= 10,000
for pinion 02
877 xP
}(bp X yp 0
1 5 9 11 13 15 17 19
_ 877 X 26 Gur ratio,R
- 0•246 X 0•795
= 116,500 Pigure 22.3. Case-hardened Steels Chart No. 1
210 THE DESIGN METHO.D
leads to the selection of t = 10. Therefore 1S6 X po-a

= Xcp x Z
T = s X 10 1S6 X S•S23
=so =
0•24 X 1•0S
p =so + 10
for pinion
= 3,420
2 X 3•S 234 X p
60 Xbp x Yp
=7
234 X 8•466S
=8·S71 0•246 X 0•S6
= 14,400
Adopting a standard diametral pitch of 8·466S,
i.e. a module of 3mm (p = 0•3711 in.), the
centre distance is 3·S4 in. approx. 1S6 X po·s
10
scw =·')( CW X z
Pitch circle radius of pinion
2 X 8•466S 1S6 X S•S23
1 . 0·334 X 1-üS
= 1·6933 In. = 2,460
for wheel
Sb _ 234 X p
2x2 - Xbw X y W
Maximum face width of pinion = 1 -6 933 W
234 X 8•466S
= 2·36 in. 0•34 X 0•S0
= 11,6SO
Tangential force per inch of facewidth
Carbon steel to B.S. 970 EN 328 would be
217 X 1•6933 suitable in conjunction with a reduction in
for wear
2·36 facewidth
= 1S6 Lb./in.
2·36 x ~:~~~ = 0 ·877 in.
for strength =326 2X •36
1•6933
or 14,400 0 8S .
2 •36 X 40 ,000 = • ln.
234 Lb./in.
=
Basic surface and bending stress factors
required whichever is the larger, i.e. 0·877 in.
Chapter 23
SELECTION OF MATERIALS
A. L. Davies
lntroduction In this particular case, the saucepan can still

It is probably true to say that a designer be used with a broken handle, but not with
always asks too much of the materials with a broken ri vet.
which he has to work. In general, this is a A more complicated situation arises where
satisfactory state of affairs since it imparts many parts are used, such as in a motor car.
to the materials scientist an ever present A motor car, from a materials standpoint,
incentive to develop improved materials, and consists of two kinds of components:
also it ensures that the designer has the best
material for his purposes available at the time, (1) Those which are moving and are nor-
within the confines of the economics of the mally replaced with ease.
exercise. (2) Those that remain static, many of
Many commercial operations are scaled which are not replaced with ease.
to using the material of minimum cost for
doing the job. In some cases a nurober of Thus an automobile may have two or three new
materials meet this requirement and then other engine units during its life, but once the body-
factors of material utilization come into play, work starts to corrode the end is usually near.
e.g. local supplier, long association of company In composite articles, particularly those
with supplier, ease of fabrication. containing a !arge nurober of components, it
For many manufacturers, the ideal material is extremely difficult to build the final product
is that which gives excellent service for a using materials that deteriorate all at the
given period, and then fails quickly. A material same speed. Certain obvious discrepancies
which gives trouble requiring servicing or can be eliminated; for example, no one would
replacement, but then lasts for a considerable dream of making a car body out of stainless
period is less desirable, as users normally steel, as the body would still be intact when
prefer an article that is replaced in its entirety tlte engine design had been superseded. On
at the end of its life, rather than a material the other hand, exhaust systems on modern
which needs constant attention, but lasts for cars could, with advantage, be manufactured
a Ionger time. from a more heat resistant material than they
Any article whichincludes parts assembled are at the present time.
and manufactured from a variety of materials At the other end of the scale, articles
is only as useful as its weakest part. Taking are made in which the cost of materials is
as a simple case, a saucepan made of deep not important, but the performance of materials
drawn aluminium and the handle made from in service is paramount. Rocket motors must
bakelite: in all probability the handle will be be developed to give a certain thrust, and it
the first part to fail or, if the handle has been is usually essential that this is obtained.
given extra strength, then the rivets holding Very often this may result in the use of exotic
it to the saucepan will probably fail first. materials, with a high cost.
211
Between the above two situations, there In general, the value of Young's modulus
lies a wide field where although cost is not falls with increase in the melting point of the
all important, it must be kept within reasonable meta!, but magnetic and other phase changes
bounds. The development of aircraft for civil sometimes introduce anomalies, for instance
use would normally come into this category, in cobalt.
whereas the development of aircraft for military Other mechanical properties of importance
use might weil come into the 'money no object' are the plastic and forming properlies of the
class. material. No structures remain completely
rigid under applied Ioads and some degree
Criteria of Selection of plasticity is normally required in all metallic
The following criteria would seem to be components. However, a lamp-post would
important in selecting a material for a specific require less ductility than a bridge member.
purpose. It is unlikely that all the conditions Thus it is possible to produce a lamp-post
mentioned would apply to a particular part; out of an easily castable material, e.g. pig
indeed, in most cases only one or two will be iron or concrete, whereas the bridge member
relevant. Ilowever, it is probably true that the would have to be produced by some wrought
more demanding the requirements of the process.
designer, the greater the number of criteria The ability of a material to deform is of
which will be brought in to play. vital importance in the selection of materials.
At the present time, British industry still Returning to the example of a saucepan, it
uses rather ad hoc methods of materials can be seen that this article may be produced
selection, but great strides have been made by one of two methods, by casting, or deep
in this field in other countries, particularly drawing from sheet. At one time all saucepans
in the USA. were made of cast iron and were produced in
the foundry by melting and sand casting. With
M echanical Properfies cheap labour, the method was reasonably
A great number of materials need to with- economic but produced an article that corroded
stand a certain stress. A simple example readily, had little resistance to impact, and
would be structural steel. In many cases the was heavy for housewives to handle. At the
strength-weight ratio is important. Although present time much kitchen ware is produced
aircraft could otherwise be made from a very from aluminium which is rolled into a sheet
!arge number of materials, in fact the weight and then deep drawn to produce a saucepan
of the final machine is so important that the which shows great improvement over its cast
number of materials is limited and at the iron equivalent. It should be mentioned that
present time most airframes are made from recently more expensive saucepans produced
aluminium and its alloys. In previous years, from stainless steel and/or copper have been
of course, wood was used a great deal. With produced. This, of course, shows an improve-
the advent of very fast aircraft the heat ment in appearance, but heat transfer may be
generated by the passage of the aircraft through questionable.
the air will become an important factor and it The temperature at which a material is
will no Ionger be possible to use aluminium. being used in service is also of great impor-
The likelihood is that stainless steel will tance. The tensile strength usually increases
have to be substituted even though this will with lower testing temperatures but the
have an unfavourable effect on the strength- ductility, which is related to the toughness
weight ratio. of a material, falls. In body centred cubic
Sometimes the elastic behaviour of a metals such as iron, a brittle transition occurs
meta! must be considered. In such cases in which there is a marked change in ductility
Young's modulus of elasticity is of importance. over a small temperature range. This
phenomenon is not experienced with face are sometimes of importance. Probably the
centred cubic metals, but sometimes occurs most well-known situation where some of these
with metals with a hexagonal lattice structure. properlies are of importance is in the electrical
The brittle transition is of particular impor- transmission field. For example, much work is
tance in materials used in cryogenic appli- being carried out at the present time to produce
cations. dilute alloys of copper with such elements as
A very relevant factor is the maintenance zirconium and chromium in order to produce
of mechanical properlies under changing materials that have almost the same electrical
conditions: The most important of these factors conductivity as pure copper, allied with an
are creep and fatigue. Creep is the very slow improved tensile strength.
extension of a material under the continued
application of a static stress. Normally it is 'Resistance' Properties
of importance only at elevated temperatures, In the metallurgical field, many materials
but a few metals (e.g. lead) have a measurable have to possess an inherent resistance to
creep rate at ambient temperatures. Since certain external agents. The most well-known
creep takes place at a very slow rate, infor- of these are resistances to chemical attack
mation regarding the phenomenon takes a long (corrosion), abrasion and temperature. The
time to obtain. However, a considerable amount last named of these has already been dealt
of information has been obtained on the creep with in the paragraph on creep.
properlies of materials used in turbines, for
example nickel base alloys containing chromium Resistance to Chemical Attack
and cobalt. In its broadest application this includes
Fatigue, the failure of a material at a what is loosely called corrosion resistance.
stress weil below the ultimate tensile stress, Many metals corrode under atmospheric
is of importance in any situation where a conditions, the most well-known being iron
material is subjected to an altemating stress. and many of its alloys. Because ferrous
Most ferrous materials exhibit an endurance materials are used for a very high proportion
limit, the maximum stress a material is able of articles produced at the present time,
to withstand no matter how many reversals of corrosion problems are focused on these
stress take place, but non-ferrous materials materials. In steels, corrosion is mitigated
usually have no such limit, and theoretically by one of two methods (i) external protection,
would fail at any stress provided the number and (ii) inherent protection.
of reversals of stress was large enough. The External protection includes the appli-
presence of a corrosive environment greatly cation of some corrosion resistant material to
reduces the resistance of a material to fatigue the surface of the steel. These materials may
failure. In all cases the endurance limit is be another meta! (e.g. nickel), a polymer or
drastically lowered, or destroyed completely. even paint. They may be applied to the steel
In general, fatigue is cured more by good in a variety of ways ranging from spraying
design, for instance the avoidance of sharp to electroplating.
re-entrant an g 1 es, than by materials Inherent protection is afforded by alloying
development. steel with another meta! or metals to produce
a finished product that in itself is corrosion
Physical Properties resistant. The most well-known examples of
Physical properlies such as specific these types of materials are the stainless
gravity have already been considered on the irons and steels which are being used in
preceding section. However, other physical increasing quantities in Great Britain.
properlies such as electrical and thermal Where the corrosion resistance is very
conductivity, emissivity, or thermal diffusivity severe, for example in the chemical industry,
special materials have to be used and alloys plastics and the replacement of metallic
of nicke!, chromium and cobalt are popular. materials by polymers has by no means finished.
These materials are extremely costly and are
only used where absolutely essential. Materials Selection as a Management Technique
It is reasonably common in the USA,
Special Properfies though much less common in Britain, to follow
Occasionally a material must have a a philosophy of product evolution, and to
certain special property in order to perform break up materials selection into a number of
the function for which it has been designated. stages. One example of this technique has
In many cases, cost is of secondary importance. been developed by TRAINER and GLASGOW
An example of these special properties are (1%5). Materials selection may be broken
certain electronic configurations which render down into four stages:
the material useful in atomic reactors. Niobium
and beryllium were initially produced in quan- (1) Concept formulation,
tity because they had properlies which atomic (2) Feasibility,
scientists wished to utilize. Germanium was (3) Development (final and production
produced in quantity because it found use in design),
semiconductors. (4) Production (or manufacturing).
In all cases, materials with special
properties produce heavy development costs Concept Formulation
and these are not always recovered. It is First, a designer develops a generalized
doubtful whether any money has been made out system (system image) to represent the general
of beryllium. Conversely, transistors are now functioning of a device to satisfy a defined,
paying their way. specific need. Next, he visualizes different
ways in which the individual functions of the
Cost, Availability and Ease of Fabrication system image can be performed.
Once it has been decided that a certain When the tentative concept is analysed in
number of materials are suitable for a particular terms of applied stresses, fabrication, etc.,
application (and in most cases there are a certain classes of materials can be eliminated.
number of materials which would be suitable) In essence, what the designer really wants to
then the three factors of cost, availability know is whether the concept is seriously
and ease of fabrication become of considerable limited because needed materials and processes
importance. are not available.
It is manifest that some of the newer A materials limitation may suggest the
materials have made inroads into the fields of modification of a tentative concept. On the
materials used twenty years ago. An excellent other hand, it may not be a case of changing
example is the replacement of metallic parts the material, but of changing the technology of
by plastic materials. Before the second world a material that would otherwise be suitable.
war, a telephone hand set was constructed The use of the so-called 'Nimonic alloys' for
almost entirely of metal; at the present time turbine blades in jet engines is a good example.
only a very small portion of the instrument These materials possess the necessary heat
is made of metallic material. Plastic materials resisting properties, but are difficult to
are easy to produce, exhibit a low specific fabricate. As a result of research to improve
gravity and are non-conducting. Their dis- the process technology of these materials,
advantages lie in the fact that they are more their use is now no problem. Other new tech-
brittle than metals, have a lower tensile niques such as diffusion and friction bonding,
strength and have less heat resistance. How- vacuum Casting, high energy joining and
ever, the improvement in the properlies of explosive forming are still only in their infancy
and are further examples of improved process times it may be advisable to change materials
techniques making available hitherto unusable and manufacturing processes as sales rise,
materials to the design engineer. though it is desirable that these contingencies
should be foreseen and appropriate arrange-
Feasibility ments made at the outset.
When a prom1smg concept has been
selected its feasibility must be determined. Production
Stress analysis work is important at this stage, Materials changes or selections made in
and the designer must be sure that no factor the production stage often have a profound
is overlooked which might place the whole effect on the success (especially financial)
project in jeopardy. After the first detailed of the product.
examination to determine whether all critical One technique that is often applied is
stresses are reasonable, a second examination known as value analysis, and although the
is needed to determine what types of materials value engineer does not confine hirnself
are candidates for the different components. entirely to the production stage, the technique
Stress analysis will have to be combined with applied to production analysis has produced
such factors as corrosion resistance, electrical astounding economies. Other significant
conductivity, and methods of manufacture as aspects of operation which influence materials
appropriate. When items are critical ( e.g. modification are tooling, purchasing, and the
bearings, electrical contacts, components in pilot run. Although work is clone in each of
vibration) the more accurately the conditions these areas during feasibility and development
are known the more specific the materials stages, the greater contributions on the part of
specialist can be. Tests and experiments will experts in tooling, purchasing, and production
have to be carried out. It is important at this usually occur when a product is actually going
point that the designer works in close col- into production.
laboration with a materials scientist. Simply stated, value analysis is an
organized method of finding the least expensive
Development way to make a product without compromising
Ideally, the development stage is only quality or reliability. It is a systematic, step
concerned with how design functions can best by step, method designed to eliminate hap-
be performed, and not whether they can be hazard cost reduction approaches, and to
performed. allow no cost reduction alternative to escape
Often the primary objective of the develop- without examination.
ment stage is to maintain a certain level of A typical approach, popular with large
product performance, while reducing cost to companies, is to have a policy committee and
the minimum. Sometimes a particular aspect an operations or project committee. The policy
of performance such as reliability is as committee is composed of manage m e n t
important as cost. The probable materials to personnel from research, engineering, materials,
be used have been determined in the feasibility and processing departments, whereas the
stage, and the development stage will include operations committee is made up from engineers
a detailed analysis of possible methods of and scientists from the same departments who
manufacturing components from the probable actually do the experimental work. Users of
materials. value engineering have found it useful to
Other factors which must be considered divide the programme into three phases:
are existing facilities, for example types of ( a) definition of product function; (b) creation of
machines and sales predictions, which might alternatives; ( c) cost comparison and final
be needed because of the effect of production selection.
quantities on production economics. Some- The purpose of defining product functions
is to find out exactly what the part or product fabrication and manufacturing processes
is supposed to do and what is required of it. necessary to produce the product. Failure
A study of the alternatives available models are constructed in which the failures
consists of a search to find out what other are described in terms of the causes of failure,
designs materials and processes will do the e.g. functional failure of materials, failure
job. Alternatives are suggested regardless of resulting from inability to withstand a particular
feasibility, and anomalies are eliminated in environment, or failure caused by improper
the cost comparison stage. manufacture or processing. Once the failure
In the cost comparison phase each of the m e c h an i s m s have been determined and
alternatives is broken down into individual analysed, a materials selection procedure
items which are separately cost analysed and can be worked out.
then totalled. In a typical analysis the items In value analysis the primary goal is
are separated into two main groups (i) recur- minimum cost, whereas in failure analysis
ring costs, and (ii) non-recurring costs. the primary goal is reliability.
Recurring costs are unit costs that are incurred
each time the part is made, non-recurring costs Existing Products
are initial costs. Although materials selection is most
Although value analysis is probably most often thought of in the framework of new
important during the production stage, naturally product development, it is very important in
it operates best when integrated into the changes in existing products. There are many
whole process of materials selection. different reasons for making materials changes
in existing products. Some of the most important
are:
Failure Analysis
Another method of materials selection is (1) To solve materials processing problems
failure analysis. This method is based on arising in production.
predicting and anticipating all the ways in (2) To reduce production costs.
which a product can fail, and then selecting (3) To reduce basic materials costs.
materials so that failure does not occur. The (4) To make a functional change or improve-
method can also be extended to specific ment.
techniques, such as prevention of fatigue (5) To improve service performance, in-
failure (CLAUSER, F ABIAN and MOCK, 1965). cluding longer life and higher reliability.
Failure analysis can be used in either (6) To use a lower cost material.
new product development or in the review and (7) To take advantage of a new material
reappraisal of existing products. More problems or processing development.
will exist with new products since little or no
experience will have been acquired. The view Since the materials selection function cuts
has been put forward that two types of failure across many departments, and engages the
exist (i) immediate failure, and (ii) delayed attention of many different people, problems
failure. Immediate failures cover many obvious of liaison and control may arise, particularly
sources of breakdown, such as rupture or in a large organization. In smaller companies,
buckling from direct loading, chemical attack however, the company metallurgist or ehernist
or high temperature softening. Delayed types often advises on all materials problems, but
of failure are more difficult to anticipate and if in difficulties he can always bring in
include failure due to stress corrosion, creep, outside help.
fatigue, or solid state transformations.
The technique of failure analysis involves Future Developments
the study of the environments to which the A development which has already com-
materials will be subjected, together with the menced is the use of computers as an aid to
materials selection. Provided great care is and non-corrosive finishes. Plastics of much
taken in the preparation of information to be higher strength and heat resistance will be
fed into the computer, and it is kept up to developed and will have a much wider range
date, it should be possible to carry out simple of uses than at present, particularly in the
materials selection with speed and accuracy. light engineering industry and most probably
In complicated situations a computer will in the building industry. Forecasts, such as
be invaluable if it is able to give the sources those of BRECH (1964) provide pointers.
of information available regarding a particular
topic, so that the materials scientist can Faced with the ever increasing nurober of
come rapidly to a considered judgment. materials available, the materials scientist,
During the coming years, developments in order to achieve the best economic, techno-
in materials will include advances in alloy logical and ethical solution, will have to
theory to produce stronger and more resistant spend more time in the selection of materials.
alloys, the synthesis of what are at present In the coming years many companies that do
regarded as natural materials (e.g. wood, not concem themselves with materials selection
paper), and the greater use of materials but adopt the 'what was good enough for
mixtures. It would seem that there is room for Grandpa was good enough for me' attitude,
development with plastics/paper, plastics/ will find that they either have to employ a
fibre, fibre/metal and plastics/fibre/metal. qualified materials scientist or make wide
Metals will be mainly used in the alloy form use of consultants. If this is not clone it is
to give strength, together with toughness, likely that they will find themselves in serious
lightness and plasticity, with easy cleaning difficulties.
Chapter 24
RELIABILITY AND MAINTENANCE
C. T. Corney
lntroduction for reliability is of greater importance than the

There is frequently considerable confusion others - design, research and development
in peoples' minds on the difference in practice engineers, sales and field service people and
between Quality and Reliability; interpretation so on. Therefore inevitably any chapter on this
is not made easier by the fact that 'reliability'
is often tacked on, so to speak, to the word
'quality'. Let it be said at once that, without
effective quality control, no product will be
reliable; on the other hand, the finest quality
control, applied to a product whose design is
inadequate or which is wrongly applied under
service conditions, will not ensure reliability.
Thus the simplest definitions, on which this
chapter is based, are:
Reliability - The product should look right,

Ouality control
work right and should last for the specified
life.
Quality Control - A system to ensure that
(b)
the product, through all stages of manufacture,
conforms to the specification.
Thesedefinitions are illustrated in Figure 24.1.

Although it has already been said that the
two are in a sense complementary, until quite
recently the emphasis has always been on the Figure 24.1. Definitions of (a) reliability,
'conformance' side, i.e. conforming to the (b) quality control
specification and drawing, but with the
extension of management control techniques to subject tends to portray these spheres of
commercial and manufacturing activities. activity in greater detail than the role played
Special attention should be paid to the methods by people on the shop floor.
of achieving reliability, including a system of ReHability is so much a first essential in
quality control. British products that its existence has long
There is a slogan 'quality is everybody's been taken for granted. As a result any
business' and, broadly speaking, so is reli- deficiency in this attribute becomes even more
ability. But there are certain sections or conspicuous in the customer's eyes. The
departments in every firm whose responsibility reputation we enjoy today was built on good
219
design practice, and it is therefore appropriate which decides, as far as the user is concerned,
to evaluate the problern of achieving reli- whether the product is a good or a bad one.
ability by considering a number of techniques For many reasons, the reliability required
as practical design tools, which are capable of a product by the customer is increasing;
of general application. for example expenditure on capital equipment
The financial cost of unreliability is is making very big demands on everybody's
quite staggering, though very often this may pockets, and a purchaser naturally expects
not immediately affect the manufacturer. For that the higher his capital investment the
example, it has been estimated that roughly greater the performance and reliability of the
half the Air Vote (two hundred million pounds) product. In the commercial vehicle, equipment
is the annual cost to the RAF of unreliability. has had to meet increasingly severe service
On the civilian side, the cost to the Coal requirements. Advanced engine development
Board of maintenance on mining equipment has resulted in Ionger periods between engine
amounts to hundreds of thousands of pounds overhauls, higher annual mileages, higher
a year. speeds and heavier loads, frequently under
In a recent study in the author's own arduous climatic and operating conditions.
company, the rejection rate in steel castings But while these are factors, they are not the
was as high as 9 per cent; the major cause was main cause of unreliability. Complexity does
condition, with chemical content second. All cause unreliability, and modern equipment
this work was reject to us and scrap to the
supplier and merely represents money thrown 1928
down the drain.
Reliability and the Customer

In the vehicle industry, too, unreliability
is a form of waste that can be measured in With 238 critical parts each having
a reliability of 99 95 per cent, then
monetary terms; this waste not only costs overall vehicle reliability would be
the manufacturer money for repair of the 90 per cent ·
component, but in the case of a revenue-
eaming unit such as a bus, truck or train, it
also costs the operator money. A failure 1964
affects the attitude of those who hear of it as
weil as those who suffer it, and once a repu-
tation for reliability is lost it takes many
years to regain. The result is customer dis-
satisfaction with its ensuing effect on sales, With 696 critical parts each having
which will frequently be a much greater indirect a reliability of 99 95 per cent, then
financial loss to the supplier than the direct overall vehicle reliability would be
73 75 per cent
cost of maintenance or replacement.
A reliable product is one that gi ves Figure 24.2. Ef/ect of increase
customer satisfaction, but there is no absolute in complexity of trucks
standard of reliability; for example, a com-
ponent with a mean life of 100,000 miles on a is becoming increasingly complex. In simple
railcar can be regarded as unreliable because terms, there is more to go wrong and con-
of its 'short life', whereas a similar product sequently the reliability of each component
on a light truck might be perfectly adequate must be greater than its simple counterpart of
with an identical life. It is the ability to meet some years ago, even if the target is only to
a given duty for an acceptable length of time maintain the overall reliability of that time.
Figure 24.2 shows the growth in complexity ponent or complete unit. It is the manufacturer's
of a road haulage truck over the last thirty duty to sell goods with a known performance
years. In order to maintain overall reliability and reliability; this will obviously be related
at 90 per cent in 1964, the average reliability to cost, but will give the customer the infor-
of the critical parts had to be raised to 99•985 mation on which to decide whether to purchase
per cent. But even this is not good enough, or not.
for the overall reliability has to be improved If better results are to be achieved, they
in spite of the problern of added complexity. must first be designed and where possible,
fewer parts must be used in products; it is
clear that improved reliability is unlikely to be
Systems reliability (%) fortuitous (probably the reverse) and it is the
!ndividual No. of components design that sets a ceiling to what can be
components
reliabilityrtt~ 10 60 100 250 500 achieved.
99·99 99·9 99·45 99·1 97·5 95·2
The Reliabil ity Programme
99·9 99·0 94·23 90·5 77·5 60·6 The concept of reliability must be present
from the first stages of introduction of a
99·0 90·4 54·75 36·6 8·1 1-0"' product so that it is built in from the start; it
can only be added afterwards at the expense
98·0 81-7 29·77 13·3 o* o* of redesign. The importance of the part played
"'Approximate by the feedback of information from the user
Figure 24.3. System reliability as influenced

by component reliability
Figure 24.3 shows that with 98 per cent
component reliability (about right for 100 per
cent inspection) a system with sixty such
components has a reliability of 30 per cent,
and with 100 components 13 per cent. 250 such
components give approximately zero reliability.
Thus to get system or vehicle reliabilities
6f 90 per cent, component or product reli-
ability of 99·9 per cent is needed; or put Figure 24.4. Reliability programme
another way, product failure rates of less than showing the importance of feedback
of information from the user
one in a thousand.
Thus it can be clearly established that
industry has to face an unreliability problem. is of paramount importance and is clear from
It costs !arge sums of money and of the many the block diagram shown in Figure 24.4.
contributing factors the main one is complexity. It is therefore of value to go through these
To overcome unreliability the manufacturer headings in some detail. (A systematic
must really 'know' his product and be able to procedure developed in the USA is that of
prove his 'knowledge' correct. The specifi- MCCRORY, 1965.)
cation may be satisfactory in that the product
functions as required, but for how long will it Knowledge of the Market- What does the
function? A free maintenance guarantee to Customer Want?
replace worn or broken parts is no assurance This question seems so obvious that it is
to the customer who has no desire for 'down- sometimes not answered in sufficient detail.
time' on his purchase, whether it be a com- With increasing specialization of knowledge
Figure 24.5. Two views of a starter subjected to water from a Eire

hose, to simulate the effect of wheel splash at 90 m.p.h.
and experience, and with rapid changes taking designer can understand and work to. Thus.
place, it is necessary to take positive steps operating miles or years must be translated
to ensure that all who contribute to design into hours or number of operations; environ-
detail are aware of the operating conditions ment into temperature, humidity, vibration
of the final product. For example, it may not conditions, etc.
be realized that the force of a wheel splash
on a component of a vehicle on a motorway is Application of Existing Design Life Knowledge
almost the same as that from a fire hose (this While the provision of a specification is
can go up to 90 m.p.h.); consequently, instead a big step forward, there remains an even
of sitting on a chassis going down the Ml, one greater step which must frequently be taken
may conveniently observe the effects of a fire 'in the dark': designing, on paper, a product
hose (see Figure 24.5). The design specifi- with the gi ven reliability but not incorporating
cation provides considerable information, but such !arge safety factors that it becomes
arrangements should be made to obtain first- uneconomical. If the design is largely based
hand and up-to-date experience and quantitative on previous experience, then generally that
data on vibration, temperature and other factors experience can be applied to the new design
involved, including assessment of conditions with appropriate correction factors. If the
of use. design is novel then the designer must relate
such component information as he can obtain
Full Product Specification to the overall design reliability and rely to a
It is sometimes said, and is frequently greater extent on prototype testing.
true, that too little attention is paid to deciding Published data and previous experience
what reliability is required. Hence the first dictate life parameters in some instances,
step when designing a new product, is to e.g. bearing life-load relationships. Rating and
prepare a specification which, besides the pressures, mechanical and electrical wear
usual details of external dimensions, will properties and similar data are available for
include weight, performance, the range of electrical contacts, brushes, commutators,
ambient temperature which must be allowed for, slip rings, etc. Mechanical properties, fatigue
the degree of weather protection required, and lives, ultimate stress and other material
so on. It must also include in detail the life properties are normally available. All this
requirements under varying operating conditions; information related to product requirements
it is on this last part of the specification that must be immediately available to the designer
the actual reliability will greatly depend. in an acceptable form.
To design for reliability it is essential Published information on the life of
that all the design staff, not only the con- branded items may not be in the form required,
ceptive designer, are fully in the picture at this but one exception is bearing life on which
stage. This is where designing for reliability there is a great deal of knowledge; manu-
starts and where it is most important. facturers have for years expressed life in
terms of hours, loads and speeds. However,
Translation into Engineering Terms this data is commonly given for a 10 per cent
With a new design, the first conception cumulative failure rate, whereas it is nowadays
in specifying what is required - and this really necessary to operate to much lower component
means what the customer requires - is in failure rates than this, in order to achieve a
practice a sales market research function. product failure rate of a fraction of a per cent
Normally this conception is unusable as a within warranty and a satisfactory service
working document for a designer, and sales life afterwards.
engineering must translate the outline into Studies have been carried out in America
the definite specification terms which a (see Figure 24.6) to determine the source of
trouble in equipment supplied by government In the rig testing of prototypes the

contractors. While the percentage varied from engineering design is clearly making a major
product to product, studies on different kinds contribution, since at this stage production
of products showed similar distributions and quantities are not generally available to
these figures are probably representative in a assess manufacturing conformance - essen-
large number of cases. tially the province of the quality engineer.
Rig tests are thus primarily the responsibility
of design and development functions, and the
type of rig shown in Figures 24.7 and 24.8 to
simulate cycling conditions are in themselves
fairly complicated units.
Most new designs or conditions create
new problems, and experimental rig facilities
must be available to provide the answer. Such
testing facilities should come under the control
of a reliability engineer who, with the research
and development team, also has the chance to
assess the type of tests and equipment required
in future when the product is in manufacture.
Furthermore, a great deal of effort must be
put into problems of environmental testing,
Figure 24.6. Percentage distribution of to ensure that the tests reveal any real weak-
sources of trouble of a number of different nesses under the extreme conditions in which
products investigated in America the product may have to perform.
Figure 24.7. Duration test machines for simulating years of service

life of a product
away by shop floor concessions, or production

will be a nightmare of 100 per cent inspection,
excess scrap, low operator earnings and the
rest.
Some designers say that - like Oliver
Twist - they always ask for a little more than
they need, as a kind of safety factor; for
example a slightly tighter tolerance, a better
finish, a tighter performance specification.
This is said to cover the grey areas between
the black and white of accept or reject. Not
only is this attitude. an insult to inspection
mentality, but it is uneconomical in production,
tooling and inspection costs, and cuts away
!arge chunks of allowable tolerance which
make statistical quality control possible: and
statistical quality control is the handmeiden
of reliability. This attitude also reduces the
drawing or specification from an inviolate
deposition, to the status of a mere guide,
subject to interpretation by personnel in-
competent to do so, or the subject of inter-
departmental wrangles.
It is essential that the drawing (and this
includes customer drawings) represent the
design requirements and no more.
Figure 24.8. Control board for proving tests on
CAV product
Product variab1lity
Control of Manufacturing Techniques and
Quality of Bought-out Material
The capability with which the specifi- .r 1
cation can be met in the production phase is a
design characteristic. A satisfactory design
is one which, besides meeting the performance
specification, can also be made economically
on the equipment envisaged by the designer.
This forms a vital link in the chain of functions
contributing to product reliability.
It is sometimes argued that design should
not be restricted by the limitations of existing
processes and that it is a production engineer-
ing responsibility to provide the techniques to
achieve design requirements. While there is
some justification for this view, the fact
remains that if the process is incapable of
economically maintaining the parameters Figure 24.9. Opening out of dimensional
required, either these are going to be whittled tolerance
In stipulating the design requirement for accelerated tests. Testing to failure should
reliable design it is necessary to be aware of always be the goal but, when time does not
and to consider the variation of the process permit, accurate measurement should be made
concemed (see Figure 24.9). There is no of wear, etc., which, when related to the length
virtue in setting a tolerance T, about the of the test run, will give an indication of
mean for an operation whose known performance product reliability. In a design devised to meet
leaves no room for other Variations, even
assuming that its variability is within the
total tolerance.
Statistics are an essential tool in design
when consideration has to be given to cumu-
lative component tolerances. lt is surprising
llow frequently tolerances can be opened out
to .T 2 , far more than was originally supposed,
resulting in reduced scrap and yet still allow-
ing reliable assembly. Knowledge of process
capability assessment techniques and the
interpretion of their results into detailed
drawing dimensions should be an integral
part of the training of drawing office personnel.
It is essential that the design requirements
are clearly laid down in the drawing, and that
it is understood by everyone that there should
be no departure from them unless authorized by
senior management.
Full Design and Test Specification

This follows in engineering terms from Figure 24.10. Synthetic reliability testing
the previous steps and provides the designer
and the quality engineer, at the earliest a new requirement, the tests can be made
possible stages, with the basic information more effective by synthetic reliahility testing
to design both the product and the tests and in which the reliability of detail components
test equipment to be used in design proving is evaluated before they are assembled into
and production quality control. The latter are the whole complex. Figure 24.10 shows the
easily relegated to too late in the development specific example of oil seals being subjected
stage and can become as much a design to a stringent performance test before assembly
problern as the product itself. After designing into fuel injection pumps.
a prototype to the required specification (at In general, however, only some aspects
this stage the project has only just begun), of a design can . undergo synthetic reliability
a development and proving programme to verify testing, and it may be uncertain whether all
that the product performs under the full range the relevant factors have been recognized and
of specified operating conditions must be taken into account. In such cases it is prudent,
started. if not clearly essential, to test a proposed
In a great number of products, such as design under actual or closely simulated
domestic goods, the service conditions are service conditions. Even very simple com-
completely known and understood, enabling ponents can exhibit service problems which
design parameters to be specified with cer- can be minimized if they are adequately tested
tainty, or at least proven by accurate before being released for production and sale.
This takes time and slows down delivery of matical short cuts which save the labour while
samples to customers, but it is time well keeping the principle the same.
spent. The chances of getting the different
The properties of products made to a possible numbers of spades are shown in
given design vary, often to an extent that Figure 24.11, from which it can be seen that,
matters. This is particularly true of .Product although everything from 0 to 13 is possible,
reliability. In any test, therefore, the test values outside the range 1 to 6 are unlikely;
batch must be regarded at best as a random they would in fact occur only once in some 44
sample of the production which a line is deals. If on joining a new school, one was
capable of producing. A random sample is, dealt a hand containing (say) seven or more
by definition, one arising by chance. Fortu- spades, one would not be justified in shooting
nately the effects of chance are calculable - the dealer, but it might be wise to take
but they are considerable. precau tions.
The problern of judging a new design by Sample test results must be judged in
testing a sample batch is similar to that of exactly the same way: namely, in terms of the
judging the honesty of a new card school from probability of the actual sample test result
the first deal. How can it be done? Nothing arising by chance from a product with a given
is certain - except that a hand cannot contain average reliability; decisions to accept or
more than thirteen or less than zero spades - reject a product must be taken on the basis
but the probabilities of the different possible of reasonable probability rather than certainty
hands can be calculated. For instance, the about its reliability. This applies to any form
chance of a hand containing four spades can of sample testing, whether it is measurement
be calculated quite straightforwardly by work- of a continuously variable parameter such as
ing out the total number of different combi- wear, assessment of a product life curve, or
nations of cards that can be dealt as a hand, simply assessment of the proportion failing at
and finding the proportion of these that contain some particular life. In all cases statistical
exactly four spades. The chance of being techniques by which the effect of chance may
dealt exactly four spades on any given deal is be allowed for are available and should be
simply this proportion. Writing out all the two used; in most cases the results of the neces-
million (or so) possible combination of cards sary calculation are already available in
would be very difficult, but there are mathe- tabulated form. These techniques are not
limited to ·'normal' distributions and to 'random'
97 · 7 ~r cont cl duiS produood failure, although it may sometimes seem so
- betwHn 1and 6 spados
( i.P. 43 deats tn 44) from the literature: in particular, the Weibull
dis~ribution allows a wide range of life curve
shapes to be dealt with readily. Unless the
designer has a comprehensi ve knowledge of
statistics he needs to seek guidance on the
interpretation of test results from someone
; ~ ; who has. The point to be emphasized is that
8. 3.
~ ~
~
S> such interpretation must be done on a statis-
~ ~ ~ tical basis.
"' "'
l l Scale of Testing
11 12 13
Allowance for chance means, in effect,
that a margin of error must be allowed on the
Figure 24.11. Chances of getting various actual test result. This margin is surprisingly
numbers of spades in one hand ~~ cards large, or - to put it another way - the scale of
testing necessary to make it reasonably narrow would be inadequate), and test batches of some
can be surprisingly large. The worst eireum- thousands of samples may be necessary in
s tance is when the test is limited to deter- order to prove that this order of reliability
mining the proportion of sample failing within has been attained.
a given test period, as frequently happens when Such high reliabilities might seem to be
progress towards a failure is not measurable special requirements but, as was seen in
and the time available for testing is limited. Figure 24.3, to achieve 90 per cent reliability
SufJpose 300 samples are tested on this from a system of 100 components the average
basis. The best possible result is that no component reliability must be 99·9 per cent.
failures occur, and the most stringent accep- When one considers the reliability required of
tance criterion would be to accept the product the individual components in a complex piece
only if no failures occur. However, it by no of machinery - particularly components with a
means follows that the failure rate would then nurober of alternative modes of failure -
by zero per cent. It can be shown (by the same improvements from quite modest overall reli-
kind of argument as for the pack of cards) that ability levels are, in practice, often found to
if a product with a failure rate of 1 per cent is hinge on improvements of individual fault
presented for test, there will be no failure rates of the order of 0·1 per cent or less.
among 300 samples once in twenty such tests: The multiplicity of items of electrical and
i.e. such a test allows a 1 in 20, or 5 per cent diese! fuel injection equipment supplied for a
chance of accepting a product with a failure modern motor coach is shown in Figure l4.12.
rate as high as 1 per cent. If failure rates of Many of these items are themselves relatively
the order of 0 ·1 per cent are hoped for this complex assemblies, with many components
chance will be unacceptable (i. e. 300 samples and many different modes of failure. The major
Fiuorescent tubes Horns

I A1r corculating !ans Loudspeakers (3)
11
\ \
I
/
I
E.P unit
Alternator /speed gener ator lube
Battery cut-oll swo
Stop
Figure 24.12. The multiplicity of equipment on a modern motor coach

Produclien rate= 20,000per month Ttme reqUtred:28to38 months

Normal fault rate= 0-T3per cent H"lwarranty Samptes requtred = 15,000
S\<[19~ Foeld IE'S! - - - -Servoce tn al - - - - - - - - Fuil rPIE'ase
100'1.------------------,------
Proporlton 1
of out~t 1
modified 0 ...__---- --=======::::::::::L_ ___;._ _ _~-
oMost probablettme
: cJ f ,nal resull
Stop d ilution · further : ftnal result w•li oo
'>amptes wou ld be too : obla<nE'd SOTIE't <me
:ate to spet>C result dunng thos perood
'
:Star! d1luti<Y" at l!n 25.
•assumtng no foeld te;st Iaiiures
Ttme scate 0::-------1:-':2,.---------=2'-:-
4-----~36"..---
Months
Figure 24.13. Predictable course of service trial of a modification,

assuming it not to affect the Iailure rate
modes of failure found generally show rates of shown in Figure 24.13 was adopted, the steps
the order of 0 ·1 per cent during the product in the procedure being:
warranty period of twelve months. In circum-
stances such as this, the direct proof of the (1) All known internal tests and safeguards
full required reliability is impossible ahead relating to this fault were applied to the
of design release unless testing is quick and modified design. The modification caused no
cheap, or there is the time and the opportunity change to the results, otherwise it would have
to conduct a substantial service trial. been abandoned there and then.
(2) One hundred samples were field tested
for one year, under controlled conditions. No
A Service Trial Procedure faults resulted, indicating a less than 5 per
A service trial is particularly possible cent chance that the true fault rate was as
and desirable in the case of design modifi- high as 3 per cent.
cation to an established product. The procedure (3) The modification was introduced as a
and scale of test required is illustrated by the 1 in 25 dilution of normal production. At this
following example, taken from actual experience. dilution, if the failure rate of the modified
A particular product suffered a low but product is less than 3 per cent then the overall
persistent fault rate of 0 ·13 per cent, but the failure rate would be raised by less than
fault was of such critical nature that no sub- 0·12 per cent (1 in 25, or 4/100ths of 3 per
stantial increase in its rate could be tolerated. cent) i.e. it would be less than doubled, which
A proposed modification, desirable for other was the condition stipulated.
than reliability reasons, was expected not to ( 4) The service failures of modified units
affect this particular fault rate but, because were compared to those of the normal units,
the mechanism of failure was not completely and the strength of the dilution was increased
understood, this could not be guaranteed. It since the results showed this to be safe.
was therefore decided that the modification (5) The trialwas continued until the service
must be put to statistical test, and could be results satisfied the condition stipulated for
accepted only if the results allowed a less acceptance. If they had shown that the modifi-
than 5 per cent chance that it would as much cation almost certainly worsens the fault rate
as double the failure rate. The test procedure would have been rejected.
The time-scale and numbers involved in this likely to Iead to product unreliabili ty; by
trial are both too large to be contemplated concentrating effort in these areas, audit
except in special circumstances. The product activities can be carried out economically
was in production at the rate of 20,000 per and with maximum effectiveness.
month; for smaller outputs the trial would Accelerated tests on complete units will
take longer. frequently reveal premature failures. These
What are the implications of inability to are of value if they indicate a trend in a failure
prove conclusively on a statistical basis? Past pattern which will be repeated under actual
experience shows that it is painfully easy to service conditions; otherwise such failures may
find that expected reliability improvements be red-herrings. If service conditions can be
make the situation worse, due to unforeseen nearly reproduced for monitaring tests, then
secondary effects which were just not observ- accelerated testing can give valuable infor-
able on the quantities originally tested. In so mation to design staff ahead of first field
far as full proof of reliability is not possible results.
in advance of design release, the other steps
in the development process assume added
Customer Guidance on Application
importance. In particular, all reasonable
and Main tenance
possible steps must be taken to foresee and
This is essentially a sales engineering
guard against failure through care in design -
activity governed by the product specification.
particularly where a substantial safety margin
In the case of vehicle accessories with a
can be allowed at no added cost.
great di versity of application, the necessity
In general, customers have some under-
of a sales engineering manual for each product
standing of one's business and will not expect,
has become evident, and this makes a suitable
or pay for, the practically impossible; but,
check-list for design approval.
equally they will not forgive failure to take
The problern of specifying the right product
all reasonable precautions.
for the right job is necessarily beset by
The Follow-up of Initial Service Results continual competitive pressure, without which
This is a most rewarding area of reli- everybody would become inefficient, but
ability activity, since isolated failures which in itself can sometimes Iead to equip-
examined by design staff are extremely ment being used above its reliable rating.
meaningful in interpretation of possible Adequate product operation specifications,
trends, and in instituting immediate remedial which clearly show the ratings and conditions
action. There may be a tendency to 'explain for which each product has been designed,
away' single failures and the statistical are essential in the reliability field. These
significance that these represent is not widely will enable the user to recognize the correct
appreciated. It is therefore of great importance equipment to fulfil whatever conditions he
that all early failures should be examined in requires. It is worth bearing in mind that, by
detail and properly explained. paying a little extra at the beginning for a
more conservati vely rated product, considerable
operating economies, through greater reli-
Monitaring Production Quality and ability, can be achieved.
Service Reliability Guidance from the designer must include
Audit schemes for production 'conformance' the location of the equipment. In the worst
may be carried out from sample unit strip-down. cases, this can be so bad that it is almost
These should again be statistically based to impossible to carry out normal maintenance so
satisfy acceptable risk factors. Design has a it just does not get clone. On the other hand,
responsibiJity, based on user experience, in the product is sometimes located for ease of
guiding quality control on those aspects most maintenance but in such an exposed position
(see Figure 24.14) that water, rnud and grit, products are serviced by dealers, special
etc. rapidly cause the product to deteriorate, arrangernents to secure information should be
leading to eventual failure. The problern of made. The author's cornpany represents an
intermediate case: service and repair of its
products are carried out by its depots and
agents, and a twelve rnonths' warranty clairn
scheme ensures a return flow of data on
failures during the first year of service. The
inforrnation on the clairns forrns provides a
diagnosis of the nature and cause of service
troubles and, if it is accurate and is carefully
analysed, provides an essential pointer to
the rnatters that should be investigated; it
also of course, indicates the reliability during
the warranty period. Exarnination of out-of-
warranty failures gi ves the ultimate life pattem
and indicates the progress towards the com-
pany's own target for cornrnercial vehicles of a
quarter of a million miles without rnaintenance.
Fairly elaborate analysis is required for
(b) two reasons. First, the range of operating
conditions of vehicle equipment is wide and
failure can depend on factors such as the type
of vehicle to which it is fitted, geographical
Figure 24.14. A ventilated altemator area, month of the year, and length of service,
(a) before, and (b) alter field testing in as well as on the quality of production. Second,
a selected unfavourable mounting pos-
there are unfortunately variable delays between
ition on a vehicle
Oetay to commsS10n D•l• from fa1lur• to r•port

'where to put what' is a sticky one, but the
design specification is the yardstick against M•an • 3·9 montha .r:; M. .n • 2·4 months
which every suggested installation rnay be
judged.
'0
~f
l
~
!lc
"'O
~E i
t:
Feedback of Service Information
E
EC
u
0 ..
0
a:
Aceurate inforrnation in service experience,
12 0 12
to be used in the developrnent of future designs, Months smc• m•nufactur• Montt-•• ••nc• fatlur@
cornpletes the design for reliability cycle.
lt is essential for confirrnation of the reli-
Total d•I•Y from manufactur• to r•port of fa•lur•s. occurring
ability (or otherwise) of new designs, and for dur1ng 12 month warranty pt!r,od
waming the rnanufacturer of the onset of any
new trouble due to changes in quality or in use.
It is also irnportant in keeping the designer
generally abreast of the rnarket conditions
and requirernents.
For some manufacturers - particularly
~ 6 12 18 24
Month• s1nc• manufactur•
30
those who service their own products in situ -

the problern of securing adequate information Figure 24.15. Delay distributions for CAVfuel
is straightforward, but for others whose injection pumps in service in Great Btitain
failure and the report of the failure, as illus-

trated in Figure 24.15. The result of this is 1 0 per cent
that the warranty claims coming in at any given 1·5 per cent /
' I
II I'
time cover failures on all vehicles, in all Actual number 1
1
I
I
areas, which actually occurred anything up to Number expected I /

for fa1lure rate / / lb·7 per cent
ten months earlier, and on products made shown / ,' /
I ' I
anything up to three years earlier. The effect 1

I
1
'
I
I
1
of any sudden change in failure rate for any / / 1/ /0·5 per cent
,l ,/ // ,,''
particular cause, therefore appears only I
I
I
.I
I
II
,
I
,,'
/ I I ,
gradually and is difficult to see among the ,,' "' ,,"'

, "' ,'
mass of data still coming in from units not
affected by the change.
---:~::~:>--'
0 12
To overcome this and to get the earliest
Months since start of production
possible, or even reasonably early, warning of
a change, routine analysis of the data is Figure 24.16. 4ssessment ol Iailure rate lrom
necessary with respect to each factor. From first Iew Iailure reports
this analysis the normal pattern of report, and
the variability of this pattern, can be ascer- new design where knowledge of the pattem
tained; significant departure from normal can of delays can allow the failure rate to be
be recognized; and the rate of failure can be predicted with reasonable accuracy from the
predicted. Figure 24.16 shows the case of a rate of build-up of the first few returns.
~
~rcentage
reit abi!tly
Be-fore next phase

___________ quantitative assessment
ot diminished reliability
(1)0evetopment
(i1)Mode! shop
samptes
(111)Design drawtngs Pre-production Initial sates
(,. )Short order
samptes
(1)Work to drawing (i)Work :o drawing Re-appraisat C<Jn"plete and

(ti)Procedu~ for · and lest to ot service rapid tec:hntca!
custo~er and orovmg specification
Dtt!O+ conditions tt!O+ claims
samp manuf~t ure 1) Machine capabilit anaiySIS by
(m)Qua!ttv deot. ,~ ch~ck and
guarantee.samp es to '·")Q 1.1 d' apptica!tons service dept.
drawtng vll ua t y au t!
Tooling Limited produc:tion Full
release retease release
Figure 24.17. The lour phases ol reliability achievement

Analysis in such detail can present an opportunity to check process capability.
fonnidable clerical problems, and some (3) InitialSales - Re-appraisal of service
mechanical aid is almost essential. In the conditions and application engineering.
author's company, a digital computer is avail- (4) Production - Rapid analysis of early
able and is used for this purpose. At present failures using statistical techniques gives
this is programmed only to sort and tabulate upper and lower limits of ultimate failure rate.
the data in the forms required, but when
sufficient data is eventually available on the The reliability seen by the customer is some-
natural variations in results, it will be possible thing less than the 100 per cent envisaged by
to programme the computer to analyse the the concept designer. The extent to which
results and to throw up only those features denigration will occur must be estimated and
which are likely to be significant. allowed for at the design stage. A quantitative
check on the actual diminution should be
Co-ordination carried out at each stage by the function
It can be seen from Figure 24.17 that a concerned, and compared with the design
wide variety of line functions are involved in estimate. While this emphasizes the importance
the reliability programme: service, sales, of team effort, the designer has a right to
design, development, buying, production expect that the activity will be co-ordinated
engineering, production, quality control and by someone else, and that the designer will
finally service again. These may be grouped be presented with the information in usable
in four . phases of reliability achievement: form and will be allowed to approach the
design problem, inhibited but not impeded.
(1) Engineering - Including design, devel-
opment detail drawings and sample manufacture Acknowledgment
responsibility. In presenting this chapter, the author
(2) Preproduction - Manufacturing quality would like to acknowledge the help given by
plays an essential part in ensuring that what his colleague Mr. D.B. Wedmore, and to thank
is tested is of known quality, to predetermined the Directors of CA V Ltd. for permission to
drawings and specifications. This also provides publish the contents.
Chapter 25
DESIGN OF MATERIALS
S. A. Gregory
lntroduction justify the special design effort and any

In respect of materials there is a situation associated research costs (SMITH, 1965).
which in some ways resembles the problern Such situations will tend to be in flow
which has to be faced in other regions of production jobs, for instance the manufacture
design. Is it better to take a standard material, of motor cars; in high cost batch production
or to design something specially? Designers such as the construction of aircraft; in complex
generally tend to take materials for granted, one-off manufacture as in the building of
as a fixed point. chemical plants.
The arguments which have to be settled This does not necessarily imply that the
partly reinforce this resemblance. A designer material design process will be carried through
or design system engaged in producing a by the final construction organization. Some
scheme for a batch production job, or a simple manufacturers of materials see the possibility
one-off job of low value, will avoid introducing of exploiting their skills and facilities in
uncertainties or expense into the task and supplying specially designed materials to fit
will opt for the use of standard materials in so certain engineering requirements. The pro-
far as these are available. It is at this Ievel duction of such materials Ieads to general
that techniques of material selection first availability and the gradual recognition of a
come into play. Here the need for exotic new standard material. There is likely to be
materials has stimulated the development of more 'fall-out' from materials design than
relatively simple techniques. from specialized part design because materials
Design of materials consists in the are more elementary, having no specific shape.
production of some arrangement of constituents As might be expected, the design of
of one or more materials to yield desired materials may lead to the design of fabrication
properties. Standard materials are those for processes. lndeed, in many of the practical
which the design has been developed and cases there is an inextricable interrelationship
tested, usually with a substantial record of (PEARSON, 1965).
application. The important variables which Much of the development of design of
affect the design of materials are the possible materials lies outside the conventional Iimits
materials and the arrangements which may be of supply of engineering materials, namely
made from them; also the specific functional metallurgy (ALEXANDER, 1965). Substantial
requirements which they have to fulfil. pressures towards the new approach come from
Material design, which starts from a thinking about polymeric materials and about
foundation provided by materials science and composite materials - but this is not to suggest
the techniques of selection, sees its principal that metallurgy is without interest in the design
outlets in those design situations where the of materials. Metallurgy provides materials
magnitude of the cash turnover is such as to generally with more reproducible characteristics.
235
Pressures to change the design of materials can Analysis of Function of Materials

come from several sources: Just as in the better-known areas of
design, a practical starting point for the design
(1) Manufacturers of materials who wish to
of materials is adequate specification of
extend their markets, particularly in competition
function. This may be precise or, more likely
with other materials of construction, whether
in the case of pioneering design, an indication
traditional or not.
of the preferred limits of a solution. These
(2) Engineering designers who wish to
might come from market investigations in the
provide special characteristics to the system
case of a materials manufacturer, or from
they are concerned with, particularly at extreme
preliminary engineering/economic design
conditions. These may be associated with high
studies in the case of the system constructor.
cost batch manufacture, or with complex one-off
systems. As far as can be seen, all engineering
(3) Engineering designers who are attempt- functions (excluding aesthetic and other sub-
ing to bring down product cost or user cost. jective functions) can be classified as either
This is particularly the case with flow system functions (transformation or transfer),
production. Container functions, or support functions. For
example, a kinematic system might involve at
In general, the materials gJ.vmg special least a torque requirement and a ruhhing contact
characteristics at extreme conditions make up requirement as system functions. In addition,
only a small part of the cost of the product; for all kinds of function, there is a persistence
the majority of the total product cost is taken obligation which may be fulfilled in a variety
up by the materials of category (3) - those of ways, of which rigidity is one. A container
associated with searches for reduction in function might require the maintenance of
product cost. These are often the materials pressure and the prevention of cooling by loss
used in the common types of construction. It is of heat.
bulk materials of this kind that tend to attract A designer attempts to isolate the details
the interest of the manufacturers in their search of each function so that an attempt may be
for large markets. made to supply what is needed against each
Where special characteristics are required, detail. The extent to which a design may fulfil
the design of the materials is almost in- the details depends upon the state of the art
separable from the design of the systems and the class of solution attempted. Composite
needing them. Further, because of the tendency materials provide more possibilities than
of materials manufacturers to avoid small homogeneaus materials. Homogeneaus materials
markets (unless the materials have a very high tend to lead to decisions requiring compromise.
value), their design must almost inevitably In some situations, and in certain industries
gravitate to the systems firm. This may even the design requirements of the materials are
lead to the development of the production taken almost for granted. For example, the
method. Where bulk outlets seem possible the heavy electrical industry sees most of its
materials manufacturer will take an interest in function requirements fulfilled in terms of high
the design of the material. Unfortunately, conductivity copper and mild steel. The greatest
there will be an underestimation of the com- variety probably occurs in materials which
plexity of the needs to be satisfied, arising execute the containment function, the in-
partly from an absence of direct practical sulation. In electronics the range of transfer
interest in the application of the material, but materials is much greater.
probably more from the diverse modes of In trying to fulfil design requirements, the
application. initial approach will probably be to use a
homogeneous material or a material found make modifications to structure and synthesize;

naturally. Where there are multiple function test; if successful, investigate problems of
requirements such materials will lead to application and problems of economic manu-
compromise solutions. facture; modify structure as required.
The response of the materials technolo- It is through the application of design
gist will be to attempt a redesign of the basic procedures of this type that there is now
materials. Such redesign, or new design, may available, in addition to nylon, such polymers
have possibilities at the molecular level, at as polytetrafluoroethylene, the polyester fihres
the micro level, and at the macro level. and films (e.g. 'Terylene' and 'Melinex'
respectively), polyurethanes, organic and
Materials Design at the Molecular Level inorganic polymers with high thermal re-
This is largely the field of the polymer sistance, many kinds of synthetic rubber,
specialist. The man who may be seen as the polyoxymethylene, and the base materials used
pioneer in this kind of design. - Wallace in a number of modern paints.
Carothers - called his sphere of activity Materials designed at the molecular Ievel
'tailoring the long molecule'. His name is may themselves be exploited in further designs
remernbered through his successful achievement for materials, either at the micro or macro
of a new polymer whose use is now widespread Ievels. The possibilities of developing new
in engineering, although it was in the replace- properties in metals in this category are
ment of natural silk in stockings that nylon relatively restricted.
came to fame.
In this kind of design, the polymer
specialist attempts to build up a molecular Materials Design at the Micro Level
structure to provide desired properties. He has This class of design is traditionally the
at his disposal what is potentially a very preserve of the metallurgist and the ceramist.
large number of alternative types of building It is perhaps too positive to term this activity
bricks. From these he has to choose those 'design'; during the last century, knowledge has
which are already available, or which promise progressed about the way in which micro-
to be commercially availahle on terms com- structure of a material influences its engineer-
patible with the limits suggested by the polymer ing performance - and this has led to control
selling price forecast. of production processes to provide the
From his chosen bricks he must build up structures thought best. However, the last
molecules which provide the desired tensile twenty-five years have seen the rise of
properties, resistance to heat, ability to absorb conscious design in the fields of metallurgy
energy, impermeability to gases and solvents, and ceramics: 'alloy design' is now recognized,
electrical resistance, etc. A great deal of its name being self-explanatory.
information is now available about the bricks In metals and ceramics, structures are
and the ways of articulating them in order to formed which may either be self-composite or
provide the specified properties, although it is hetero-composite at the micro Ievel (HOLLIDA Y
difficult to be quite sure what the overall and MANN, 1964). This is the level of particle
result will be. size at which surface energy effects are
The procedure in design is therefore likely considerable.
to be: sketch on paper the structure or class of Pure metals and solid-solution alloys (or
structure thought to be most likely to provide their ceramic analogues) give little ground for
results within the given constraints; attempt manoeuvre, although the obvious manipulation
practical laboratory synthesis; test products; of crystal size may be supplemented by
review design in the light of test results; orientation effects and, it is hoped, the in situ
development of whisker crystals. These any pressure, or liquid at any pressure, provided
additional effects give rise to fabrication that there is no continuity between dispersed
difficulties. Where there is more than one units. In the most concentrated condition
basic material the possibility of producing there will be a tendency to interparticle
new properties of value increases because the contact. At this extreme it is possible to
number of degrees of freedom is greater. arrange a composite with the phases reversed,
Hetero-composites first developed from i.e. with the disperse phase solid and the
alloys giving phase separation. lmpurities continuous phase some other state of matter.
also assisted. The aceidentel discovery of The plum-pudding structure is not normally
peculiar modes of behaviour in complex alloys, suited to tensile or torque functions because
such as age-hardening in the complex copper- of the deficiency in continuity of the structure;
bearing alloys of aluminium, and successive but the structure is excellent for compressive
precipitation in the creep-resisting low-cost performance, as in supports such as found-
alloy steels, illuminated the freedom of ations, provided that rigid 'plums' are used.
manoeuvre in some of the more complex Rigid porous structures of this class, provided
systems. Design was possible, but design the continuous phase is adequate, lend them-
with a large element of experience and selves to thermal containment. Suitable
intuition. combinations of phases give materials with
Recently there has been something like a interesting electrical properties, such as
retum to first principles. lnstead of complex controlled resistance, or suitability for heavy
alloys being used, attempts are being made duty contacts. Non-rigid porous structures lend
to work with single metals and insoluble themselves to certain classes of Containment
additions. or support duties involving mechanical energy
absorption.
Materials Design at the Macro Level
Composites at the macro Ievel give many Spaghetti Structures
facilities for design which are missing at the Spaghetti structures are built up from
micro Ievel, although they exist to some fibres or filaments in some continuous phase.
extent at the molecular Ievel. Macrocomposites The fibres may be randomly oriented or
provide opportunities for Variation in size of regularly arranged, and may be straight, wiggly,
components, ratio by volume or mass, shape, or flexible. Their length distribution may be
spatial relationship, and continuity. narrow or broad; the length-diameter ratio
Designers are familiar with many common influences the composite properties.
composites and may have designed some With spaghetti structures it is possible to
without knowing that they were designing develop some kind of rigidity with very small
materials. Perhaps the commonest are concrete fibre-continuous phase ratios. This property
and reinforced concrete. of the structure is exploited in papers· of all
The three broad classes of composite are sorts, and in most textiles. With high porosity,
defined in terms of structure: plum-pudding, good thermal insulation and other requirements
spaghetti, and sandwich. may be achieved. By specific orientation of
the fibres, directional properties may be
Plum-pudding Structures developed. Under conditions of orientation,
The plum-pudding structure has variations it is possible to provide high packing densities
in terms of size and size range of the disperse of fibre and thus develop valuable tensile
phase; in shape; in phase-ratio. The disperse properties. Pressure vessels formed by winding
phase need not be solid, and may be gas at suitable fibres at the correct angle on mandrels
to a high density, with an impregnation of selves have been specially designed in

polymer, now find many critical applications. considerable detail.
Sandwich Structures Conclusion

Sandwich structures are rich in possi- The basis of a rational approach to the
bilities and many are commonly exploited. general design of materials has still to be
The successive layers in the sandwich may be established. In certain areas such rational
similar to earlier layers in the sequence, ·or design is possible but much more information
completely different. The layers may them- has to be gathered and specifically directed
selves be composed from complex structures. to the task of aiding design. Up to the present,
Many common timher products are like this. much investigation has been devoted to the
Common sandwich composites which are replacement of natural products. A faculty of
purpose designed by the engineer include the technologist is (in von Karman's words) to
furnace walls (for which rational design and imagine what does not yet exist. The poten-
optimizing procedures exist) and corrosion- tialities for new systems through the invention
resistant or heat plus corrosion-resistant of new materials is such as to suggest that the
materials which can take a tensile load. The time is now come for a systematic exploration
normal motor car tyre provides an example of of material structures in relation to engineering
a layer structure in which the layers them- function and use.
Chapter 26
NEW IDEAS IN THE DRAWING OFFICE
P. McMullen
lntroduction and too long. Surveys have shown that as much

The term 'drawing office' goes back a as 30 per cent of a man's time is spent in
long way, and until the advent of technical obtaining information.
education this was the place where 'drawings
were made'; there the whole of the engineering Information to Designers
thinking was done - and all too often still is The following is an abbreviated list of
done - by very few people, and in many cases information that can usually be prepared before
by one person only, out of a drawing office of a job is started and put in a folder for reference
ten or more. The result was that a draughtsman by the designer draughtsman:
was purely a 'drawer of lines' and was judged
as such. The emphasis was on the drawing (1) Specification sheets giving precise
and not on the design. Also, because the requirements for the piece of equipment to be
drawing office head would not delegate even designed.
the simplest design, he usually became the (2) Mass balance sheets (these refer to
boUleneck thus encouraging the draughtsmen aspects of chemical system functions) giving
to over-elaborate just to fill up time; hence quantities, sizes, rates of flow, horsepower,
over-elaboration became 'established practice'. etc.
(3) Physical properties.
(4) Equipment schedule with leading
The Object of the Drawing Office dimensions.
The main object of a design office is to (5) Standards book for the job. This will
clothe technological ideas in hardware, and to include a selection from the D.O. Standards
provide clear instructions to others as to how book, plus photocopies of manufacturers'
to fabricate and erect equipment, buildings, standards suitably marked up.
roads, instruments, pipelines, power units etc. (6) Target completion date of his piece of
design and how this fits in with the programme
Communications as shown on the critical path scheme.
This is the problern of clear communication
from one group of people to others. There are
many methods, all of which have their weak- Orthographie Projection
nesses and advantages - there is no one best The universal basic technique in design
way. Clear communication starts in the design offices is the orthographic projection and,
office. Most supervisors do not consider care- despite its many weaknesses, this is likely
fully what they want to put over to their to remain the prin:cipal means of communication
assistants. They do not put them fully into from the design office to fabrication and con-
the picture and assume a knowledge of a lot of s tructian. The main shortcoming of ortho-
background; the instructions are mainly verbal graphic projection is that it tries to represent
241
three-dimensional objects by two-dimensional In many cases in the author's design office

means. As a result only those people skilled it has been possible to cut out checking
in the art of reading engineering drawings can altogether and in only a few cases is 100 per
obtain the proper picture and even they often cent checking still done. The average time
fail. spent in checking is now between 40 and 50
Drawings, particularly arrangement draw- per cent of what was done five to seven years
ings, are viewed by very many non-engineers ago.
and in most cases they convey virtually nothing.
Perspective drawings as frequently produced Models
These can be divided into flow sheet
by architects can be helpful, but they are often
models, arrangement models, and design models
deceptive and are expensive to produce.
showing pipework, electrical installation
Simplification of Dr.awings including lighting, and instruments.
Models are three-dimensional design
Before proceeding to the new ideas in the
arrangements of a three-dimensional object.
drawing office, a few ways in which productivity
They are readily understood and have many
in design can be improved while using orthodox
other advantages over conventional drawings,
methods are suggested:
not the least of which is that several people
(1) Simplify. can discuss design points round a model and
(2) Do not make unnecessary views. know that they are understood.
(3) Eliminate. The design is done on the model: it is not
( 4) Do not repeat the same detail. copied from drawings. Models are expensive
(5) Do not redraw. but only a quarter to a half the cost of con-
ventional design methods. Speed of design is
A useful leaflet entitled Functional Drafting
many times that of conventional methods
is available, but even this does not go far
because people understand the design and, as
enough: for example, cross-hatching can
a result, only a fraction of the usual late
easily be eliminated. changes occur.
Apart from the waste of time in the design Electrical and instrument equipment and
office, some other reasons for eliminating runs are shown, as well as pipework; snarl-ups
drawings are that they have to be (i) registered, are avoided and planning is better. Since using
(ii) filed, (iii) filmed, (iv) printed, (v) distri- design models for chemical plant design, the
buted, and (vi) checked. author's design office has met and bettered
target dates, and the overall costs have been
Checking
This is a time-consuming, non-produclive less.
occupation which occupies a large proportion
Preparation of F low Sheets and Engineering
of the most reliable and experienced man-
Line Diagrams
power in a design office. The following
The preparation of flow sheets and
questions can be asked:
engineering line diagrams form the basis of
(1) Is it necessary to check at all? chemical plant project design.
(2) If it is, to what degree? There is a basic set of standard symbols
(3) Where should the design be checked? which should be used, but these need supple-
(4) When should the design be checked? menting by outlines of equipment which make
(5) By whom should the design be checked? the flow sheets and line diagrams more real.
(6) What design methods can be used to A lihrary of basic equipment outlines, in some
reduce checking? cases showing valve arrangements, can be
prepared and added to as required. These instrument departments to add their design for
outlines are photographed on to clear film and cable runs and instruments layouts.
can be produced to any desired scale in the
numbers required. They are set up on a 'Melinex' General Arrangement
film with a squared backing sheet from which By using the cut-out techniques mentioned
a master is printed. A print is taken from the above, it is sometimes possible to make
complete general arrangement drawings. This
master on to which are drawn the pipelines
is particularly applicable to plants where the
for the particular service and, in the case of
same equipment is repeated.
engineering line diagrams, all valves and Sometimes illustrations and photographs of
instruments are shown. equipment have been taken, reduced to the
The result of this method has been to required scale, trimmed, and included in a
reduce the man-hours required to as little as general arrangement.
one-eighth of the previous times, not only
saving manpower but calendar time in the Modification
early stages of design. One of the advantages of printing on
Melinex sheet is that the print is on the
Pipework Drawings reverse side of the sheet and erasures are made
on the back, leaving the face side intact.
Isometrie drawings are frequently used
Where major alterations are required, a
nowadays for pipework fabrication. They do
print is taken, the parts to be altered are
give a fair idea of the particular line drawn but literally cut out with a knife or scissors, a
usually lack any information of local obstruc- new master prin ted from this, and the alterations
tions and limitations; they are not self drawn in on the blank spaces.
checking.
By the use of selected basic plant arrange- Notes, Specifications, Tables and Symbols
ment drawings on Melinex reproducibles, it is These are often repeated on a series of
possible to take dimensions from the model and drawings. They should be typed with a type-
show lines to scale in plan and elevation in writer on to the film (there is a machine, made
the standard orthographic projectiort. A few in Germany, which can be attached to the
key dimensions only will be shown. Some of drawing board for typing on to drawings). The
required number of copies may be printed and
the advan tages of this method are:
stuck on to the basic drawings. Note that this
requires one checking only; a machine is used
(1) All engineering contractors are familiar to do a machine's job.
with orthographic projections.
(2) The master general arrangement needs
Photo graphy
no further checking. Generally in industry, photographs are
(3) The pipelines are very largely self regarded as mainly a medium for recording
checking. completed things, but a photograph can also be
(4) Owing to the stability of Melinex, the used as an objective and functional tool that
pipe fabricator can scale all dimensions from is complementary to design and drawing office
the drawing for his shop details. work. Photography, used in the broadest sense,
(5) Speed of communication is improved. is the tool that has helped to improve drawing
office methods.
Electrical and Instrument Layout Stereoscopy

Using the 'to scale' pipework arrange- Stereo photographs are a more advaooed
ments produced early in the design programme, form of visual aid; pairs of pictures are
master copies can be passed to electrical and arranged so that, with a suitable viewing
9
instrument, the picture can be seen in three are lodged where they are most needed - in the
dimensions. The value of stereo pictures is design office.
that they permit the observer to separate ( 4) The photographs need not be analysed
different planes in the object and to estimate until the information is needed - dimensions
spatial differences, which is often difficult, can be taken from the pictures within a few
if not impossible, with single pictures. minutes at any time.
Stereo pictures can be presented in
several forms. Stereo colour slides that can be Two photographs of the subject are taken from
examined through hand viewers give the most slightly different viewpoints. The photographs
realistic impression, but only one person at a are analysed on a plotting instrument by stereo-
time can see the picture. Slides can alter- scopic examination. A mark in the viewing
natively be projected on to a screen and viewed system appears to float in the three-dimensional
through polarizing spectacles, so that any picture. The position of the floating mark can
number of people can see the picture simul- be moved through the picture and made to
taneously and can point out details to each follow the outline, or to match the position of
other. any feature that can be seen in the picture.
Another convenient form of presentation is The plotting system also moves a pencil
stereoscopic prints in black and white, or in that draws the plan shapes of the features being
colour, mounted on cards and viewed through probed by the floating mark. The mechanical
a simple magnifying device. Stereo prints can geometry of the plotter eliminates perspective,
be provided as either individual pairs, or as therefore the drawing is orthogonal and true
multiple strips to make a form of line-overlap to scale.
that presents all aspects of the subject. The
overlapping of any pair of prints can be viewed
Photogrammetry of Piping Design Models
stereoscopically, and, due to the overlap of the
One application of photogrammetry in the
area included in each picture, any detail can
be viewed from three or four different angles. chemical industry is for making scale drawings
of piping designs that have been produced by
Enlarged prints of stereo pairs may also be
viewed with a similar device in which the modelling. The problern here is that whilst the
eye-base is extended through mirrors. design has been modelled in three dimensions,
it must eventually be recorded on paper to give
instructions to the piping fabricators and to
Photogrammetry the erectors of the plant.
Photogrammetry is a photographic tech- The design on the model can be extracted
nique that provides a means of measuring by photogrammetry and drawn more accurately
precis.ely, in three dimensions, objects In a than by any other method. Models are photo-
picture. Now well-established, with appli- graphed in sections of about twelve inch
cations in several non-topographical areas, cubes or larger, according to the building
photogrammetry was originally developed for module.
making maps from air survey photographs. For this work, the plotter is fitted with a
For the engineer who similarly needs second drawing table. This allows both plans
'maps' of designs and illustrations photo- and elevation sections to be drawn simul-
grammetry offers a means of obtaining dimen- taneously, to provide the information in the
sional records with the following advantages: required form. The plotted drawings are scaled
to match a background drawing that has been
(1) Photographs are made from a distance prepared from equipment drawings and Co-
without physical contact with the subject. ordinates, and from structure drawings showing
(2) A complete and permanent record is the required sections. These arrangements are
made in the photograph of all details that can the basic drawings from which the plant is
be seen in the camera position. constructed and are the most accurate source
(3) The visual three-dimensional records of information available.
Photogrammetry of Existing Plant to be outstanding. Compared to times estimated

Photogrammetry can also be used for for conventional methods, design office man-
recording and dimensioning existing plants. hours for modifications to chemical plant are
Here, the problern is that after a plant has less, at a conservative figure, by 15 to 50 per
been in production, even for only a short time, cent, and workshop effort is less by up to
it has been added to and altered. It is no 20 per cent.
longer exactly represented by the original The outstanding merits of photodrawing
design. are:
From the poin t of view of designin g major
(1) The photograph shows the existing
modifications, these differences and other
details that do not appear on the model or situation up to date, including details and
arrangement drawings, can be a major obstacle. changes not shown on existing drawings.
Other photographic aids are used as visual There is therefore no need to check and correct
references to inform the design office of the existing drawings before new work can be
present situation, but photogrammetry can designed, and repeated visits to site can be
often go a stage further, either to bring the eliminated.
design office records up to date, or to provide (2) Site measurement is reduced to the bare
a means of obtaining exact dimensions in essentials for the job. None is required for
specific areas just when they are needed. the sake of detail otherwise necessary for
For this work the photographs must drawing the background.
(3) Visual representation makes the design
naturally be taken from a greater distance than
problern clear and easier to recognize.
when the subject is a model, and the cameras
(4) Drawing time is reduced because most of
are transferred from the short base to a longer
the drawing effort is confined to new work.
base bar that is mounted on a surveyor's tripod.
(5) Because the presentation is clear,
locations and instruments are quickly and
Photodrawing
precisely understood by the craftsman and
Photodrawing is !!otentially the most
site Supervision time is reduced.
valuable new drawing office technique that
has appeared in recent years. Photodrawings
Photogrids
are engineering drawings in perspective,
superimposed on a photographic background. There are occasions when it is useful to
They provide an ideal means of communicating be able to measure dimensions from an ordinary
instructions for modifications to existing photograph. With care this can often be
situations. The technique comprises three achieved to an extent sufficient for approxi-
main stages: mations. A simple case has been taken from a
straight-on viewpoint. On the basis of one
(1) Photograph the existing situation and known dimension the photograph can be printed
print on transparent film. to a selected scale, or a scale can be drawn
(2) Tape the film print on to the back of a to match the scale of the picture. When the
translucent drawing film, and overdraw the subject exists in more than one plane, a
engineering information. dimension must be known and a scale prepared
(3) Reproduce the combined photograph and for each plane in which dimensions are required.
drawing on ordinary dyeline or print paper. When photographs are taken from oblique
angles, dimensions can still be obtained. The
The use of photodrawings has several advan- perspective of the photographic image is in the
tages. 8oth design time and construction time same central projection as that normally used
can be substantially reduced. Exact com- for drawing in perspective. It is therefore
parisons are difficult to obtain because the possible to superimpose upon the picture a
same job is seldom clone twice using both old scale grid drawn in the same perspective as
and new methods, but results obtained on that of the photograph. Such a grid is prepared
jobs when photodrawing has been used appear by applying in reverse the rules for constructing
a perspective drawing. The lines of obviously ivity in the design office, it is foreseen that
horizontal features in the picture are extended the present rate will again be doubled in the
to their respective vanishing points. next few years. It should be noted that, with
the exception of photogrammetry, no expensive
Conclusion equipment is used.
There is a whole range of further points
to be considered, such as the reasons for using Acknowledgment
the material Melinex, types of print, sizes of I would like to acknowledge the help given
drawing sheets, pencils, lettering, etc. - but in the preparation of this chapter by my
this would occupy at least a further chapter. colleagues in ICI Plastics Division: Mr. R.G.
A few years ago, the design rate in the Farrand of the Photogrammic Department,
author's office was considered reasonably Mr. A.S. Monk - Standards Engineer, Mr. J.
good, but the rate is now well over twice what Masterton- Head of the Drawing Office Records
it was. Although only a start seems to have and Drawing Reproduction Section, and Mr. H.
been made of tackling the problern of product- Bennett of the Drawing Office Methods Section.
PART V
MANAGEMENT AN D DESIGN
Chapter 1.7
DESIGN POLICY FORMULATION

B.T. Turner
lntroduction now and in five or ten years time? What should

That progressive companies must have an be the scope of the product market - is it to be
overall design policy for innovation and global, continental or national? If a company
progressive development goes without saying, has to tender for 'turnkey' contracts it is
bu t even minor broad policy statemen ts on new essential that a policy of progressive product
products or systems often require courageaus improvement, in line with the overall systems
decisions to be taken. The consequence of requirements, should be pursued. ßut even for
these decisions may very largely determine concerns selling individual products, develop-
the survival and prosperity of the enterprises ment direction must be specified at the company
concerned. level in order to ensure a commanding sales
l<ARGER (1960) has stated that 85 per position for the future.
cent of business volume today comes from There is good reason for believing that
engineering products which were unknown some firms would rather run to be second than
about ten years ago. Furthermore, there is the take the risk involved in being first. This
ever-increasing tempo of technological change, could be not so much a design policy as a way
and the lag between discovery and use is of death; although, if the second is close to
beginning to shorten. Again, fresh ideas for the first, it could pay off (Comet v Boeing 707).
products and systems are tending to be more However, a real policy for design which would
complex and of greater söphistication; this bring wealth and prosperity to an enterprise
leads to high research and development costs demands something more than new ideas and
and complicated testing, etc. The result of this courage alone. Restrictive practices, carteli-
quickening pace is that both promoters and zation, P.revention of newcomers entering in
investors are finding that new techniques of competition, maintenance of cost or production,
surveillance, screening and forecasting are and selling are all factors which mitigate
essential if any degree of confidence is to be against innovation. If the present production
achieved in launehing a new design. methods and machinery yield a comfortable
Even when such a decision has been taken profit which is assured by rings and cartels,
and communicated to a design group, there tariffs and quotas, then incentives to redesign
still remeins a need for clear detail policies or design de novo are small or non-existent.
regarding any new enterprises or even for Sometimes the cost-benefit factors for
modified designs. Hence design policy can be introducing a new product through a particular
considered under the two headings of company industry do not provide adequate incentives.
design policy and design group policy. If such products are in the national interest
g o v e r n m e n t action will be r e q u i r e d
Company Design Policy (DUCKWORTH, 1965).
At the company level in a large enterprise With the ever-increasing need to export,
there is a need to co-ordinate and guide the many British firms will have to reformulate
various product divisions with regard to their their campenies' design policies to become
future products. What should they be making more progressive for the nineteen- seventies.
249
For example, because some large authorities manufacture and operation phases through
demand a product which is double-foolproof which a product passes. Vital experience
with lush finishes everywhere, designs become and data which is suffused in the whole design-
uncompetitive for foreign markets. It might. to-use evolution is often lost because it is not
pay an enterprise to consider what the true recorded and coded so that it can easily be
needs of the circumstances really are for retrieved for future use. Cost-benefit curves
foreign as well as home customers. It might should be produced for past products (SITTING,
pay to design a product so that the extra 1963). This is not easy to do since most good
finesse required can be added to a less engineers tend to move to new pastures as
stringent basic design, so meeting both soon as a job has been completed and are
markets. Alternatively, separate designs may reluctant to record mistakes, errors and
be required for the home and overseas omissions and are poor at tidying-up operations.
customers. British standards may be fine for But if cost-benefit curves can be done well,
this country when they are eventually issued, they can considerably aid future design policy,
but too rigid to allow compatibility with for they help to identify strengths and weak-
foreign requirements. These factors can lead nesses.
to over-pernickety design and so hamper
exports. Data on Customer Requirements
It will be necessary to invest with fresh Any forward-thinking unit must have
drive the past industrial dynamic which this customer contact as well as research contact.
country possessed, by applying a scientific It needs to determine, through systematic
approach to the problern of design innovation. techniques, the real needs of markets and
Every company in the manufacturing industry individual customers (Chase Manhatten Bank
requires to create within itself a mechanism Report on European Markets, 1964;
which will constantly revi vify its designs by HOGMANDER, 1962). The business is deter-
applying fresh ideas. mined by satisfying customers' needs, not by
A forward-thinking unit, or product policy producing technical perfection of a product
committee, must be formed which can take that has become outdated. It is useless to
into account past performance of the companies' make super rat traps if rat poison is what is
products and the economic, social and other needed, or elegant drawing pins if drafting
external influences which will affect the tape is really required. The cycle starts with
products' future (CORNFORD). Designers the customer and ends with the customer; the
should be included in the unit as well as
representatives from sales, marketing, research Th~ bus1ness. e.1usts to deliver
and development, and production departments. value sat1sfdcl 10ns at a proflt
Such a committee would be responsible for

providing top management with a design brief.
This would embrace such factors as an estimate
for the Validation work necessary to establish
technical feasibility of a new product, a survey
of the market giving probable sales and profits
for the new product, and an estimate of the
capital investment required. For this work
certain internal and external data must be
collected by the forward-looking group.
Proftt, reputat1on, growth
Data on Past Products ~formoney\

One of the undoubted weaknesses in
British industry today is the poor communi- Figure 27.1. Design in the total business
cation feedback from the design, development, cycle
MANAGEMENT AND DESIGN 251
complete business is centred round the Figure 27.2 shows the necessary balance
customer, not the company. The 'correct' that is required when looking into the long-term
product is one that sells in a competitive future. The possible market requirement must
market and wins a profit and a reputation for be balanced against research and development
the company concemed. Figure 27.1 portrays probabilities. Surveillance of the field can be
in simple form the complete cycle of operations. carried out by engineering market research,
In this cycle, information is fed and Validation with careful interpretation (WILLIAMS, 1963).
work is carried out by several other depart-
ments, such as development, accounts, person-
nel and other service sections. But the cycle
starts and ends with the customer, and design
is in the mainstream of translating ideas to
hardware for money. The money obtained from
the selling price has to be sufficient to yield
a profit for survival.
Data on Resources Research probab1lities Market poss1b1lities

Not only should the objectives be spelt RISk Reward
out in broad terms at company level, but the Prest1ge Prof1t
resources of a company must be known in
terms of strengths and weaknesses. The design Figure 27.2. Dependence of Iong-term policy
policy is governed, in part, by building on the upon balancing commercial possibilities against
strengths and minimizing the weaknesses of research possibilities
any concern. Special expertise and skills,
which have been developed over the years, This must be clone on a continuing basis,
give certain firms superiority over competitors since markets can change rapidly. Technical
and wherever possible these should be utilized research may be conducted inside the firm
to the full and developed further. Such factors concerned or by some outside national body
as financial reserves for carrying out extensive such as NEL, NPL or RAE. Great advantage
development work on a new product must be acrues from designers contacting these govem-
considered, together with any necessary legal ment research centres to find out what is going
agreements (patents and licensing), factory on and what likelihood there is for practical
capacity, existing plant capability and appro- design embodiment of their researches. Quan-
priate costs of manufacture, etc. Sometimes, titative assessment of possible cost-benefit
when entering a new field, additional skills ratios can help to optimize capital expenditure,
may have to be added to a firm and a merger etc.
or takeover may become necessary. For short-term policy considerations, it
may be necessary to create markets. There are
Maintaining Balance many relevant factors which have to be con-
With the data provided it is necessary sidered and some of these are listed in Figure
to obtain a balance between the market poten- 27.3. Design policy is influenced by most of
tial and the company potential. For example; these factors to a greater or lesser degree.
the pressure for a new design may come from For example, advertising is essential to
a research and development department, or increase sales of not only soap powders but
from a sales group which has scanned a also nut-and-bolt engineering products.
particular market, or from some other source. Ultimately, it will not be the country that can
The balance is influenced by uncontrolled produce most that will win the battle for world
external factors of supply and demand, but may markets, but the country that can sell its
be controlled to a certain extent by a company products most efficiently - and advertising is
supplying a need or creating a demand. necessary to boost sales. If sales increase,
it may pay to design special tooling to cope Possib•lrtles Danger pornts
with the increased demand; if direct selling is

used, servicing will be at the customers site, (1) lncreas•ng the sales pnce Marke! may decl•ne
and consequently design policy will be affected. (2) Increasing the market by Too Iew a prof1t marg1n
reducing profit per uni!
Political activities may include govemment
(3) Reduc1ng total cosls and Qual•ty may decline and
contracts, perhaps imposed for defence reasons, so reductng saies prrce- c.apttal may be necessary
and making use of government surveys, etc.
Short-term boosts may require design policy
Figure 27.4. Profit increase in a static
changes, which can include collaboration with
competitors or price cutting and dumping. mark~t situation
Taking a licence will also require careful
design handling, for often drawings have to be also help as unprofitable lines are cut out and
anglicized and redesign may be necessary to profitable ones become standardized.
use existing production facilities. Product In considering all these factors, the
ability to achieve a profit is of paramount
importance. A profit increase with a static
situation may also require attention and the
possible ways of achieving this, with the
inherent danger points, are listed in Figure27.4.
For any changes made, the break-even point
should be determined.
C ompany cantrot C 1s achreved
by etther creatrng a market or
When all these factors have been carefully
sat1sfymg a market considered in the light of the strength of the
Creattng a market by: Sat•sfy1ng a market by. concern, a company policy for design can be
(1} Sales acttvrttes (1) lncreasmg productron formulated. Such a policy must ensure that
(2) Pol1tical act1Vtltes. (2) Decreasmg product•on any projected products meet the market at the
(3) Short-term boosts (3) Givmg lttences for manufac.tuf'!' right time, hence proper planning to phase in a
<4} Taktng l•cences (4) Altenng channels of dtslr•button new product and drop an old one are important.
(5) Altenng product l•fe-
cycles
(5) Ma•nta•ning standard•zahon It must be remernbered that the problems asso-
ciated with discontinuing a product are as
Figure 27.3. Dependence of short-term policy
upon balancing potentials
Product tdeas trom any s.ource
life-cycles will influence design considerably

as obsolescence is considered vis-a-vis •
New product propo!.al Feedback
longevity of life.
Changes in market conditions may require
action to satisfy a market, and this can be
done in a number of ways. Some of these are
listed in Figure 27.3, and it can be seen that
these also demand design effort of one kind or
another. For instance, increased production
may be achieved by additional capital expen-
diture or by sub-contracting, both of which
require extensive design liaison; decreased
production may require diversification or
closing down of certain lines. Yet another way Phue (I) ends
of satisfying a market might be to give licences
for manufacturing and to alter the channels of Figure 27.5 Phase (1) in the introduction of a
distribution. Rationalization of products may new product
MANAGEMENT AND DE~GN 253
great, and sometimes greater, than introducing parts and make the product containment as
a new one. small as possible. The technique of value
Phase (1) of the operation of introducing engineering can be applied to this aspect with
a new project is now completed, and the great effect if it is used in the design process.
discrete steps in formulating the company For some engineering work, the use of nota-
design policy are set out diagrammatically in tional models can be a valuable aid to
Figure 27.5. It should be noted that, before achieving the required simplicity.
design and any development are authorized, a
considerable amount of data collection, Cost and Weight Control
collation and sifting has been carried out. Targets for cost and weight should be
The product policy committee has considered set as a matter of policy, and should be con-
and weighed carefully many factors given them tinuously monitored. For example, with heavy
by specialists. Only by doing this, can a electrical machines the effective use of active
correct company design policy be formulated. material - copper and coreplate - cannot be
reduced if performance guarantees are to be
Design Group Policy met. The required amount of copper must be
A policy is not a directive or a command, present, but the other components may be
but a guide. The company design policy controlled by weight and cost factors so
clarifies the viewpoints of top management that good utilization of material for fabricated
concerning direction and provides a pattern or parts is achieved. On certain stator frames it
framework within which the design group may has been shown that material wastage can be
operate. Such a guide establishes latitude and as high as three times the net weight. Here
longitude of product design decisions, antici- again all the company's past designs should
pating future trends and conditions; but detailed be analysed so that realistic figures can be
design policies must be made at the design set as targets. A typical example for a
group level. It is as well to have certain generator is given in Table 27.1. Both cost and
policies written out, so that all members of the weight monitoring on !arge projects can be
group clearly understand what is demanded of greatly assisted by the use of computers.
them and what a're the product or system
objectives. Standard Parts, Components or ltems
When considering this subject without Standardization should be part of every
reference to a specific product, it is not design group policy. Wherever possible, a tried
possible to be too specific about design policy and proven designed part or component should
proposals, but some of the most important be used. This applies for systems designs as
common aspects are given here. well as product designs. Non-standard parts
Of course, every product can be looked have high production costs, high inventory and
upon as an organism with a definite life-cycle. work in progress costs, as well as difficulties
The hard effort required to bring a new product in planning and inspection. A typical example
to birth from abstract ideas needs an ex- showing the relative costs for a standerd
pensive launehing period and, like a child, machine and a special machine for a small
it may require considerable support before it industrial motor can be seen in Figure 27.6.
is fully developed and self supporting. In order Control of the use of non-standerd items is
that the growth phase may be reached as soon probably best done by a committee approach
as possible, the following factors are con- where full justification has to be given.
sidered important. Drawings of proven designs need to be
coded, not by piece numbers, but by shape and
Simplicity size, so that designers can easily retrieve
In general, the simplest design that meets past designed components, parts, etc., to use
the specification should be used. The policy in the new products. Similarly preferred sizes
should always be to reduce the number of of raw material should be used in new designs,
Table 27.1. Material, Labour and Factory Overhead as percentage of Basic Product Cost
Factory
Material Labour Total
overhead
Field system Frame 5·46 0·44 0·95 6·85

Main pole punchings
and assembly 3·08 0·44 0·95 4·47
Main field coils 2·69 0·22 0•46 3•37
Compole punchings
and assembly 0·90 0·27 0·60 1·77
Compole coils 1·56 0·40 0·84 2·80
Compensating winding 3·30 0·64 1·40 5·34
Field connection 0·55 0·19 0·42 1·16 25·76
Annature Shaft 6·95 0·51 1·ll 8·57

Annature hub 0·81 0·40 0·86 2·07
Armsture core and
endp1ates 4·32 0-'75 1·67 6·74
Armsture winding
equalizers and assembly 3·84 3·25 7·08 14·17
Half coupling 1·79 0·25 0·55 2·59 34·14
Commutator Commutator hub and baffle 2•36 0•16 0•34 2·86

End rings 3-42 0·12 0·28 3·82
Commutator bars and risers 6·10 1·52 3·30 10·92
Commutator assembly 0-43 1·15 2·52 4·10 21·70
Brushgear Brushgear 2·08 0•99 2·14 5·21 5·21
Covers Endbells 0·92 0·91 1·99 3·82 3·82
Pedestals Pedestals and bush 1·87 0·39 (}84 3·10 3·10
Miscellaneous General erection,

painting, etc. 0·29 1·85 4·13 6·27 6·27
Total 52·72 14·85 32·43 1()().00 100.00

ITime compari!.on I efficiency, it can lead to an inflexible product.

It may sometimes pay to have additional
lc_
strength added to scantlings in order to allow
I
~r---o
for higher compression ratios later on. In some

cases this conflicts with the cost and weight
control, and necessary reconciliations have
lrrmTTTTTm-l' tobe made.
L=
Production
lnsurance that Designs can be Maintained,

UW.llll.UW:ltion
Operated and Cleaned Easily, and are Pieasant
to View
Standard Non-standard Cost effectiveness of any technical product
does not only depend upon pounds per output
(.t/kW or h.p., etc.), but also upon cost of
serv1cmg and installation. The estimated
outage cost of a 500 MW steam turbine for one
month is about n million. If the duration of
such an outage is largely caused by the
difficulty of dismantling to install a new part or
component the design must be considered in-
adequate. A servicing policy must be laid down
for the product. Will replacement be by assem-
costs
blies or components? What holding of spares
will be at site, at base or elsewhere? What is
spec1at design to be their shelf life? What is to be the warranty
L tests
period for the product? Will the .warranty be
tied to regular servicing conditions by
Standard Non-standard approved dealers or agents? Operation of
certain engineering products too often assume
a superhuman dexterity: control levers are
Figure 27.6, Relative costs of a standard
difficult to operate and meters are placed in
design and a one-off design
impossible positions. Such ergonomic factors
are vital and basic data sheets on space
since this allows flexibility of purchasing and requirements and optimum movements for
cost reduction by bulk buying. drivers, pilots and operators must be issued to
For some products, careful considerations the design team (PILDITCH, 1964). Cleaning
should be given to the setting up of the correct is also important, and aesthetics are becoming
standards so that overlapping is avoided. more and more important even for capital goods.
Here, preferred numbers and geometric series A design based on a sound policy which took
of sizes should be used. Where some form of these factors into account is shown in
variety is required for specials, the design Figures 27.7 and 27.8.
should be such that these can be introduced to
a standard framework on the last 20 per cent Manpower Considerations
of the final assembly. Every design group policy must consider
the manpower available to tackle the new job,
Allowance for Flexibility (uprating, etc.) and whether the men are trained for the work?
With an entirely new design on such More often than not, little provision is made
products as prime movers, a group design for training or retraining. Members of the
policy should initially be laid down about forward-thinking unit must gi ve talks to the
possible requirements for growth of perform- group so that the complete background know-
ance. If the initia~ design aims at too high ledge of the proposed project is understood.
As new tools and techniques become available At the other end of the manpower scale,
they should be committed to software for it is necessary to have operator-training under-
programme-learning. Teaching machines may way well before delivery. Initially, this will
be used to teach designer/draughtsmen the probably have to be done at the designers
importance of such aspects as statistical works. 1\fodels or analogues may be required
tolerancing, reliability factors and bearing to familiarize operators with control procedures.
design data. Certainly handbooks will need preparing, and
it should be a design policy to issue these
when delivery of the product is taken. For
many engineering products these are either
forgotten or are produced very late.
Use oftheBest Method to Communicate

Design lntent
While the drawing has been accepted as
the universal language for communicating a
design, too often a two-dimensional presentation
has been used. A good design group policy
would include the necessity of using repre-
sentational as well as orthographic drawings.
Wiring diagrams may be replaced by
computer print-outs, piping diagrams by models
and photographs, testing schedules by tape
recorders, and so on.
Production Considerations
No design group policy can afford to
neglect new procedures and present process
Figure 27.7. tngineer's control cabinet capabilities. Designers must know where and
in open position for servicing how their creations are to be made. It may be
profitable to set up a 'make or buy' committee
which consists of production, purchasing and
design representatives. If items are bought
outside the company, then careful specifi-
cations are necessary, and a project quality
survey of the sub-contractors facilities should
be made before a contract is placed.
Setting up of Adequate Check-points

in the Organization
In order to control the process of producing
the correct quality hardware, appraisal check-
points must be provided and these can best be
illustrated by referring to the flow charts given
in Figures 27.9, 27.10 and 27.11. If these are
iead in conjunction with Figure 27.5, it will be
seen that altogether there are eight distinctive
check-points in the translation process from a
Figure 27.8, Testing and replacing a logic feasibility design study to the issuing of
circuit manufacturing instructions. This is for the
Ph&s• (2) t>.gins Ph&s• (4) b~ins
Sales
estimates
rev•ewed
by sales manager
D•sign &pproval
by
chief ~ngin•eor
Ph&s• (2) •nds Ch•ck-lists

compl•t•d
Figure 27.9. Phase (2) in the introduction : Sates est•mates, marke-1

of a new product ] Information, Ptc. checked
Phas• (4) •nds
Figure 27.11. Phase (4) in the introduction

Phase(3) beg1ns
of a new product
Prov•S•onal ~gtneer•ng spec•ftcat•on

Phas• (5) beg•ns
F~rst produclton bAtch
---raltfied by ~ales manager
OPve!opment model approvat

C~ck by
po•nt cl"uef engmMr, works manager, sates manager
V
V~s
Figure 27.10. Phase (3) in the introduction Figure 27.12. Phase (5) in the introduction
of a new product of a new product
introduction of a batched or flow-production atmosphere. The accelerating pace of inno-

product. vation will continue, and consequently
Of course, these may vary from company companies will have to consider even more
to company, according to the nature of the carefully their choice of products. The
business undertaken, but it should be a design scientific method must be applied not only
policy to fix where and when these checks are to confirm technical feasibilities, but also to
to be taken. Only by using such 1 gates' can market considerations and financial aspects.
errors and omissions be avoided and a selling, Company policy must lay down the broad
profit-making product be produced. It should be direction for design groups to follow, but the
added that, in line with the production, there groups themselves must pay attention to many
must be a proper marketing plan supported by detail policy matters, narrowing the framework
adequate publicity and this is indicated in of operation so that optimum use of the
Figure 27.12. Many good engineering projects company's resources takes place profitably to
have failed either because they have been satisfy an identified need.
poorly handled or have been launched at the This chapter specifies some of the require-
wrong time. To combat the former there must ments for successful design policy formulation,
be a publicity promotion programme, and to but it does not claim to be exhaustive. A
avoid the latter, adequate planning throughout possible flow chart of the operation of
the design and manufacturing processes must introducing a new product has been suggested.
be set up as a design policy. This emphasizes the need for providing
adequate check-points throughout the complete
Conclusion operation of converting ideas to money.
In this age of rising expectations, custo-
mers for capital and consumer goods are
becoming more discerning. They are better Acknowledgment
educated, more sophisticated and more affluent. I wish to thank the English Electric
Design has to be tailored to suit this new Company for permission to publish this chapter.
Chapter 28
INNOVATIVE DESIGN AS A POLICY FUNCTION
R. Davis
lntroduction depending on technology. It atrophies creativity

This chapter concems the optimization of by putting the control of creative facilities in
conditions to encourage improvements in the wrong hands and can only be rectified by
design for manipulati ve machinery. It attempts directors resuming immediate responsibility
to indicate some of the reasons why standards for technical progress, accepting a feedback
of design in such work are not good enough from creators, and using this feedback as a
and calls attention to successes in other factor in setting policy. This type of change
fields such as aircraft design, nuclear plants, would do much to rectify the status situation
and the automobile industry, where design is for such creators as innovative designers.
regarded as of sufficient importance to be the
immediate concern of directors. Range of Design Functions
In the FEILDEN report (1963) it is sug- The term design covers many activities,
gested that the designer should be given ranging from detailing to innovative design, and
increased status and reward, but no Suggestion may even include some branches of research.
is made as to how to do this. Such gains are Table 28.1 shows in simplified form how these
frequently the result of self-advertising and activities interlock with other activities in
political teamwork. Good designers are often an engineering firm. In practice they have all
not interested in applying themselves this way been lumped together because they all depend
and become bitter and disaffected on realizing on the use of a drawin g board and eventually
their losses. lead to a detail drawing, despite the fact that
With the expansion of organizations they depend on skills as diverse as academic,
utilizing technology to make profit, control craft, or creative skill. The table shows the
moves from the technologist to the financier, range of work which is concerned with imme-
organizer, or accountant. These people make diate profit, and should come under day-to-day
efforts to select technologists but in recent management, and the range which involves
years, perhaps due to lack of self-confidence, policy and should be the cortcern of directors.
they have relied on the technologist to sell Such a division would enable the director
hirnself to them, and then have delegated, to profit in setting policy, and the creator to
with many restrictions, the work of maintaining improve his performance by knowing the
progress to the successful applicant. These director's problems. To bring innovative design
successful applicants are often organization under a director would shorten the lines of
men who have more in common with the people communication concerning hidden problems and
who accept them than they have wi th the potential solutions. These lines are often
typical technical creator. They often depend weak and faulty and are among the weakest
on the prestige use of qualifications and the links in modern industry.
selling of second-hand schemes rather than on Figure 28.1 shows a director's possible
the use of technology as a creative tool. approach to evaluating design staff. This
This trend is not confined to mechanical could develop into meetings called by the
engirteering, but is common in many industries director for his proved and potentially creative
259
staff, either as a group or indi vidually, at and probably achieve much more with well-
which they would learn of his problems, and understood traditional engineering techniques
they could contact him first by memorandum before making recourse to the invaluable but
and then personally in order to put forward expensive techniques of modern technology
potential developments. He would also leam (DA VIS, 1964).
of their difficul ties in acquiring non-engineering These group meetings between a perceptive
knowledge essential to designing, be able to director and his staff could do much to break
replace committees with essential consultation, down the barriers that have grown between
Table 28.1. Interaction of Design and Other Activities
Ile!Up- Ievels, factors, cootrols, skills, associated ectivities
Type Origin General need Cha11ge ia Change in

Example Chatacter of woric Objective
of design of need for change eoo.trol need skill need
Detail Shop drawillc Productioa Almost • cralt activity CommUDication Camplex and Leaner manage- More shop experience
needing production reeords, futu.re cosUy systems ment; more cost on production and
experience. serYice, liz:inr; need revision; c:onsciousness 1nspection
Answerahle to detail and 6xing time wastl!d on on both methods
plOduction m~~~;ager cost of mano- prestige appear- aad products;
facture ance of better incenb.ves
drawings; tDOte to draughtstt~en
ptacb.calmanu·
facturing know-
bow aeeded
Routine Structures of Customer Routine cralt work on Realization of More attention More critical Field e%perience
different size; ..d oales buis of technoloBY ud requiremeats on to users n~; appraisal of needed;
new models of experience. servioe aDd cost. better iastrumen· product llftd more perseverence
existin1
machines for
Answerabte to aeneral
manageme~~t
to meet
c:ompetition
tation, appear·
aoce, servicing
completeness
of the design
to complete the
desig:n
iu
diffetent sized and mechani-
products :ration
lmproving Speed-up of Soles, Partly routine using Progress, Determination \Iore technical More knowledge of
instruments, competitioo. engineerins expM:ience, quality, .ad rtOt to fall directors with users needs; I
mechanintioa aDd fonrard· technoloCY, some security by bebtnd real power; more engineerins I
I
thinkins by creativity and e2:perience meetinc awareness of more knowledce knowledge;
directors ofuse. Free to resch competition progress of users needs; i.naight and courace
l1
director to over-ride demand for
manapr improvetnent;
search for ability
and innovatlve
i.deas
Innovative
.
Completely
new project
,
teehniques
Oiaectof's
foresight;
director"s
aceeptar:w::e of
new ideas
Creative, b.sed oa
ezperience and
technoloCY
Desire for
progress, hiP
profit th10uP,
monopoly,
security, and
Setter commu·
nications
between ere--
ators and
authority; need
Allocallon of
money to un.pre-
dietable i.nn~
vation.s;
the search fcx
Access to more
information on usert:
and modern teeh-
niques; consultation
with other skllls;
.
'i:
~
frorn creative cood imase for nabonal innoV11tions and abdity to contact u
st.t'f or users awareness of innovators; and communicate to
resistance to the
I
importance of directors;
industrial build-up of high training in cteativity
II
prestige; more parasitic coats on and observation
awarenrss of budgets, estimates,
1mportance of manapment and
breadth of feasibility insight
thought and
attenbon to detail I
Developmeat
research
Aquisi tion of
data by mock·
ups; use of
advanced
transducers on
Designer,
director and
Advanced instrumentation;
mathematical skills to
solve complex pmblems,
consultant.
Acwrate
information to
ald creative
thousht
Lim.itation of
control of
f.cilities anc'
pmgrammes by
Appreciation of
worth and the
lirnitations of non·
creative techn~
1
More practical
traning;
more trauung tn use
of Imagination and I~
u
I
A.nswerable to ctirector non-creati ve lo&ists; observation;
production ot and desipet people, control understandins of less abstruse
p1lot plants; being sranted the needs of commttnicatiOil
mathematics; on s trength of creative designers technique <
computers; eu.m successes;
new scientific bettet efforts
diac:oveties to feed COtlpt'-
tent but non-
creative techßOoo
logists with work
I
progr•uots of
imporUnce aad
needi.ag their
lmowled&e
= Constder tf suttable on produettan-not

for drawing offic•
D•tatl draughtsman or
routine deostg~r
Us•ful t•chnologtst to
•ld tnnovattv• d•stgner
Personnet up to 42 years of age, 1 e post war
Part-tome. e.ducatoon- HN C or Ctty and GUIId

I
I
Wants a Expresses •deas Uses h•s
safe job, as good sk•tches technology
Not H N C Ievel
hopes to I as a formula
Draws unexpected for solvtng
be a
manager opmions about problems
tundamentals,
Fallure LPft shop Wast•d Came up Interest Pollttcat
leildmg to
j_Out' r•queosts for tn maths for a early y..ars from ot type,
I
respectable through shop practtcal destres
research or
Must not control
projects
JOO tn Iack of late nature. power
destgn drawtng QUidance alter must ·see' _L
ls he good off1ce changmg cannot
? [f hts tdeas
at graphoc
methods, _L 1deas
~·
1mag1ne
::L
Out 1
are sound can he

A good technologtst /
wtth team sptnt but
teach htm thtnk ' öUt•
destgn under
has no orrgmal tdeoas
a creator Exam1ne for
Re.gards ortgmal 1deas
Regards H NC Interest '" Tryong to
Machinery Try under H N C as professional and powers
a good
as a tool and creoate
research key to probably takong of observatton
tnstttuttons
under a destgner, safe JOb post-graduale I
creator
_L
moght
aevetop j_ courses !f for
power
I
Good?
I
None
:::::=
Out! th~n
~
try to
Des•re to Oestre for power educate
create f lf not' creafor
!I creator .J.... Out 1
Further educatton, conferMces, vtsits, process
proJects of mcreastng cost as abtl1ty grows and of
Encouragement to suggest own tnnovattons, and Travet a,broad to

mnovators to carry proJect through and recetve othe-r compan.es
the credtt J I
!f id~as
1wtld or If tdeas baS@d on observat•on and
based on wishful creattve cross- ferttltzatton of
thonktn~ ideas -even more responstbtltty
on creatrve werk
Otrect .contact wtth board as weil as techmcal director -mentton '" te-chntcal batance
sheet -will Iead to recogn1t1on of worth tn salary -feed dtrectors wtth advanced
tdeas for tuture
Figure 28.1. Suggested survey of drawing office staff by technical director
graduate and non-graduate technical personnel. less now than formerly owing to the change in
These barriers spring from the fact that a the nature of management, and optimism about
graduate often has had a more expensive current and pending changes in technical
(university) education and expects, and often education. Table 28.2 shows how incentives
receives, a greater reward regardless of his differ between creators, craftsmen, and labourers.
effective use. The non-graduate resents this, The behaviour of creative people and the
having probably been promoted to a creative methods of management necessary for getting
post by dedication and successes in this field the best results from them and for developing
rather than on qualification. The importance of potential creators have recently been discussed
the powers of observation and imagination in practical detail by McPHERSON (1965).
developed during extensive practical experience A good creator questions existing practices,
and the value of these qualities to design teams looks for weaknesses and alternatives, and
is not always adequately recognized, possibly has a pugnacious and self-assured temperament.
These qualities are the opposite of those of can apply the technique of involvement, which
the organization man so readily accepted by used to be known as leadership. Companies
day-to-day management. A firm which wishes which succeed without the apparent application
to be progressive must have at least one of any of these techniques will probably be
member on the board able to over-ride confound to operate with a considerable hidden
ventional management attitudes to the creative involvement.
mind and who understands creativity and The word involvement rather than leader-
continually searches for it. ship is used, as the latter implies that labour
follows a Ieader, whereas in fact management
The Management of Design involves the labour in the objective to such an
Table 28.2 also shows how the various extent that it drives itself to get on with the
levels of design activity fit into the pattern of job and largely do its own thinking. It is an
management. In considering this it is worth- application of the fundamental desire of man-
while to remernher the structure of the average kind to be creative (FRIEDMANN, 1961;
company. The chairman of the board is the DAVIS, 1964). A use of this desire was re-
principal representative of the shareholders ported recently by a French company which
and is concerned with the financial well-being reversed the trend of breaking down assembly
of the company and with the choice of the of computers and instead permitted a single
managing director who is to carry out the man to complete an entire unit and attach his
board's policy. The directors of a board are name to it. Not only did quality improve, but
concerned with fixing the company's policy a remarkable interest in advanced technical
and may include specialists in various fields. education was induced in the people so
As they are concerned with policy, and inno- employed.
vation and innovative design may have a pro- In another case, the managing director of
found effect on the company's future, these a large company which used much machinery,
should be the concern of a specialist director. exploited his skill as an engineer to set up
It is probable that the full effects of the his own engineering development unit. He
director's activities on policy will not be rejected suggestions that a bonus system be
realized until up to seven years after many introduced, on the grounds that the pay was
decisions have been taken. Innovative design reasonable and that the work was of such
trends often have this delay before their effects great interest that good leadership should
can be measured. The managing director dele- provide all the incentive necessary. The unit
gates much of his work to a team of managers, expanded to forty-five craftsmen and twelve
advisers and experts, such as accountants, draughtsmen or designers. On the managing
work-study men, technologists, and sales director's death his successor, who was an
managers. The effect of their activities may accountant and new to the company, continued
frequently be measured in up to two succes.sive with the projects already in hand and, on com-
balance sheets. J AQUES (1956) considers pletion of the main one, referred to the unit as
methods of gauging the time for corrective the place where miracles were achieved. Then,
action by immediate authority. in an attempt to improve the unit, he started an
The work of such experts is largely incentive scheme accompanied by modern
concerned with the organization of facilities, organizational methods under the control of
labour and methods, to effect the making of an outside industrial consultant. The consultant
immediate profit; to do this various induce- installed an organizing expert in place of the
ments are used. These include fears of losing existing manager, who was a creative engineer.
the job, of proving inadequate, of losing The unit soon looked smarter, and issued well-
prestige and security and fear of criticism; produced folders on suggested projects showing
there are incentives of promotion, increased incentives, completion dates and budgets
earnings through bonuses, and personal credit (which were usually subsequently over-run).
with increased security. Finally, the experts However, after four years it was decided to
Table 28.2. The Management of Design
Levels of management- innovation and some production only
Type Character Supervision Current Management

Eu.mple Applicable incentive Supervision
of labour of work other than men weaknesses bias required
General la'oour Job is codified Line productioa Pressure of oncoming work; Line foteman to Report slackness Lack ol plannin&. Accouutancy;
fear of loss of job and productioa manager of discipline; of best equipment, prodoc:tion engineer-
security; piece money and stop for faults; and of technical IIDd in&.; experience
wage rate; personaUty of foresee stop- practical know·how
supervisor pages
'3
J~I
~
Skilied lobour, All methods Machia.e.tool lnterest, promotion Foreman to Choice of right Technical Practical ezperience;
routine work known andin fitting; detail prospects, security, piece
i!I
managet ma.~ for job; education and simple direct control ll
· regular use drafting money and waee rate; a work method, tool faciliües orp.nizer;
small h.cklog of work;
5e tools, supplies, human attitude;
"2 personality of supervisor; quality and accountancy .!!
11
~ Co--operation with fellows discipline
!I Ii:
-8§ Skilied lobour Unexpected Complicated Inte.est, involvement, Foreman to Observation and More technical Practical experience
on novel work problems repair work; bonuses, security, managet, or assessment of knowledp, need· coupled with enoudt
~
I
with random
problems
using known
sk.ills
pilot or proto-
type work;
prestige, promotion,
communication of
deftlopmeot
en&ineer to
faults; devising
work programme;
ing continual re-
assessment of
technical knowledge;..
1)1 >
Cl
.9 awareness of passing
;; simple design; objective; team spirit; director disci pline and progress to pre- of time and growing t<l
~ manufacturing desire to improve ability; quality; creative vent drifting with costs; flexible and Ii:
dependent on personality of Supervisor; approach; drive; not much papet· observant organi· t<l
"' poorly under· wage rate commurtication work; jobs often zation; choice and z
stood princi· upwards drift and progress trairting of men ~
ples witb poor data eilher mean-
I
instruments and ~
ingless or i&OOted; u >
variable vested in terests 'E z
material J! tl
Skilled labout Routine work Design; Involvement, prestigeo, By-passes Development; Bi& paraaitic oYer· Creotivity, technicol tl
with with problem:s experimental salary Ievel, security, geDeral manace-- assessmeot ol heads; directors insight, crystal ball t<l
technical demanding ! rJ)
machine increased experienoe, ment; engineer successor besitate to invest gazing, practical ....
education initiative or developments; credit; courses of to technical failure; in progess; experience, cost Cl
i, training technological education and visits; director or Observation of poor communi- consciousness, z
~ investigation contact with top manag,i.ng director technical cation; misuse of cowaee and I
~
.~ executives; team spirit methods;

personal inter·
labour dedicatioo
~
1
~ est in achieve-- '3u
i ments of staff;
"'a communications 1
up and down I
1 ~
!I
Innovative Observing, New tecbniques, Prestige, credit, flattery Direct to director Set company Too many non-- Creotive of product. l!
creators thinkinc and new machines, of ego; enough money to for authorization, policy; search creators gettinc ond poliey;
f creating; ne:w products; be free of family worty, knowledge of for creators and control of focili- ezperieace of I
~ feeding tracinc cause security; freedax~ to problems and innovations; ties by virtue of customer'a work;
§ directors with of failures choose own approach and display of ideas; estimate qualilications or open-mindedneoo;
possibilities; some selection of work generat mana~· feasibility; polltical activity; courop Ollj! iDsl&bt
llJ
I
solving coot.cts; direct COiltact
>.
problems with authorizing
ment to have
nacging powers on
finance develop-
ment; to sell
crowth of • .",..
creaü ve techno--
illto the future sE •
~
" authority; travel of discipline and to suocesses lol)'; poor COfD.o.
biasiness; fac:::ilitiea to &et wark dooe but mwücatiou; bic
leam; possible pro- oot to hinder or overheada of noa-
motioe to clirector eriti~ae pi'Ojecta techa.ical services;
Ievel ia.hadorp~ failure to iavest
~ed iD propess ~
w
close the unit down and expand another small problem's root. This approach is independent
engineering unit to do some of the work. Three of technical knowledge, but technical know-
years later still, a new development unit was ledge in addition to clear and direct thinking
started with some of the original labour, but will be required to interpret what has been
the old spirit of involvement was not recovered. seen or postulated.
Today, the projects started by the old engineer- A simple example of this directness
ing unit are amongst the most profitable carried occurred recently when a member of a workshop
out by the company. group overfilled his lighter and discussion
It is interesting to note that the engineer arose on why it would then not work. A graduate
who started the original development unit did who had joined the group at tea-break didacti-
not bother with any great degree of accounting, cally announced that failure was due to an
but he fixed his targets, the size of his unit over-rich mixture refusing to fire in the same
and the rates of pay; he kept down the over- way that an over-choked car will not start.
heads and drove the uni t hard. He never had to A fitter, after some thought, tentatively
drop a project because expenditure was getting commented that you could ignite the lighter
out of hand. At the end of the year he could with a match. Could it be that with so much
easily see what had been spent and what he wet petrol the spark was quenched before any
had earned for the expenditure. region reached a temperature high enough to
The director in charge of innovative design ignite? Here the fitter had collected some
must remernher that the vital objective is good relevant information and used his imagination
design, and that systems of organization and on it.
management are only there to assist this and Graduates who lack early practical ex-
reduce costs. He should hirnself have reliable perience quite often fail to examine the
opinions on feasi bility. He must be aware of practical details of a present situation. This
the fact that if management procedures become may be due to the use of a syllabus-based
the most important objective of the development examination for the choice of people to be
unit, they will carry the most prestige and trained to graduate level. There is a lot to be
distract good designers from their vital said for a double filter in the choice of
business. The greatest prestige must go to the graduates: one part being based on class work
designer who does the most difficult and and the other upon examination in which the
important job. Morale will be good if the questions are both hypothetical and highly
auxiliary organization is arranged to help the im probable.
designer rather than hinder him. A similar weakness is that of failing to
The technique of involvement depends on examine all factors and then to apply well-
fastering the feeling that initiative is expected known techniques in a novel manner before
from everyone; that credit without any dis- turning to some modern miracle method which
tinction regarding background will be granted is not yet fully understood. It is true that many
in full to those showing initiative; and that the modern developments are not used readily
initiator will be involved all the way in carry- enough; but people have, for example suggested
ing his ideas to fruition. In applying this the use of infra-red, dielectric or microwave
technique, it is necessary to know the people heating as answers to a drying problem, without
involved weil, to look for creativity and to knowledge of the absorption characteristics
develop it, and to have everyone really believe concemed, and without having studied the
that the unit does operate in this way. possibility of overcoming the difficulties by
The successful innovator is the man who using cheap steam available as a by-product
has made a habit of trying to solve problems of power generation. This resort to the use of
other than those thrust upon him: he finds incompletely digested technology as a formula
problems to solve if none appear to be avail- for solving problems, instead of applying a
able. Experience develops in him a peculiarly creative approach coupled with a systematic
direct approach to the vital factors at the examination of alternatives, is much too
common; graduates are possibly more prone proposal, and should have a wide experience
to it than HNC engineers. of the background, current and past practices,
Directors often do not know the extent of and materials, of the processes concerned, in
the weaknesses in the machinery they employ, addition to the necessary pure engineering
nor how much could be gained by redesigning techniques. In many cases this demands a long
such machines. This is usually only possible association with the process, best obtained by
to people with the requisite technical know- non-productive service as part of an operating
ledge, and with the opportunity to take part in team under the eye of a creator who is looking
the running of such machines. The easier a to his company's future.
project is to understand the more likely it is Such directors also need clearly defined
to gain acceptance. Thus, non-technical resources to support work they think important.
managers find it is easier to organize a small There should be money, staff paid from an
store containing a static stock of no great overhead fund, and other facilities, sufficient
value, than to deal with the improvements to enable the director to follow up any idea
which might be effected at similar cost to thought likely to lead somewhere without
a process plant, although with much greater having to make a detailed case of where he · is
potential return. This tendency to deal with going with estimates of costs and incentives.
the understandable and petty rather than the Day-to-day management should not be able to
difficult and important has been alluded to by hinder him or stop him when he thinks he has
PARKINSON (1958). found a trend which will lead to progress, or
The task is to perceive possibilities, both when he is investigating what is actually
in situations and in men. It is usually better going on in a process. This is a situation
to rely on a proven creator to perceive situ- which may be pregnant with difficulty. He
ations than to depend upon a committee. Such should be answerable to the chairman of the
proven creators will surface if they are at first board whose concern should be how much the
encouraged by a congenial atmosphere, the company can afford to spend to ensure its
assurance of credit, and facilities to carry technical future. Enlightened production
through their early ideas under their own mariagements are entitled to press for develop-
control. They can be given increased responsi- ment to be clone in the light of their own
bility as their score of successes rises. experience, but not to hinder the development
Creators, too, have a facility for recognizing department from following its own ideas.
other creators. If such men with the requisite
experience and character are raised to director Possible Future Aids to Design
Ievel, they will recognize and develop other More than propaganda will be needed to
creators. Such a build-up will also require a convince many boards of directors that they
tradition of continuity of service with a should make the best use of design in their
company, and will need non-creators to be organizations. Even legislation might be
fitted into service teams while creators are required to help, and there are indications that
given freedom to follow their particular bent the government might support this if forward-
as far as possible. thinking directors would formulate an appro-
Directors in charge of a company's priate scheme.
technical policy must have qualities other than My own approach as an engineer would
pure creativity. They must be open-minded and be to examine the entire system to find out if
interested in people; they must have a direct there are any factors which could be introduced
approach to the root of problems, a great or used in a different way to change the
breadth of knowledge, and a clear under- operation of the system in the direction re-
standing of the fundamentals of techniques and quired. I offer this opinion with as much
principles rather than a specialized knowledge trepidation as a production engineer should
of application. They must be quick to observe have in suggesting a major machine change to
and build upon all the implications of a a designer. One such factor which does not
seem to be used to the full to forward the how he had to learn the successive theories
cause of design is the shareholder. The share- of flight as the areas of interest changed: thus
holder is interested in profit and a successful he passed from the lighter-than-air field to the
innovation can give high profit margins for a heavier-than-air field, from subsonic, through
period. It costs money in the early stages to supersonic to hypersonic.
forward an innovation, and profit may not This receptive ability should be con-
materialize for up to seven years after invest- sidered in forming postgraduate courses. These
ment in an innovation has started. This money courses should encourage the cross-fertili-
must come from current profit, tending to zation of ideas between designers, users of
reduce the price of shares in the company machinery, and purchasers of machinery. The
concerned if the shareholder decides to sell first move would be to encourage users to
his holding. This loss in share price leaves employ more qualified and suitably trained
the company exposed to takeover bids from engineers. These men would feed back criticism
people who know the financial position of the of designs to manufacturers, a service that is
company and who know that by dropping the needed in Britain; at the same time they would
investment in innovation they can raise the suggest new ideas. This recognition of the
dividend and gain an immediate profit. If importance of engineers in industry is more
immediate profit is their objective they are not common abroad, and could be copied here with
interested in the future of the company over a advantage. I have worked as chief engineer to
long term. A change in the structure of the new a group of small companies under continental
capital ~ins tax might improve this aspect of employers in this country, and was surprised
the working of the system. at the high status of the post, the importance
If each individual field of industry had attached to my opinions, and the weakness of
the annual capital gains of the companies an HNC alone as aid.
working in this field averaged to provide a Comprehensive post-HNC or equivalent
capital gains index, and if shareholders who part-time courses should be available to train
had held a share for five years were entitled men for the specialized posts of chief engineer,
to reduce the tax payable by the sum of gains designer, machinery research and investigation
made annually above the capital gains index, engineer, production engineer, and sales
they would have a real interest in holding on engineer. Creative minds would grasp the
to the shares of companies investing in the importance of the mastery of several of these
not-too-distant future. This would help to courses, the importance of combining practical
stabilize share prices. Shares would not have experience with design ability, and the im-
to be sold to realize the gains, the bonus for portance and sheer interest of continually
holding accumulating each year the company increasing one's span of knowledge.
made a gain in excess of the index.
Interested companies might publish a
technical balance sheet dealing with successful Conclusion
achievement and current spending, perhaps Engineering innovation and design is not a
mentioning successful innovators by name if matter for day-to-day management but is a
they were pursuing new projects for the matter of company policy and long-term survival,
company. and should therefore be very strongly- repre-
Another possible way of encouraging sented on the board of directors. The achieve-
creators is by way of life-long education. ment of innovation successes is a long-term
Contrary to some professional opinion that the business and ranks in importance with growth
mind loses its receptive ability with age, policy, finance · policy, and company image.
creative minds seem to retain this ability, at Its problems are so different from those of
least in their creative field. An interesting day-to-day management that it must be inde-
example of this was given by Dr. Barnes pendent of this function. Too many companies
Wallis at a recent conference. He described regard innovation as a short-term activity,
MANAGEMENT AND DESffiN 267
capable of having its return predicted accurate- encouragment of diverse experience, and a
ly, whereas much of it is a trend activity with rational approach to examinations. He will
a new feasibility materializing out of the encourage independence and openness, and
successful realization of the previous inno- prepare for the arrival of those ideas which
vation. Such progress comes from the visuali- have yet tobe formulated (DA VIS, 1963).
zation which JOlfis functions and makes The importance of design in a company's
balanced compromises in a single dedicated future cannot be too highly stressed. The
mind: teamwork is needed once the innovation shareholder is the final arbiter on many of the
is visualized. A committee might control the matters concerning the future of a company,
teamwork needed for the execution, but the and it is important that he should know of this
creators should be answerable to a single importance of design and should demand some
man responsible for the specific policy. means of measuring how the board of directors
This man should be a genuine creator of which he supports is handling this factor.
high calibre, not a good organizer or a salesman The quality of engineers will improve under a
of ideas; he should be answerable to the leadership which produces involvement in a
ahairman of the board and be in very close company's business. Since so many criticisms
contact with his creative staff. This close have been made of our engineers by directors,
contact will do much to implement the frequent it is surprising that more thought has not been
suggestion that the designer's status should given to the part that directors may play.
be raised. This chapter is based, like most engineer-
Through his knowledge of the ways of the ing design, on the facts of experience and
creator, he ~ill appreciate and assist the work personal interpretation of these facts. There
of his team. This will need the adequate supply seems to be little of an objective character so
of information, facilities for visiting customers far published on the subject and a case can
and places where work of interest is carried be made for research studies on the proposals
out, the provision of specialist courses, the given, in so far as this may be possible.
Chapter 29
EFFECT OF ORGANIZATIONAL PROCEDURES ON DESIGN

AN OUTLINE OF THE PROBLEMS
I.M. Ross
lntrodiJction human designer must play and how bis skills

The last few years have seen the begin- are best acquired, developed and exercised
ning of a shift in emphasis in the study of must also be known. As part of this, the extent
design - from the end product as such and the to which he is influenced in bis work by the
components, materials and knowledge which organization and environment in which it takes
come together in the evolution of its form, place must be known, and it is this aspect
to the process itself and the methods of reason- which is the subject of this chapter.
ing and imagination by which the form is con- In looking at aspects of design which
ceived and refined. Many factors may have have tended to be taken for granted in the past,
contributed to this change: the increasing it will be desirable to adopt an attitude of
complexity of techniques and function and the doubt and suspicion: to refuse to accept any
widening and sophistication of user require- notion, other than very tentatively, until it has
ments are forcing the use of a more systematic been shown to be not only internally self
approach; the low standard of much recent consistent, but to key into the evolving pattem
design is focusing attention on the decision of knowledge as a whole. A final understanding
processes which may have led to this outcome; is going to require the contribution of many
the entry of the computer offers the possibility disciplines, particularly from the social
of automatic design in suitable cases, but to sciences, and is not something I could hope to
do this the process has to be broken down into attempt. Rather, this chapter is written from
its bare essentials. There has also arisen a the point of view of one with a fair amount of
feeling that the technological tower of Babel practicar experience in the conduct of design,
created by over-specialization in the pursuit a consciousness of some of the problems and
of facts is probably inimical to good design, an instinct to question; but without the detailed
and this is leading to a search for generalized and documented knowledge required to provide
methods of training which will emphasize and a firm framework of understanding. Where I do
develop the common features of the design suggest answers, it is more with the intention
process, even if it means relying more on of illustrating the form these may take and in
machine methods of information storage and the hope of stimulating others, better equipped,
retrieval to provide the detailed knowledge, to rectify and complete the picture.
experience and philosophy required.
Thus a phase has been entered in which
design, as design, is a subject for study. The The Nature of Procedures
goals of such a study must be to understand Procedures are the rules by which the
the nature of the design process and how its various processes in an organization are
various elements can be developed and assem- sequenced and interlinked, and which promote
bled, motivated and controlled to give the and regulate the flow of activity and determine
greatest overall benefit. The role that the under what conditions and in what form it takes
269
place. The rules are based on some precon- fluence the process in the manner laid down.
ception of what is best, modified similarly The influence signals may be of different
from time to time in the light of experience. patterns. They may depend only on M, with
They may be explicit, or merely implied by the dependence changing infrequently in discrete
existence of other rules, or they may take the steps as procedures are reviewed; they may
form of convention, tradition, or be purely have a fluctuating component superimposed,
psychological. It is desirable to know what
procedures are in use for the processes in- lnlets
volved in design, what effect these may have

on the output, and whether (and if so, how)
some degree of optimization may be obtained.
Particular processes may show peculiar
sensitivity to procedures and it may well be
that design is of this nature, but first it is
Outlets
advisable to look at the problern in a more
general way, narrowing it down later when Figure 29.1. Network of elementary
there is a clearer idea of what to look for. processes characterizing design
It is also best at first not to be too pedantic as
to which influences should be called 'procedure' for example as the environment changes; or
and which by some other name (but distinguish- they may be experimental or adaptive following
ing where necessary between influences and the some procedure laid down at a higher strategic
processes as such). It will be sufficient to level.
distinguish different classes at a later stage If the process were purely mechanistic,
when it is clearer what type of classification the network problern would be that of process
best suits the problem. control and hardly relevant to the present
Consistent with this general approach, consideration. However, since people are
environmental influences can, for the present involved in the process, there is both a work-
purpose, be regarded as forming a sub-class of study or ergonomic problern of treating people
procedures. This is probably reasonable, as machines, and a motivational one concerned
since they certainly represent the outcome, with the psychological aspects. It is the latter
extension, or manifestation of procedures which may be the more important, since it is
'written' in the form of bricks, steel and glass, well-accepted from general experience that
patterns of light and shade, intellectual, social enthusiasm can make tolerable a considerable
and psychological background, and the like. degree of apparent inefficiency or confusion,
They are certainly not processes as such, whereas the most perfect system, on paper,
but may have a profound effect on them. may easily be wrecked if it achieves only
At a later stage of study, it will be perfunctory, or even hostile, compliance.
desirable to know what processes specifically In essence these effects arise because the
characterize design. For the present, one can operation of a machine is an end in itself,
think in terms of a network of elementary whereas to a man work is merely a means to
processes of the form symbolized in Figure29.1; some entirely different end. Any procedural
there is a basic process (complete with storage control which thinks only in terms of adjusting
capability), two inlets to allow blending, two the means, forgetful of the effect on the man's
outlets to allow switching, and a pair each of aspirations, is not likely to be very successful *.
supply and signal links. The latter provide
monitor signals M outwards, by which the
state of the process (and in total the network * The nearest simple mechanical analogue of man
in this respect is probably the gyroscope, which
of processes) is determined. Knowledge of the also has its own extemal 'aspirations'. Notlee the
state is conveyed to a procedural box (not cudous, resistant, oblique behaviour under direct
shown) and this issues signals I which in- applled constraint.
This is probably particularly true in the design desire for security is, of course, a direct
field because of the high aspiration Ievel of surface manifestation of the survival instinct;
designers as a class and their vocational happiness is connected with the temporary
ability to sense the presence, and foresee the achievement of security, or rather with the
implications, of constraints. temporary avoidance of danger and particularly
Thus, in looking at procedures and their with the accomplishment of something which
effectiveness, particularly those concerned may have this effect.
with design processes, it seems especially In addition to these four goals, the problem-
important to Iook out for psychological solving nature of man's existence should be
difficulties such as may reveal themselves by noted. He is like a computer which is forever
low morale, Iack of enthusiasm, care. or origi- seeking the best solution, up-dating the answer
nality, or in difficulty in holding or recruiting continuously as his experience evolves and his
staff. To this Iist other diagnostic factors may environment changes. Because of the importance
be added when the problern has been more of this process, he likes to use every opportu-
carefully considered. Important to this approach nity to practise - hence, perhaps, his interest
also, will be a knowledge of the aspirations in games of skill, his attraction to puzzles,
or motivating forces of the typical designer. his acceptance of challenge. Linked with this
This is a problern for the social scientist, is the competi ti ve instinct, which has also a
but for the present purpose and to show in fairly obvious direct connection with survival
more detail the implications of such aspects, under conditions of limited resourc:es.
some observations from general experience Of these forces and instincts, the desire
may be useful. for security would be the most powerful were
it seen to be really imperilled in the near
Man's Motivationsand Moods future, but under modern conditions and in the
It is suggested very tentati vely, that the present context it is probably a lot less power-
significant forces which on the surface motivate ful than the desire for respect, including self-
a man are directed to the pursuit of esteem, respect. The other two forces are probably
security, knowledge and happiness. Possibly still less powerful and may need a rather
these could be traced back to deeper forces sophisticated appreciation to have an important
and perhaps to a single force - the wish to effect, though they should not be neglected
survive and the pleasure associated with in dealing with the higher levels of intelligence.
incremental achievement of this. For example, The competitive instinct remains a powerful
esteem, which is the degree of respect in factor.
which he is held by his fellow men (and self- Before discussing the importance of such
esteem, by hirnself on . their behalf), may well points in consideration of procedures, it should
be related to the primitive need to keep in be noted that man has usually had a choice of
with the herd if he was to survive: to show the groups he might aspire to join, and that in
qualities entitling him to membership, in its modern form he can even be simultaneously
particular willingness and ability to help, a member of a number, for instance his family,
ability to Iead in a direction thought to be his friends, his neighbourhood, club, union
safe, no unstable tendencies such as might (or institute), firm, creed, culture, country,
endanger the group safety, but always allowing etc. His attachment or alignment to any par-
the possibility of his being able to convince ticular one will be related to the benefits he
the group about some other course of action. visualizes, which will depend on view-span,
Similarly, the desire for knowledge is doubtless and to also the extent to which it may imperil
connected with its importance for survival - a his membership of another to which he attaches
curious object may contain da~ger and need importance. To illustrate the significance of
investigation - and more generally, knowledge this, a procedure which effectively prohibits
is the counter intelligence against nature on publication of his work to a man who attaches
which man bases his strategy of life. The importance to his cultural group, is likely to
weaken his alignment with the organizational The sub-headings chosen are more for con-
goals or, put another way, his 'specific co- venience of paragraphing, that with the intention
herent activity' for the organization falls. Fear of systemalle mapping.
of security may maintain a minimum alignment,
or where this is not effective - or is only Personal Factars
negatively effective - he may be powerfully Freedom - Man has an instinct for any-
influenced to leave. thing that might limit his freedom of action or
In this example the motivational aspect trap him in any way, and since procedures are
is fairly obvious, but frequently the effect of almost by definition of a restricti ve nature, his
procedures seems to be less directly con- first reaction is one of suspicion and antipathy.
nected with an end result. Instead, they set up He will comply if he has to, but it will be with
a mood, spirit, or state, a kind of mode of bad grace and he will be alert for the defici-
operation of the mind, which colours or in- encies, will be readily confused by the obscuri-
fluences the approach to problems whilst it ties, and quick to notice and experience the
lasts. The mode may be either harmful or absurdities. The di"afting and introduction
beneficial to the organization. For instance, a therefore require care: the importance should
procedure may create a resentful mood, likely be clear; the restrictive element should be
to be harmful; equally another may create a confined to what is essential; there should be
competiti ve spirit, possibly with good results. user participation in the final drafting,
Still more generally, a procedure may create a particularly on 'sensitive' detail; a controlled
state of tension in the mind, a focusing of 'leak' should be incorporated to cover the
attention on to a specific end point, which may absurd case and so on.
either be beneficial or harmful, and contrariwise Self-respect - This powerful factor is
another may confuse, or bewilder, or possibly invoked to some extent by all procedures, and
soothe. strongly by those relating to status, status
It would be presumptuous for me to attempt symbols, salary and salary scales, promotion,
to disentangle such considerations and set side benefits, trust and responsibility, and in
them out in a coherent way: this is something particular decision-making. It is also involved
which is going to require professional attention. in questions of publishing and exhibiting his
But it may be useful, as a temporary bridging work, when it can easily over-ride the import-
operation between the realm of the mind and ance of commercial secrecy. It is usefully
the practical world of design, to discuss the increased by procedures which ensure that
interaction briefly through a set of terms such good work is noticed and credit gi ven when
as those now introduced. This set makes no deserved.
pretence to provide a logical framework, The fear of loss of respect is an important
although this is something that must be hoped negative factor if the work is very challenging
for in due course. or difficult. It is allayed by good facilities,
wide experience, support, co-operation, friend-
Factors having a Psychological Content ship, understanding and the sharing of risks.
lnvolved in or Conditioned by Procedures Progress - A man likes to feel that he is
With the reservations of the preceding making some progress, that his work is being
paragraph, the following factors are suggested. used and appreciated, and that he is increasing
In order to give some semblance of order his potential for the future. He will therefore
these are broadly grouped into three classes; be interested in the width and variety of ex-
perience he is gaining, as well as in the depth
(1) Personal factors- freedom, self-respect, or special knowledge involved. He will want
progress, creati vity. to see also that the possibility of advance
(2) Task factors - worthwhileness, diffi- exists, and that he is being given the opportu-
culty, methods of dri ving. nity to acquire the background necessary to
(3) Organizational factors confusion, take advantage of this. Such points will be
politics, frustration, morale. involved in procedures relating to training,
further education, transfer within the organi- change their field periodically (after three or
zation, staff grading, recruiting policy, and so four years, say). Some of this change can be
on. If the prospects look too stagnant he may hierarchal, for example component-equipment
leave rather than complain, so it is important system or vice-versa, some 'product' class and
to see things through his eyes and sense the some a gradual change in specialization. There
danger in advance. are reasons to believe also that a cycling
Creativity - This is a very important between research and application is desirable,
factor in design and needs special discussi6n, in givingfull exercise to the naturalcapabilities
particularly since it is likely that many current of the human mind. Such a change in method
procedures tend to inhibit rather than promote may also have other advantages: it allows
this faculty. There is a widely held view that better assessment of ability; it provides
this is a gift gi ven only to a few at birth, and practice in the approach to new problems,
that the best one can do is to find such people forcing a more systematic approach; it provides
and then pander to them. In fact, there is width of knowledge and experience which is
plenty of evidence that this is far from true, very desirable for rapid insight; it helps to
that originality is at its best under dire neces- break down petty fashions and snobberies
sity (the 'mother of invention '), that it is quite between jobs.
amenable to training and experience, and that It is probably easier for a team than for an
the initial difference between one person and individual to innovate, because of its greater
another is much less than might appear. width of knowledge, its greater confidence and
A man joining a group is motivated to its relative resistance to psychological pres-
justify his membership and to show his worth. sure. To be effecti ve, however, it must be
With his fresh point of view there will be a carefully composed to embrace the experience
tendency to innovate. On the other hand, the and skill required, and must be trained as a
group will tend to regard innovation with team. To maintain its creativity there should
suspicion - will it usurp or destroy? If he probably also be a slow drift in field, and a
comes in with special knowledge not available slow replacement of individual members. These
to the group it will be easier for him - he can problems will require study, particularly with
use this as the core of his innovation. It will the importance of the team method in the
be especially easy if his confidence has b~en design for automation, a characteristic of which
built up by the excellence or aura of his trai·n- is likely to be short periods of intensive and
ing, and particularly by having worked for a highly specialized development, followed by
time with a great innovator, watched his long runs in production.
methods, seen his human side ·and his failings, A major deterrent to creativity is probably
and lost the awe and feeling of inferiority the type· of procedure which requires a strong
which he might otherwise have had. case to be made before any work can be started
Later when he becomes established. in the on a project. A new idea often contains an
group the need to innovate becomes less, and unknown element and even a trace of the
he may in fact find it psychologically difficult ridiculous, points which will only be resol ved
to do so, unless he has meanwhile become later under the intense challenge of the job.
accepted as a leader or perhaps as a hannless Equally discoursging is the knowledge that a
rebel. It will then be better for him to move if new idea is going to be under close scrutiny
he wishes to retain his creativity. This will before it has had a chance to develop, or, in
create fresh challenge and a further need to the opposite extreme, that it is going to be
leam. given too much importance too soon. An atmo-
The lessons for procedure would seem to sphere of faith, trust and understanding is
be the need to move people araund in such a required, with gentle pressure to innovate
way that they work closely for a time with the rather than not to do so.
best designers; the need to avoid narrow speci- Mention should be made also of the possi-
fications in recruiting; the need for staff to bly inhibiting effect of 'checking the literature'
before beginning to think creatively about a if it is known that all reasonable help has been
project. Although this may be good work-study provided.
practice there is the danger that it may channel
the mind into a groove which then becomes Drive - One of the functions of procedure
difficult to leave. A fresh approach is probably is to set the pace - to get the work flowing
preferable, leaving the study of what others smoothly at the maximum practicable rate.
have clone until later, when ideas have had However, man does not like being monitored
time to flow and take shape, although this may and, still less, being pushed. He tends to react
mean some back-tracking on occasions. By this in such a way that the effect of such methods
method, the Iiterature will also be found more is self cancelling. But he can be inspired and
meaningful. drawn on by good leadership, the essence of
Finally, it is probable that the bottling-up which is a sense of purpose and direction,
of ideas which occurs when commercial or strength of character, good tactical jl1'dgment,
national security procedures are applied, fails an almost sacrificial sharing of risks and
to make room for other fresh ideas. Publication, rewards, and an ability to let each man feel
apart from being a powerful stimulus, both inwardly the importance of his own contribution.
frees the mind and creates a vacuum to be This has a remote, but none the less powerful,
filled by fresh thinking. There may be many procedural aspect in so far as methods of staff
other detailed environmental factors having an selection, training and assessment determine
influence on creati vity which will emerge on in the long run the leadership structure.
closer study. On a different plane is the quest~on of
direct incentives - .bonuses, profit-sharing and
Task Factars the like. These can be strong short-term stimu-
The Worthwhileness of the Task - For a lants (and sometimes sedatives), but they are
man to give his maximum effort to a job he difficult to apply fairly and in the long run may
must be able to appreciate its part in the generate as much resentment as satisfaction:
scheme of things and understand how his own a carrot in front may be a good stimulus for a
contribution is helping towards this. Pro- donkey, but how long is it going to be effective
cedurally, this means that he must not just be for a thinking man? If there is to be any
fed with information technically sufficient for feedback of extra benefits, it should probably
the purpose; he must be given sufficient in- err on the side of liberality, and uniformity,
formation on background, progress, costs, etc. rather than consist of a few awards to selected
for him to take a live interest in the project people.
and in his own contribution towards it. Not A man is at his best when he is driving
every job can have the same excitement, of hirnself - the volunteer, rather than the con-
course, but the procedures should ensure as script. The aim of pacing procedures must be
far as possible that each man gets his share to create this condition, or rather allow the
of those which are. This will react on the condition to occur spontaneously, since any
methods of loading, of staff transfer and so on. artificial forcing or cunning will be self defeat-
The Difficulty of the Task - Just as .the ing in the long run. There must be the maximum
job should be exciting there should also be a giving of trust and responsibility and the
reasonable degree of difficulty attached. Not encouraging of a man to commit himself, but not
too much, or fear of failure will inhibit his of course in any rash way, with full under-
whole approach, but sufficient to form a strong standing of the need for unobtrusive help and
challenge, and yet be within his powers, with encouragement and the sharing of risk and
a little help at critical points. This will focus danger. It must also be appreciated that it may
his interests and bring out his best. Again, this take time to convert from the driven to the
is something that must be taken into account self driving mode, and that there may thus be a
in the movement of jobs and of people. dip before the rise.
To make a job artifically difficult by lack Typical of the legitimate aids in this
of aids will not help: morale will only be high process, and the aim of procedures, should be
the appeal to the problem-solving instinct, the importance in design, for which both width and
game-winning instinct (i.e. the competitive insight are highly desirable qualities.
spirit), the team spirit, the sense of occasion, Confusion also creates alibis for failure,
the sense of personal accomplishment, and so and in this way tends to undermine the sense
on. For example, competition is a powerful of commitment which has been stressed as
stimulus, but has to be seen and feit by the important. On the other band, the organization
competitor to be effective. It may be necessary must not be so 'tidy' that it becomes claustro-
to encourage some internal competition where phobic: each man needs his own little 'castle'
it would not otherwise be available. The view in decision-space in which he is able to puzzle
that ali overlap or competition is wasteful is a things out, free temporarily from the stares and
dangerously narrow one, but all too common. lifted-eyebrows of his neighbours.
Similarly, the creation of teams, particularly Politics- A concomitant of confusion is
where these compete in some obvious sense, politics - the prevalence of unobjective, selfish
provides the maximum self disciplinary force, thinking and behaviour. Sometimes this state
by combining the desire to partake in victory can develop from a single infecting source, and
with the desire not to lose respect by letting it must be the aim of procedure to detect and
colleagues down. eliminate it at an early stage. Frequently,
The importance of the other two factors however, it develops spontaneously from the
mentioned is seen from the extra effort which misunderstanding and misinterpretation asso-
can be drawn out in preparation for exhibitions, ciated with confusion (particularly with the
conferences and similar immovable and prestige- spread in view-span referred to above), and the
involving external events. Not only can such feeling of insecurity thereby created.
stimuli and pace-setting means be encouraged Frustration - This can be regarded as a
(instead of being regarded as interruptions in mood generated when work, or ability, is
the work), but the process can be extended impeded or wasted through procedural defects.
within the organization and used as a principal It leads to low alignment with the organizational
energizing mechanism. goals, reduced enthusiasm, initiative and sense
of responsibility.
Organizational Factors It is important that procedures be screened
This term is used as a convenient title beforehand for this possibility and also watched
for some factors which characterize an organi- carefully in operation, looking for signs of
zation as a whole, but undue significance frustration. A typical defect is the operation
should not be attached to the classification. of a design approval procedure too late in the
Confusion - An organization can be design network, or lack of care in the agreeing
regarded as confused to the extent to which it · of the original target specification.
lacks a complete, well-known and coherent set Morale - This is an index of the overall,
of procedures. Although a certain amount of deepseated, mood of a man. It is low if the
confusion gives scope for personal initiative, work is unnecessarily difficult, unrewarding
it is much better that such scope be embodied and prejudicial to his long-term prospects; if
in the procedures rather than occur for want of he is being asked to make unnecessary, or
them. Scarcely any inhibition can be more useless, sacrifices. One example would be a
powerful than not to know the limits of one's man kept at the same task indefinitely, for
authority, the means of getting things clone, convenience, because of some special skill,
who eise is likely to be involved in decisions, when someone eise could easily have been
and what factors may later upset the decisions. trained to take over. His standard will tend to
Unfortunately those most likely to be inhibited, fall and he will be strongly motivated to leave,
discouraged and driven away are those who see often without complaint.
furthest and most deeply, since they are least It is important that one of the personnet
likely to be satisfied with temporary decisions, procedures should be to visualize continually a
or superficial solutions. This can be of great man's job as he is likely to see it, not just as
10
it appears to the organization. There is a need transverse supply and communication lines in
to develop methods of measuring objectively accordance with Figure 29.1. The process may
aspects such as morale, not only for individuals be represented in the main by a single un-
but collectively for groups and for the organi- branched flow line, as shown in Figure 29.2.
zation as a whole. In this diagram the process chain has been
Procedures in Design divided into a number of zones, marked A, B,
The discussion so far has been concerned etc. Of these A represents the market study
with procedures in general, but with emphasis stage, which may or may not be regarded as
where appropriate on design. In a full study of part of design. Zone B is the design study stage,
the problern there will be a need to look at the about which there is little doubt. In zone C no
whole design process network in detail, examin- attempt has been made to distinguish between
ing the procedures likely to be involved and by design and development (which is regarded here
applying principles such as those tentatively as the experimental trying out and consequent
put forward here draw conclusions for future refinement of plans prior to completion), since
guidance. It would be premature to attempt this design cannot be regarded as complete until
here, but it may be useful to take the discussion development, if any, has taken place.
a little further, if only to bring out perhaps the The following problems on procedure are
nature and extent of the problem. suggested by an examination of Figure 29.2:
The Prncess Network
(1) Should there be a formal network for the
Any actual network will show considerable
organization, or should design be informal, or
branching with spurs and loops at various
ad hoc?
points consistent with the inspirational and
(2) If formal, should there be through-
iterative nature of design. There will also be
control, e.g. market-study to marketing, or
should responsibility be zoned (departmentali-
Study market
t Study re'-O<Jrces
. and strategy
zation); and if the latter where should the
frontiers lie?
(3) If through-control, should there be a
Correlate
A
Select ta.rget
Marke!
study
single vertical network, or several in parallel
(divisionalization)? If the latter how should it
-~~y-~~~-------------1 be defined - discipline, product, or field of
Study teas1bil1ty !
Study deslgn !i application? Should there be overlap (technical
8
Rev1se target Des 1gn competition) or purely profit (or performance)
Dec'de yes(or no) study
Plan(targets. re'-O<.Irce allocat1on. 1 competition?
dates. marketmg, ~tc )
Obta1n approva1 t
1
(4) What authorizations, check-points and
---------------------
lntt1ate
general controls (financial, programme, techni-
Suppl1es cal) should exist and where should they be
rI
Assemble facllltles, suppl1es
(mater1als.eQu1pment. applied in the network? What should be the
statt. know-haw,
Do tnal des.ign
Eng1neer
~~~atJst s.ennces., etc)
I remaining decision structure, relative to the
Product' draw 1ngs
Make models Des1gn design network, i.e. who should decide what?
and
c
Tes.t,evaluate de-Rlop!T'It'nt (5) Should the people be stationary with the
Rev1se des1gn
Redraw work flowing past, or should they move down
Prepare supp0rt1ng documents with the job: many functions, same job; or same
(speclflcations,literature.
Jnstruct,ons. etc J function, many jobs?
Qbtam apprcNal
Release to product 1on
(6) What degree of teaming should exist?
What composition, training and life-cycle?
Producl!on,marketing, ~lc.
(7) How should specialist skills and ser-
vices be fed in? Should people move in, or work
Figure 29.2. Symbolizing the design network move out? How should extra-mural work be
(skeleton only) placed? Who decides and pays?
(8) What aids should be provided: compu- up-to-date. There will also be many points of a
tation, data retrieval, special instrumentation, more detailed nature in connection with parti-
consultants, secretarial, etc.? cular processes. Throughout the whole study,
(9) What constraints are tolerable: secrecy, the psychological aspects discussed must be
publications, exhibitions, discipline, budgets, bome in mind, in addition to problems of work-
etc.? study planning, which have been rather played
(10) What environments are desirable: down in this discussion, but will become of
buildings, localities; intellectual, discussion increasing importance in the later stages of
groups; psychological; social; housekeeping, optimization.
etc.? Finally, there will be a need for consider-
(11) What procurement procedures are desirable field work, to find out what procedures are
able: consumables, capital goods; calibration, current in representative design laboratories
standardization, etc.? and what trends are taking place. An attempt
(12) What staff procedures are desirable: must be made to distinguish the good from the
recruitment, selection, grading, salary scales, bad (on the basis of actual results) and to try
assessment, promotion, pensions, side-benefits, to understand the reasons which lie behind the
status symbols, posting, training, further edu- differences. The career histories of notably
cation, cultural activities, health (body and successful designers will also be of great
mind), and so on. importance, including a study of their environ-
mental backgrounds. The discussion here will
These are some of the many aspects which will have been useful if it encourages a wide and
have to be examined, and to which a nurober of deep approach, and wams of some of the things
others may be added on fuller consideration. which may prove to be significant. Above all,
In addition there are questions such as how however, is the need for careful unprejudiced
procedures are best drafted, introduced and kept research.
Chapter 30
COMMUNICATION
C. H. Buck
lntroduction measurements are made by a development

Design without communication is impos- engineer, not by the designer, so at this stage
sible, and many modes of communication are the model becomes part of the communication
discussed in this book. Frequently the designer between the designer and his immediate
does only the mental work, and the execution colleagues.
of the design is then clone by others. It is In a manufacturing organization there will
necessary to communicate the details of the be formal communications concerning design,
design to the craftsmen so that they will know instructions from and reports to management,
what to make. This is commonly clone by means and the passage of information between various
of engineering drawings, and in Chapter 26 functions and design, such as costs(Chapter 18)
McMullen has discussed ways of improving the marketing (Chapter 6), production and service
output of the drawing office in quantity and (Chapter 27). The designer also needs a great
reliability. deal of technical information, and in Chapter 12
But the designer first needs to communi- Farradane has suggested the possibility of
cate with himself. Rarely is a design conceived using computers to provide more rapid access
instantly and without flaw. Usually the ideas to more information. Peplow, in discussing
come in parts, and not all of the parts fit. Some design acceptance, is really discussing the
means have to be found to fix the fleeting communication that goes on between a designer
thoughts, so that the designer may evaluate and his customer, In many situations the
them thoroughly by all the relevant criteria. customer will be represented by the designer's
The commonest medium is the sketch, which marketing colleague.
may be elaborated into a formal drawing. An All these modes and uses of communication
example from fine art is the painting of The are important, but one problern that continually
Lion Hunt by Rubens, which is in the collection occurs is the failure of communication between
of Dr. Count Antoine Seilern, together with the individuals who are concemed in different
original drawings and the intermediate oil ways with the same design. In this chapter, the
sketch. nature of the problern is described, and a
The engineering designer may prepare a partial solution suggested.
chart to record his mental processes, and to Consider an organization manufacturing
guide him in his decisions on the lines illus- speculative products in quantity. This seems
trated by Matchett and Briggs in Chapter 21. to be the most complex design situation within
He may construct models, which are particularly a single organization. Here there is no physical
useful when the appearance is an important barrier to communication, since the entire
consideration. Also, they are invaluable as a organization (except for distribution detach-
check on the solid geometry of the design, ments) may well be on one site, perhaps under
which may have been incorrectly envisaged at one roof, but obstacles to communication do
the sketch stage. The model in the fortn of a exist. Some of these are personal incompatibi-
working prototype is indispensible when lities, some are the bottlenecks of a rigidly
evaluating performance. Usually performance formal organization, and some are caused by
279
the naive assumption that certain notions are much information can be gained from market
so obvious as not to need stating. However, research, but often the information is very
communication is essential, because so many limited, and when great novelty is involved the
factors are involved, even in the design of a information is unreliable. For example, ICI
simple consumer product, that no one man can invested several millions in a plant for
possibly know enough to make the best design manufacturing beryllium, but after a short
decisions. period of operation the plant was shut down,
It may be that in examining this limited because the expectedsales were notmaintained.
problem, some clues will be found for dealing It is desirable to know what sort of product
with the problems of communication in more will sell, in what quantities, and at what price.
difficult and complex designs. Some allowance must be made for the time
needed to design and to prepare for production,
Speculative Manufacture so that these forecasts must usually be made
A great many of the materials and appli- several years in advance. This long range does
ances used by the private individual are pro- not add to the precision.
duced speculatively. That is to say, the Even more questions must be asked. Is
products are made and exposed for sale in the the proposed product technically feasible?
hope that Emough people will buy them. Some of Is money available to finance the design and
these products are quite simple, like packets development? Can staff and facilities be
of sugar, while others are quite complex. The released or found? Can the product be produced
domestic appliances of the western world in the quantities expected with the available
(cookers, washers, television)and the ubiquitous equipment? Can the new product be handled by
motor car are all speculative products. Mostly the existing sales and distribution organization,
they are made in large quantities and each or must a new organization be created? What
manufacturer offers a limited range of models. sums must be invested in advertising to
A great deal of expense is involved in pre- stimulate and maintain the desired rate of sale?
paring to manufacture, and much of this expense Will the new product require maintenance? If
is directly related to the detailed design of so, what facilities must be provided by way of
the product. Once the tools have been made it information, supply of spare parts and trained
is very costly to change the design except in engineers and mechanics?
minor details. But the market, that is the Until satisfactory answers to all these
public, cannot have the opportunity of judging questions are forthcoming, there is no point in
the product until all this preliminary work of authorizing any major design activity. These
tooling has been clone. Therefore the manu- are questions for management.
facturer must incur all the expense of design
and tooling before there is any opportunity of F actors lnvolved in Design
recovering this investment by selling the The questions to be considered by the
product. Frequently several years of manu- members of the design function will depend in
facture and sales are necessary to recover the detail upon the nature of the product. Funda-
total initial investment. If this kind of enter- mentally the questions are: what is the product
prise is to succeed, good design decisions are expected to do; in what circumstances will it
essential. Everyone knows of the bad design be used; what sort of people will use it; how
decision made by Ford of Detroit. This mistake, will it be made and how much may it cost to
the Edsel, is said to have cost !:120 million. produce; what conventional constraints will
Few companies could survive such a disaster. be laid upon the designer by law, by custom
and tradition, and by company policy? Several
Factars /nvolved in the Choice of a New Product different sorts of skill and experience may be
The fundamental necessity for a com- needed to deal with all these questions.
mercial product is that it should sell in Typically these skills are likely to be those of
sufficient quantity. In some kinds of business the engineering designer, the industrial artist,
the development engineer and the maintenance make the decision himself, he must still collect
engineer. all this information, and he may devise some
For both the very diverse range of way of presenting the salient points con-
considerations that exist at management level, veniently and compactly, perhaps on one side
and the somewhat lesser range at the design of a sheet of paper.
function level, all the right answers have to be He will not write the story himself. His
found in a single design. At the lowest accept- function will be more of an edi torial nature, for
able level this will be achieved by compromise; the component parts of the story must be
at the highest possible level it will be reached prepared by experts in the various fields, i.e.
by integration. Therefore at management level market research, economic forecasting, adver-
there must be communication between the tising, production, finance, costs, and of course
representatives of the various interests, and design. At the very least, in preparing the case,
at design level between the practitioners of he will need to consult his executives in the
the various skills. Because these people have fields of finance, production, sales, and design.
different interests, different backgrounds and These men are specialists. Probably they have
different skills, communication is inefficient. professional qualifications. Almost certainly
they will have peculiarities of personality and
The Management Team temperament that have led them to their choices
In an organization making a small number of profession and to their careers. Each will
of different products, each produced in quantity, have his own jargon. 'Development' means one
the decision to initiate a new product is of thing to the product designer, another to the
major importance, not only because of the cost production engineer, and something different
of design, development and tooling, but largely again to the salesman. (The sheet metal worker
because the new product will be expected to also uses the word, in a fourth and entirely
contribute a substantial part of the company's different sense.) So each time these men speak
income when it comes on the market. Therefore to each other they may easily be misunderstood.
such decisions are frequently taken at the The words have different meanings, and
highest possible level, that is to say, the communication is inefficient.
board of directors decides. In other companies,
the decision will be taken by the chief Application of Information Theory
executive who may be styled 'general manager' In discussing engineering communications,
or 'managing director'. In the former case, ROSENSTEIN, RATHBONE and SCHNEERER
the chief executive is likely to be responsible (1964) introduce the jargon of information
for putting the case for the proposed new theory, by writing of signals and noise. For
product to the board. If a proper decision is efficient communication the signal-to-noise
to be made, the case presented to the board ratio must be high. Noise arises in many ways:
must be complete. The chief executive must when, for example, a salesman talks to a
state what sort of new product is envisaged; the designer, the original message is in the
expected selling price and the rate of sale; salesman 's head. To communicate, he must
the estimated works cost; the launehing date; first encode his message into words, then he
the estimated costs for design and development, must speak the words, then the designer must
tooling and other preparations for prdduction, hear the words, and finally he must decode
and launehing promotions; the nature of the them in order to receive the message. But if,
competition. to be faced; the sales organization as is usual, the designer and the salesman use
needed; the distribution chailnels and the the same words in different senses as part of
physical means of distribution; whether their professional jargons, there will be
servicing facilities will be needed and of what misunderstanding of the kind described by
sort; the effort needed to recruit and train these authors as semantic noise. There may
labour; for some products, other considerations well be mechanical noise also, if the con-
are also needed. If the chief execu ti ve has to versation takes place over a bad telephone
line, or in a noisy room. And because the saying: 'I cannot hear you; will you speak
salesman and the designer have strikingly louder.' If this fails, the original transmitter
different personality characteristics, there may introduce redundancy by spelling out the
is likely to be psychological noise as well. key words, Ietter by Ietter. He may introduce
One method of increasing the efficiency even more redundancy by using the A for Ahle,
of communication, that is, of reducing the V for Victor system of identifying the letters.
confusing effects of noise, is to introduce (Devotees of Hilaire Belloc should be warned,
redundancy into the signal. In ordinary written however, that such redundancies as F for
communication redundancy is always employed Vescence, M for Sis and X for Breakfast are
as a means of reducing the effects of more likely to confuse. The only acceptable
mechanical noise. The details of the letters Belloc forms are L for Leather and T for Two.)
now before the reader are much more complex Feedback can also deal with semantic
than is necessary to distinguish one Ietter noise. The receiver may reply to the message
from another. There are many differences in the by saying, for example: 'What do you mean by
forms of letters, although five pairs only of "efficiency"?'The transmitter may then define
different features would be sufficient for an 'efficiency' in words, dictionary fashion, or he
alphabet of thirty-one letters. Moreover in any may explain it by an exani.ple, or he may repeat
one language certain patterns of Ietter combi- his message, encoding it so as to avoid the
nation are common, and other patterns are rare confusing word.
or absent. Consequently, even in a badly written Feedback is particularly good for reducing
communication, the letters can usually be psychological noise, especially when the
identified from some feature of the distorted communication is between two speakers face
or debased form, or from the context. However, to face. Then the feedback signal need not be
in commercial shorthand, writing speed is in the form of speech. The designer may be
achieved by the elimination of redundancies of discussing working tolerances with a production
form, and by the elimination of some letters, man. He is saying: 'When it comes to limits
with the result that badly written shorthand is I don't see what the pro ... ' and at this point
often completely illegible. he notices that the production man has become
In an analogous fashion, when one man very red in the face, and that his mouth is
speaks to another in a noisy situation, he may moving in such a way as to suggest that if he
communicate successfully by repeating his could find the right words he would make a
message several times. If the noise varies very angry speech. So the designer changes
randomly in quality and intensity it is likely his message quickly, and encodes it in such
that in successive transmissions different words that he can continue without a break in
words will be received, the others being masked transmission: ' ... blem is if we have a Iook at
by noise; after several transmissions the the desigu together to see where it can be
receiver will be able to reconstruct the entire eased.'
message. There are always psychological problems,
The confusing effects of semantic noise because each specialist assigns most value to
may likewise be minimized by repetition, but those features of the design which most closely
for success the message must be recoded each affect him; the designer with performance and
time into different words. The receiver in his appearance, the production man with ease of
decoding will then search for the meaning production, the salesman with sales resistance,
which is common to all the varied signals. the accountant with costs. Under pressure,
This method can also deal with the mechanical which is the normal commercial situation,
noise arising from a strong regional or foreign everyone wonders why the other fellow makes
accent. such heavy weather of his particular aspect.
Another method of reducing the effect of The specialists at executive Ievel must
noise is by feedback. Thus over a bad telephone agree upon a design project or, at the worst,
line, the receiver replies to the message by the chief executive must make a decision which
will make sense in terms of all the specialized found of integrating technical and aesthetic
information and points of view which will be requirements in a design is to let the engineers
presented to him. If his decision is to be a and the artists spend enough time together for
good one (and in this situation there is un- their various criteria and constraints to be
likely to be a uniquely correct one) he must mutually understood, or at least respected.
be as sure as he can that he has got the Again, in some organizations this situation
message, that he has correctly understood all has been formalized by setting up design
the signals that he has received. teams jointly responsible for the entire design.
This is likely to occur only if he has Thus the team for a domestic appliance might
frequent talks with each of his executives, in consist of an industrial artist, a development
which there will be much redundancy and engineer, and a production engineer. They will
feedback. Further, in order better to assess spend a great deal of time in talking, and in
the influence which each specialist contri- scribbling on each others sketches. How else
bution has upon the others, and therefore upon can they do their job?
the final decision, there must be much dis- Communication between Management and Design
cussion between the specialists. In some There remains the question of communi-
organizations this process is formalized by cation between management and the design
setting up a development committee, or a new function. It might be argued that since the
products committee, whose members no doubt design executive is a party to any management
complain bitterly of the waste of time in com- decisions, he should fully understand what is
mittee meetings. But so far, no-one has found needed, and he can therefore pass any neces-
a more efficient way of dealing with the sary instructions to his staff. This might be
mechanical, semantic and psychological noise true if the management decision could be drawn,
that comes from human variability and human so to speak, out of the air of the committee
imperfection. The committee meeting provides room, and if the design executive could be
the opportunity for verbal and visual communi- trusted to communicate the design decision
cation in circumstances which permit maximum without noise or distortion. Since no-one
feedback and controlled redundancy. concerned is superhuman, neither condition is
satisfied.
The Design Team The new design often starts in the vaguest
At the functional Ievel of design, similar fashion: 'If only we could. . . . ' 'I sortathought
problems occur, even when the product does we might. ... ' It may even start with confusion
not require a large staff. A common problern concealed behind apparent explicitness:
is to reconcile the requirements of the artist 'Something like Bodwin's pencil sharpener only
and the engineers in the design of a consumer for cleaning windows.' There is a stage at
durable product. Sometimes the industrial which the management team needs some help
artist is a consultant, called in to 'tart up' an in visualizing its nebulous ideas. This is the
engineer's design. These are the works, and stage at which there can be some profitable
can he put a respectable overcoat on them? two-way communication between management
Don't bother about .the gears and Ievers; only and the design function. Communication down-
our back-room boys understand what they do. wards is likely to be made through the design
Just make it look pretty. Ideally the artist executive. He will not rely purely upon words
should refuse such a commission, but he may for this. He will almost certainly supplement
be short of clients. In another organization, the conversations with his designers with
more conscious of the sales importance of sketches, and the doodles he has made in
appearance, the artist may be asked for a committee. The designers and the modelmakers
design, and 1hen 1he engineers sweat to get will prepare drawings and models based on
the works inside the stylist's form. Both of this information.
these are almost certain methods of getting The nature of the models will depend upon
very bad designs. The only way that has been the nature of the management problem. If it is
a question of appearance, the models will be of design, purchasing of raw materials, and
quite crude wooden boxes, beautifully moulded so on.
and painted on the 'show' sides to give a Given this formal statement, design may
lifelike impression of the product as it is proceed as a purely functional acti vity. In the
conceived at that stage. If the problems are absence of the new product specification, as
technical, the model may be a 'breadboard' for example during the period when management
model or its non-electronic equivalent. If the is developing its own ideas, the designers
problems are ergonomic, the model may in- must operate within their own function as best
corporate both technical and appearance they may. As stated above, when widely
features, especially those related to controls different skills are invol ved, the best way
and the display of information. seems to be to let the different practitioners
Thus the designers can now communicate work together, and especially talk together.
upwards to management. The communication In the absence of much repetitive, redundant,
will be largely visual, the signals being in the argumentative conversation, the artist, the
form of models or drawings. Some communi- technologist and the production engineer will
cations will be expressed numerically as never understand each other sufficiently to
predictions of performance or as cost estimates. evolve integrated designs. Indeed they may not
Some communication may be tactile or achieve even decent compromise solutions.
kinaesthetic, when the models are handled. By an integrated design is meant one in which
The efficiency of communication through these" the problems of appearance, performance and
channels may be improved by redundancy in cost have been solved simultaneously in a way
the form of verbal explanation and argument. that is better than any individual specialist
With argument comes feedback, which will be solution.
helpful so long as it is controlled to reduce The converse situation is only too well
psychological noise and not to create it. Out known. Designers are often guilty of assuming
of this variety of communication should come that their particular expertise is merely common
understanding. If the designers can be present knowledge, and that it is not necessary to
with their executive in the development com- explain their problems or their solutions to the
mittee during the presentation and discussion members of other functions, or to management.
of the models they will come to understand Production engineers, salesmen and even
what management wants. Equally, in a typical managers assume that their special difficulties
situation, management may come to understand are so well known, that the designers must
why it cannot have what it wants. If the ideas know about them without being told. Others
of management are nebulous to begin with, adopt the attitude that there is no point in
they will become more definite in the presence telling the designers anyway; that designers
of a physical model based upon their first are impervious to ideas from outside their own
thoughts. Ultimately, management will become coterie. A great deal of the study of The
sufficiently confident to authorize full-scale Management of Innovation by BURNS and
design activity. Ideally it should issue a new STALKER is devoted to recording the views of
product specification to define the design many workers in design and development, and
activity. This document should state precisely in various aspects of management. These
what sort of new product is required, what reported conversations have a painfully familiar
features are essential, how much it may cost, ring· stories of errors and misunderstandings
what rate of sale is expected, and how much arising from failures of communication.
time may be allowed to complete the design Bums and Stalker demonstrate, as a matter
and to prepare the tools and other production of observed fact, that those organizations
equipment. This last item, together with the which succeed in exploiting design and
sales forecast, will also be needed to permit innovation are those in which the pattern of
the planning and programming of advertising organization is organic rather than mechanistic.
campaigns, negotiation of customer approval They list eleven characteristics of organic
systems, of which two are especially relevant can deal with the problem. The same long
to this chapter: chain is involved each time a development
engineer requires the prototype to be modified
(1) The adjustment and continual redefinition to remedy a fault, and each time a production
of individual tasks through interaction with engineer wishes to change the design to reduce
others. costs. Clearly, in this model any foreseeable
(2) A lateral rather than a vertical direction market would be lost long before the design
of communication through the organization, could be completed.
and communication between people of different
rank resembling consultation rather than Conclusion
command. The quality of design decisions at all
Ievels is likely to be greatly improved and
These two, and to a lesser extent the other expedited by the active encouragement of
characteristics described by Bums and Stalker, free communication between all those people
support the contention of this chapter: namely, involved in design, both at management level
that for efficient design it is necessary to and within the design function. The problems
have efficient communication between all the are familiar to designers, though possibly not
people concerned in design, and at the manage- so obvious to people in other functions in
ment level this means the heads of all functions. industry. Now that they have been fully
Further, for efficient communication it is documented by Burns and Stalker no-one has
necessary to introduce a great deal of re- any excuse for shirking the problems of
dundancy and feedback in order to increase communication in design. Three actions seem
the signal-to-noise ratio to an acceptable level. to be necessary:
Elsewhere, (BUCK, 1963) I have come to (1) Managers and designers should devise
the same conclusion by examining the operation means for improving communications within
of a perfectly bureaucratic model organization their organizations and departments.
in which a new product is being designed. (2) They should attempt, either by their
The organization pattern of the model is taken own efforts or by enlisting the help of social
from actual examples in industry with which scientists, to assess the effects of the changes
I am familiar. In the model, design and pro- on the validity of their design decisions, on
duction are controlled by executives respon- the speed with which these decisions are
sible to the general manager; but two aspects reached, and on the morale and profitability
of design - model-making and production of their organizations.
engineering - are controlled by the production (3) They should publish their results for
executi ve. This is quite logical since the the active help and encouragement of all.
model-makers need a workshop and the
production executive controls all workshops. Finally, it is suggested that when design is
Without control of production engineering, the revolutionary rather than evolutionary, com-
production executive cannot control production, munication may prove to be the most intractable
so the design executive has direct control problern of all. For, in order to attract financial
only of the designers and the development support and technical assistance, it is neces-
engineers. The only theoretical feature of the sary to communicate the new idea to those who
model is that communication is rigidly confined have not yet seen the light. In this situation
to the channels of authority. Consequently, semantic and psychological noise will effec-
every doubt or query raised by the model- tively mask any orthodox signal, and effective
makers when the prototype is under construction communication will require even more pre-
has to be referred, in turn, to the foreman, paration and more patience, more explanation
the production executive, and the design and more argument - simply, more redundancy
executive, before it reaches the designer who than in our rather workaday example.
Chapter 31
DESIGN, MANAGEMENT, REALIZATION AND CHANGE
S. A. Gregory
lntroduction which is observed. This unsteady situation

Much of the discussion in this book has may be comparable to a ripple on a stream, or
been directed to specific aspects of design, to to the growth and decay of a plant. In either
design as a process and to techniques and case there is now to hand substantial formal
principles con tributing to the execution of theory and techniques to deal with such
particular stages in design. Chapter 27 has situations.
devoted necessary attent!on to the way in At one extreme, there might be a relatively
which design operates as an important and stable manufacturing organization, apparently
critical function of a manufacturing company, in balance with its market. The company
and to the way in which the design system fits structure is almost certain to comprise vertical
into the company organization and participates functional divisions of the classical
in its activities, nourishing and being nourished 'mechanical' type, embedded, as in a fossil,
by them. in the description of organization by BRECH
The present chapter is concerned with the (1957). For some reason beyond the manage-
practical consequences of the fact that design ment's control a change has to be brought
is a process. It prepares the way for a detailed about, either in the design of one product
study of the technical basis which underlies which contributes to the stream of output, or
the relation between management structure and in the effectiveness of the design function.
technology developed by WOODWARD (1965). The method of design associated with product
improvement will be only too familiar by now.
Steady-state and Unsteady-state Processes The improvement in design office effectiveness
One tends to think of much of design as may come from improvement in individual
contributing to a broad stream which carries designers by training, which might be, for
on in a substantially unchanged flow, and in example along lines suggested in Chapter 21.
which the individual effect is almost un- Or it might come from an analysis and re-
noticeable. This continuous and steady flow allocation of tasks in the design office using
is one's image of the industrial process and the techniques of work-study, as outlined by
of the economy as a whole. No doubt, this CURRIE (1960). TURNER (1964) has studied
image provides some comfort, because it the activity of a drawing office, and an intro-
reflects a kind of stability and because it is ductory statement on the application of work-
an easier situation to deal with than one in study to a design organization has been made
which major variations occur. Furthermore, by FEARN (1962). Others have examined
the steady-state view tends to accord better aspects of control through costs, including
with British national outlook on industry. network methods. The rather less immediately
However, taking either a long perspective, effective changes in overall policy or in the
such as that of TOYNBEE (1960) who sees the environment and atmosphere have been dis-
rise and fall of civilizations, or looking only cussed in ChFpters 28 and 29.
at the activities involved in the design of one The implementation of such changes may
small product, it is an unsteady-state process be seen as a disturbance, and handled in terms
287
of industrial dynamics, as developed by On the other hand, more needs to be generally

FOR RESTER (1961). The detailed planning known about the life-cycle patterns of products,
may be executed by a suitable network empirically and theoretically.
technique. A number of phenomenological observations
On the large scale, the introduction of a have been collected by BRIGHT (1 %4) but it
new design may be radical and critical. It may cannot yet be said that there is information
represent the embodiment of a substantial about enough projects in sufficient detail.
change in direction or rate of growth of a NORDEN (1963) has provided details of
social group, or it may be the production and empirically based equations dealing with
use of a new supersystem; in some cases it growth. The development of the field of
may concern a complex of such things. Obvious industrial dynamics by Forrester and, in
examples are the USA space programme; the particular, its application to the elaboration
UK National Plan; the development of a non- of new products by ROBERTS (1964) provides
industrialized country. a technique which, although suggesting possi-
Such phenomena, whether large or small, bilities, probably substantially outruns infor-
may be considered in terms of individual item mation.
growth andin terms of the comprehensive social Certainly the most detailed practical
or group growth. As far as the individual item attempts at growth analysis have been made by
is concerned, a suitable sequence of comment the people concerned with project management,
seems to be: the birth, growth and death of a particularly using the new network planning
product; the project management of a complex techniques.
one-off product, including management structure
and techniques; the interrelationship of Project Management
marketing, research, design, manufacture, Those wi th experience of new product
utilization and financial operations. design and development are keenly aware of
the difference between the management's
needs to get the product established and the
The Life-cycle of a Product traditional management structure recommen-
The Iife-cycle of a product is largely dations. This difference is feit most acutely
determined by the organization which produces in organizations which have a large content
it, and by the environment into which it goes. of established lines. HALL (1962) discusses
The necessary 'profile' of a proposed new alternative methods of organization for 'mixed'
product, which recognizes the need to fit the operations in a concern with substantial new
production organization and the environment, system projects. He differentiates between
has been outlined by HARRIS (1961). It is such the departmental form of organization and the
a profile and its fit which may exercise the task-force arrangement. Students of manage-
dominating factors in investment appraisal, ment will recognize the various synonyms for
rather than any single criterion approach, as such structures. Project management is
has been suggested on the basis of empirical essentially the kind of management set up to
studies by WILLIAMS and SCOTT (1965). deal with situations of change, and is the
Indeed, such investment criteria as those which most common version of the task-force system.
employ discounting to the present date, with Project management has probably been
allowance for estimating cash flows for indi- developed to greatest effect in organizations
vidual years during the life of the product, whose concern is solely with new projects,
include in principle the possibility of making i.e. in firms which deal solely with large one-
an adequate financial appraisal. It is most off contracts, or in the special branches of
likely, however, that the life-cycle of the !arge user organizations set up to deal with
product will be incorrectly foreseen. new projects.
Some of the difficulty in estimating lies BAKER (1962) summarized his diversified
intrinsically in the uncertainty of the future. experience within !arge user organizations in
a mechanical and largely unperceptive manner. depended upon the production of charts con-
This was at a time when the Americans were sisting of points (nodes) which represented
gathering to boil down their experience in events, and interconnecting lines which re-
dealing with military projects, as brought presented activities. Some quantity such as
together by KAST and ROSEN ZWEIG (1963). time might be attributed to an activity. By
Comparable, but earlier experience in Great setting out the complete network from start
Britain is given by POSTAN, HAY and SCOTT to finish, it became possible, by working from
(1965). the 'finish' end, to determine which path from
The organizational procedures in a general a nurober of contemporaneous activities con-
sense are the same in any of the major plant stituted the critical path. In the light of this
contracting organizations. They have become information certain actions might be undertaken.
almost commonplace but there is, as yet, no A useful introduction to this method of
readily available comprehensive work on the planning and control is provided by LOCKYER
subject. Numerous papers have been given (1.964) who makes clear some of the extra
but not recorded. The joint institution sym- values that come from its practice. It provides
posium of 1966 marks a change. a discipline which forces consideration of
what has to be done, when, by whom, and in
what order. It makes clear what decisions
Project Planning by Network have to be taken and provides the basis for
In addition to the increased awareness indicating the decision-maker and some con-
of the techniques of search for market opportu- sequences which may flow from his decisions.
nities, economic and technical, perhaps the The construction of the network is
greatest influence in project management has essentially a specialized design operation.
come from the development of network tech- A sequence of future events is projected in
niques. the light of current knowledge of realization
The move from the older planning methods, facilities, and is selected to provide what
using the Gantt chart and similar devices, appears to be an optimum pattern.
began in 1956-1957. The earliest work came The initial development of network
from sturlies of possible applications of new planning techniques was concerned with the
management techniques to the engineering problern of achieving a desired completion
functions of the American chemical manu- date. The PERT method introduced the notion
facturing company, E.I. duPont de Nemours. of dealing with uncertainties. Today, as a
The first sturlies concerned plant overhaul and result of. many diverse investigations, network
similar work was being undertaken by the techniques may be applied to other aspects
C.E.G.B. in Great Britain. The USA work gave of projects than time or completion control.
rise to the CPM (critical path method) and For example, it is now possible to prepare
details of the growth are given by O'BRIEN network plans in which account is taken of
(1965), a member of the consulting organi- restrictions in availability of resources. These
zation which spread the early use of CPM. resources may be any of those commonly used,
Much greater prominence was achieved in its namely men, machines, money, in addition to
early years by the allied technique - PERT the previously considered time. Readily
(programme evaluation and review technique). available books on these wider issues are
This was developed for the US Navy ballistic those of MODER and PHILLIPS (1964) which
missile project and its points have been deals particularly with cost and with
competently summarized in a short article by statistical aspects of PERT; and of
the firm of consultants associated with it WOODGA TE (1964) which covers a wider range
(Harvard Casebooks, 1964). of applications against a computer background.
The initial network planning and control Today, the more generalized approach
techniques, although having certain points of and the study of situations involving inter-
difference whose significance is now receding, related programmes are seen.
lnterrelated Programmes and events have been displaced to arrows, thus

Within a single project, situations exist inverting the original planning network mode
where some events have to occur before others, of indication, although becoming more intel-
or where, in the case of parallel activities ligible to all those engineers who expose the
and limited resources, decisions have to be technical circuits of their processes in the
reached and implemented regarding priorities. latter way. Much more general still is the
Such situations may exist on the level of the development of the mathemaUes of networks
supersystem, or in a developing country, or and its drawing tagether from the most diverse
in an industrialized country preparing a sub- of fields. ELMAGHRABY (1964) has developed
stantial change in economic activity. an algebra for generalized activity networks.
A qualitative discussion of interactions
in the evolutionary growth of complex systems It has now become possible to look at the
is given by SALZER (1961), indicating decision problems within the whole field of
possible ways of avoiding development of technical development in some kind of rational
everything at one step. This concept of permit- way. Given the restrictions in the expression
ting growth of individual parts of a newly of some objective function by which to express
developing economic system, with phasing-in the desirability of a proposed investment of
of different stages across the whole economy, resources, as discussed in Chapter 10, the
has been outlined by SZUPROWICZ (1963). problems of decision in successive stages of
Within a developed economy, such as in the total scheme may be investigated. It is
Great Britain, the interconnection s of various possible to start with the idea of dealing with
branches of the productive system are a portfolio of investments, as developed by
extremely complex. In spite of this it has ADELSON (1965); the most promising portfolio
been possible to build up computer models of research prospects can then be selected,
of the economy. STONE (1965) and his accordin g to the pattern suggested by
colleagues at Cambridge have particularly DA VIES (1962). The course of an individual
developed this approach. In their economic research project may be plotted out. In quite
matrices they display alternative pattems of comparable fashion, the course of any project
products from a given · economy with certain may be plotted out from the state at which
resources at any period in time. SHONE (1965) sufficient data have been established to the
discusses how it is possible to develop a completion of the construction of the production
succession of these matrices against time, plant, making due allowance for loss of
setting out preferred lines of development of opportunity owing to the need for constructing
individual industries in some detail. This pilot plants along the way. This has been
kind of projection demands, of course, some worked out by Gravenor, but has not yet been
generat agreement on possible optimizing published. In bis procedure the set of alter-
criteria for such operations. native trees is drawn out against time. Money
Within any company, the interrelations of values are given to markets for individual
activities over time in connection with years, costs of pilot plants and full-scale
company growth are exceedingly complex. plants are provided, and probabilities are
However, the various topics touched upon in accorded to alternative outcomes at various
this chapter are beginning to promise possible stages along the routes. Starting from the
ways of analysis to help towards dealing with finishing end, each tree is summed in order to
such problems, particularly the interlinking find the best decision at the furthest nodes
of research, development, design, production and the system is gradually 'rolled back'.
and sales. Once the preferred route has been reached,
another network planning operation can then
More General Approaches be started, including within its scope both
The past few years have seen the design and pilot plant investigation, gradually
generalization of thinking about networks. leading to final design, fabrication, erection,
For example, activities have become nodes testing and commissioning.
Although this all sounds fine and rather Although some of these models lend them-
detenninistic, it is in fact highly uncertain. selves to application at specific points,
Any decision criterion used in an Adelson their use must be based on judgment; judgment
procedure suffers from its considerable lies indeed within their use.
simplification of the full needs of the new In all practical situations, resources
product profile of Harris. Furthermore, the provide limitations. All the overt and rational
Adelson procedure makes no allowance for the procedures, which the labours of devoted
expected relative value function as defined by people have made available, cannot be used.
FISHBURN (1964), of the particular company One cannot afford to go through the detailed
directors, or of the social system as a whole. operations: jumps must be made. It is here
Any choice of a portfolio of research possi- that skill, experience, and courage, show to
bilities relies upon some estimate of the the world who are the true designers and the
likelihood of success, which, in the last true managers. It is in the regions just dis-
analysis, will be subjecti ve. Similarly the cussed - in the ability to distinguish almost
planning of any research project is likely to instinctively between the trivial and the
be extremely tentative, and its execution will significant, with relevance to some broader
be replete with heuristic decisions. In any strategy, in an appositely heuristic manner -
case, no research should be undertaken until that the skills of manager and designer come
some exploratory evaluative design has been close together.
carried through, in order to set the result in a Much of lower management work consists
possible practical perspective. Any tentative of the operation of systems already planned
design will, by obligation, necessitate strings and constructed by others. Prescribed pro-
of decisions in uncertainty. Having gradually cedures exist for the motivation and control
worked through the research and development of the operators working within the system.
stages, expending limited resources on those Such a manager has the duty to sense situ-
aspects of the proposed design where the ations in which major change may occur and
greatest uncertainty prevails, suitably ex- to prevent such change from happening. The
pressed in expected relative values, the greatest challenge comes from the obligation
planning of the final scheme is reached. In and opportunity to improve operations. It is
this part of the network, those features are to assist the rational development of such
put down which appear to be significant. acti vities that work-study has been brought
The best use of available resources are forward to deal with smaller problems, and
evaluated by procedures which are largely operations research to deal with larger prob-
heuristic, i. e. taken in the light of the parti- lems. But ·they start largely on the basis of an
cular state of affairs. People are attempting existing situation and with existing resources
to make some of this more routine, by the committed. Many designers work within such
study of possible strategies for particular situations of limited change. They deal with
objecti ves. It must be recognized that these the minor re-organization of material resources.
strategies have no absolute rules stating when The corresponding managers modify the deploy-
they should be used. A simple introduction ment of human resources and the allocation
to research planning and heuristics comes from of individual responsibilities. Each may be
RHENMAN (1964) who advocates further study answerable to some specialist division.
of practical situations.
In the evolution of new projects, design
and management become closely interrelated.
Conclusion The designer, the planner and the manager
A substantial, growing, and justified each suggest possibilities: between them, by
amount of study is being devoted to the close discussion of the critical points, agree-
improvement of project design and management. ment is reached within suitably defined terms
Techniques and analytical procedures have of reference. The difficulties of planning may
been developed which cover much of the field. be solved by the creative labours of the
designer; the difficulties of design may be Reference has been made to the range of
settled by the imaginative response of the models available.
manager. Where one role rather than another Such models are essentially for planning
should act as co-ordinator is a point of dif- and Co-ordination. Decision models within
ficulty and hints about tackling it have been them are as applicable to the designer as to
given by HIGGIN and ]ESSOP (1965). the manager. All models, in the terms of
Within such a relationship of interaction Chapter 17, are very likely to be highly
and response, the new models of procedure simplified and this simplification must be
facilitate action: they help to prevent the continually remembered.
growth of unnecessary tensions; they indicate The networks and trees represent the
responsibili ties. In Chapter 2, McCrory has most detailed model of the planning process
pointed out the advantages that come within that the human mind has been able to achieve.
an organization from a recognition of the They provide a rational co-ordinati ve disci-
requirements of the execution of the design pline of anastomosis and yet carry within
method (task diagram for technical planning themselves the constant possibility of change,
programme). The manager knows his responsi- creative endeavour and heuristic decision.
bility to any part of the scheme. Such parts Although they were developed for planning
of the scheme may, themselves, be set out they gi ve the best models of the design process
in terms of suitable networks, whether dealing itself. It is against such models that design
with marketing, research, or other factors. science will grow.
PART VI
DESIGN RESEARCH
Chapter 32
DESIGN METHODS REVIEWED

J. C. Jones
lntroduction or explored in detail). O'Doherty's own use of

The first British conference on design the word design is in the limited sense of
methods was held at Imperial College in 1962. 'the production of conceptual structures' such
Eighteen contributors from engineering, as drawings, blueprints, etc. He distinguishes
architecture, planning, building, industrial between the generalized, reproducible end-
design, graphics, painting, psychology and product of the design process and the specific,
cybemetics proposed some new methods of unique, non-reproducible end-product of the
designing and offered many descriptions of process of art; he suggests that a scientist
the design process (JONES and THORNLEY, and a designer work from tbe particular to the
1963). In reviewing these and other papers, generat wbereas an artist works from tbe
I have tried to map out and discuss the design generat to the particular. Tbere is anotber
situation as a whole. such distinction wbicb I would like to make:
between a scientist's generalized description,
Definition of Design wbicb be bopes is true for all members of a
It is not always easy to tell what is meant class of things tbat exist, and a designer's
by the word design. O'DOHERTY (1963) points generalized description, wbicb he bopes is
out that it has been used to mean: true for a new class of things wbicb do not
yet exist but wbicb can be made from that
A visual or tactile shape. wbicb exists.
A plan or 'conceptual creation'. RESWICK (1965) bas emphasized tbis
The end-product (of such a plan). view in bis definition of designing as a
The purpose, intention or motivation (of 'creative activity - it involyes bringing. into
the planner). being something new and useful that has not
The goveming or controlling of a process ex1sted previously'. ASIMOW (1962), in seeing
(of cybemetic exploration). engineering design as tbe production of a
A symbol or a set of symbolic represen- 'model wbicb is used as a template for re-
tations. plicating tbe particular good or service as
many tim es as required', agrees witb O'Doberty.
He concludes that a term which is used in so BOOKER (1964), in a similar description of
many different senses is empty of content. He design 'simulating wbat we want to make (or
suggests that its vagueness comes of a con- do) before we make (or do) it as many times as
fusion between sensory-motor skill (which may be necessary to feel confident in the
enables one to wield a pencil), 'phantasmal' final result', empbasizes modelling and
capacity (which enables one to envisage repetition to reduce uncertainty rather than to
detailed mental pictures of things that do not make copies of the product. ESHERICK (1963)
yet exist), and conceptual capacity (which in bis arguments against too much determinism
enables one to form a generat plan to which in tbe design of regions and buildings,
one can confidently commit oneself and one's distinguisbes between (i) a scientist's neutral
resources before the plan has been imagined explanations of past events, and (ii) a de-
295
signer's committed actions which set off end- Limitations

less and unpredictable trains of events into There are limitations which are common
the future. to many kinds of designing. First, there is
Asimow defines designing as 'decision-making never time to collect and assimilate more
in the face of uncertainty and with high pen- that a part of the relevant information or to
alties for error'. MANN (1963) discusses the perform more than an incomplete and over-
inherent difficulty of designing and gives the simplified analysis (Mann). Second, there is
example of a military aircraft which has more the difficulty of spotting errors before the
than ten thousand parts. All these parts must design is well advanced. Third, there is the
satisfy their functions, must be compatible high cost of altering or abandoning designs
with one another, must be adequately strong upon which much time has been spent. As was
and yet of minimum weight and volume. pointed out by PAGE (1963), the designer has
MA TCHETT (1963) in his definition of a 'good' to 'optimize the design, optimize the design
design emphasizes the need to integrate a time and optimize the design costs . . . .
mass of potentially conflicting and possibly Rapidity is vital but this must not be bought
misunderstood requirements into a satisfactory at the expense of performance.' He suggests a
whole. This he calls 'the optimum solution to strategy of spending as much time as possible
the sum of the true needs of a particular set on analysis and evaluation, both of which are
of eireums tances '. cumulative and convergent, and of minimizing
It should be noticed that nei ther this non-cumulative time spent on the synthesis of
definition, nor any of the others quoted, implies a single solution which may turn out to be a
that good designing is measured by any ab- dud. Several alternative solutions should be
solute qualities of the final product. The developed in parallel, but not in detail, until
general view is that the quality of a designed clear evidence of the convergence of one
product is measured by its performance, i.e. acceptable solution is obtained.
by its relationship to the situation in which The aims of the systematic methods
i t is to be used and seen. discussed later are to increase the chances
of adopting Page's strategy and to make it
Design Situations less necessary to develop bad designs in
It is probably a mistake to equate design- order to learn how to develop good ones. The
ing with problem-solving. It is more like common feature of these methods is to formalize
'problem-finding' which has been discussed and make public the designer's thinking in
at length by MACKWORTH (1964). The situ- the early stages. Thus more design effort can
ations with which designers of various kinds be applied at the beginning, when it is most
are faced are less clear than the problems that helpful, and less towards the end, when it can
arise in games like chess or in examination be wasted.
questions. Asimow states: 'The designer is
presented, not with a problem, but with a Complexity
problern situation . . . . it is out of this milieu It seems that all design situations are
of perplexity that clear definitions of the complicated no matter how simple or insignifi-
relevant problems must be drawn.' Not only has cant is the thing that is designed. ALEXANDER
the designer to find problems which he hopes (1964) has discussed the complex patterns of
will prove to be relevant, but he has to en- user requirements, manufacturing requirements
visage the performance, shape and manu- and physical configurations that lie behind
facture of an (as yet) non-existent product and such apparently simple shapes as a kettle or
predict how it will influence, and be influenced an Indian village. STURT (1923) gives some of
by, the original design situation. It seems that the reasons for the tilting and curving of each
designing is not so much the solving of a set part of an English farm wagon and describes
problern as the performing of a very compli- the consequent 'interaction of its parts'. His
cated act of faith. views support Alexander's thesis that a design
DESIGN RESEARCH 297
achieves stability and perfection when the flow diagrams and related calculations of
visible structure of its parts is adjusted system engineering can be applied to predict
throughout to the invisible structures of the performance with reasonable certainty, however
situations in which it is to be made and used. large the system may be (GosLING, 1963).
The modern design situation differs from It seems that there is much to be gained by
traditional ones in that requirements, materials using these techniques outside the military
and manufacturing resources change very much and space-exploration fields in which they
more quickly than they did. New design have been applied with astanishing success,
situations constantly appear, and existing for instance Polaris, Telstar, Early Bird,
design situations tend to change faster than man-in-space.
does people's ability to design products in Products (or Mechanical Systems) - A
which these changes are properly reflected. product or mechanical system is a Slngie unit
This recent increase in pace has not been of closely integrated parts which together
matched by the widespread adoption of faster perform a set of functions. It may be a com•
methods of designing. ponent in a flow system or in an environment,
or it may be used independently, for example
Four Kinds of Design Situation a window, a tractor, a garment, a telephone,
It is useful to distingu1sh between four a pump, or a tube of toothpaste. The significant
kinds of design situation, as follows. thing about product design is complicated
Environment - The designing of regional detail. Both the performance and the cost of
plans and of buildings is intended to provide such a product are very sensitive to small
a 'strategic framework in which other people changes in the physical disposition of its
can operate in detail', or enclosures 'which parts. This difficult design si tuation requires
man inhabits as a biological species' in an very great design effort in proportion to the
'evolutionary situation over which the designer cost of a single product. In the past, this
has no final control' (Page). In this kind effort was spread over a long period, perhaps
of design situation, the design costs have to centuries, of evolution by trial and error.
be recovered from a single item and no proto- If, as is increasingly the case, there is no time
type is possible. The criterion of success is for slow evolution, the design effort has to be
adaptability to changing circumstances which paid for by making !arge numbers of identical
it is not feasible to anticipate in detail and products in a short time. The problern then is
not desirable to limit by an over-structured to simulate the trial and error process using
design. drawings, models, tests, calculations and the
Flow Systems- Thes.e are sets of separate experien~e of designers, before the necessarily
components which tagether perform a well- high tooling costs are incurred.
defined function Examples are: a missile Parts- These are single pieces of material
detection system, an airline, an administrative from which products are assembled. A part is
system, a supermerket or a telephone system; usually designed specially for a particular
on a smaller scale, the combination of washer, product, because of the high degree of inter-
drier, iron, ironing board and airing cupboard dependence that is required between parts if
which tagether comprise a not very coherent product performance is to be achieved without
system for removing the marks of use from high penalties in cost, weight and size. In
clothes. Flow system design includes the such cases the requirements which each part
specification and positioning of components has to fulfil are more easily discovered than
to perform a function, but does not include are the requirements of the product as a whole.
the designing of components. Provided the Less often, standardized parts such as nuts,
system has a recognizable throughput, and bolts, tyres, knobs, pipes, bricks, and the like
provided that the throughpu t is not greatly are designed for situations which recur or when
affected by the physical (as opposed to the interchangeability is required. In these cases
sequential) disposition of components, the it is very difficult to discover the requirements
which will have to be satisfied. The difficulties absence of criteria and methods for dealing
of achieving variety reduction and of getting with these ernerging man-machine systems as
agreement for standard dimensions, probably a whole and in relation to each other.
reflect the unwillingness of designers to reduce The design of products, although a field
their room for manoeuvre by the incorporation in which there is a long tradition of good
of standardized parts, however cheap these design, is at present the cause of failure and
may be. anxiety (Feilden Report, 1963). There appear
to be two sources of difficulty. First, some of
Changes in These Situations the talented people who in the past became
The present interest in design methods engineering designers are now able to enter
would surely not have arisen if the above four occupations which are more highly rewarded
kinds of design situation, each of which has and esteemed: there is left a need to system-
existed for a long time, were not subjected to atize detailed work so that it can be done by
some recent and rather sudden changes. What persons of less talent or experience. Second,
chan ges can be percei ved? there is the growing complexity of products
In the case of environment, there is a and their integration into more closely knit
general and mpid increase in the scale of road systems. This demands of each product greater
traffic and in the rate of building. These new reliability, greater compatibili ty wi th other
demands cannot be satisfied without new products, and greater certainty that it can be
facilities such as motorways and industrialized sold in increasing quantities to increasingly
building. Also needed are new restraints such diverse and discriminating customers. These
as pedestrian segregation, traffic guidance new demands are being satisfied, at least in
systems, dimensional co-ordination of building part, by the appearance of new kinds of design
components, and better environmental Stan- specialist, each of whom is concemed only
dards. Each of these developments calls for with one aspect of designing, such as reli-
much wider and more exact knowledge than is ability, quality, optimization, appearance,
available within the experience of any one ergonomics, cost reduction, system compati-
designer, or design profession, and demands bility, en vironmental testing and so on. These
new techniques by which to anticipate the new kinds of engineers and designers (design
ways in which new and old features of the technologists as they are starting to be called
environment will influence one another. in Manchester) require means of formalizing
In the category of flow systems, there is the design process so that their contributions
growth in the size of organizations and com- can be given their proper place. Formal design
munities and greater competition between methods should also help the general prac-
systems of transport, mass entertainment, titioner designer, who in the past dealt
education, military defence, retail distribution, adequately wi th all aspects of one design by
and so on. Ahead, there is the need to radically himself. He now needs a means of meeting the
alter the present codes of professional conduct new specialists on their own terms, so that
and practice as the activities of engineers, his central position can be recognized and
teachers, doctors, shopkeepers, clerks, so that he can make proper use of specialist
machinist..<>, lawyers, accountants, dustmen, help at the right time instead of treating it as
telephonists, pilots, typists, travel agents and unwelcome interference which he puts off
many other professionals are partly assigned seeking as long as possible.
to such machine components as computer Within the class of engineering products
networks and television links. Rethinking the there appears recently to have been a greater
human or 'software' aspects of the organized proportion of devices for which there is no
man-machine systems that are ernerging in precedent, 'nouveau designs' as they are
these areas, is often a greater design challenge called by both Asimow and Mann. Examples
than is the design of the 'hardware' components. are satellites, prosthetic devices and nuclear
The largest single difficulty is, however, the reactors. The design of products in this cate-
DESIGN RESEARCH 299
goty demands techniques of deliberate inno- adaptability, its favourable combination of

vation which, if successful, have the valuable many properlies and its low cost. In this kind
results of starting entirely new industry of design problern the simplicity of the result
(MouLTON, 1965) and extending the scope disguises the complexity of the needs to be
and opportunity of human life. The formidable satisfied.
difficulties that have to be overcome in
deliberate innovation are in assessing needs Gradual Evolution Versus Nouveau DesiW~
which have not hitherto existed; and in ex- Designers sometimes have a choice bet-
ploring simultaneously a sufficient number of ween the gradual evolution of an existing
alternative solutions to ensure a high prob- product and the sudden introduction of a
ability of 'instant' operational and marketing nouveau design. Consider the early Whittle
success for the version which is chosen for jet engine and the highly developed turbo-
the very costly processes of detailed develop- supercharged Wright Duplex Cyclone radial
ment, production and distribution. A third engines which used some exhaust gas thrust.
difficulty is that of foreseeing novel and dis- Consider also the prop-jets of the 1950's. It
concerting weaknesses or side-effects which can be seen that not only are there marked
did not occur in the less specialized devices physical similarities between the late versions
which the nouveau design will replace. Road of a traditional design and the early versions
accidents, jet aircraft noise near airports, and of a nouveau design,. but there is also an over-
'blacking-out' in the first monoplane fighter lapping of performance levels. The major
planes are notorious examples of such side- difference in the short run lies in the lower
effects. The recent practice of measuring side- cost of gradual evolution. Only in the long run,
effects early in design can be inferred from after further development of the nouveau design
the astonishing safety records of both nuclear as a pure jet, did a major performance dif-
industry and space flight, and can be seen in ference make itself felt. In the middle run,
the attempts to measure the effects of sonic the prop-jet might be regarded as a step back-
boom before the introduction of supersonic wards to get economic advantages which the
transport. pure nouveau design had not yet achieved.
At the smaller and more detailed scale at Another such comparison is that between
which the product designer habitually works, the Boeing 707 and the VClO. Is this a case
there is always a host of less destructive but of a new aircraft having insufficient advantage
no less unacceptable limitations and side- over a more traditional one to make its devel-
effects which are likely to be overlooked in a opment worthwhile, or is it the reverse? It
nouveau design. There is clearly a need to seems to be very difficult to make intelligent
equip product designers with formal methods of judgments on questions such as this, and very
checking for side-effects that may fall outside costly to make a misjudgment.
the experience of anyone in the design group.
The designing of special purpose parts is The Adaptability of Products to Situations
inseparable from the designing of the products These examples suggest a promising
of which they are ingredients. Standardized subject for design research - a study of the
parts, however, introduce new problems and evolutionary paths of man-made things. Are
are in greater demand than was formally the there some laws of artificial genesis and
case. Movement of people and new inter- evolution which could be inferred from the
national agreements call for greater inter- histoty of engineering design? Could a know-
changeability of such parts as electric appli- ledge of such laws permit a better choice
ance connectors, nuts and bolts, tape spools between the many paths which lie open to the
and office stationery. The long-delayed designers of a new product? What factors
mechanization of the building industty calls govern the frequent extinction and the in-
for modern equivalents of the traditional brick, frequent survival of a new product or invention?
which has proved so difficult to equal in its Analogy with biological evolution, taxonomy
and genelies suggests that there is scope here products. The designers of industrialized
for a fundamental advance which could be of buildings might gain much by loosening the
practical importance. many dependencies between the components
Presumably the factor that settles the of a house before tooling-up for quantity
fate of a new design is the response which it production QONES, 1965). It is interesting to
elicits from the situation into which it is see that the furniture industry is following the
thrust. This response can be such as to inhibit lead of Eames in allocating each major
further development, as in the case of early function of a product to a separate component.
gas turbine cars; or, as in the case of the j et Chairs, tables and cupboards are no Ionger
to prop-jet sequence, the situation can induce single entities but instead consist of standard
partial return to the previous norm. Successful body support shells, storage units and legs
new designs, such as railways or digital which can be mixed in many different ways.
computers, however, seem to have elicited Customers can thus be offered many choices,
from situations, which appeared unready to and components can be made in larger
receive them, a reorganization of existing . quantities.
practices and attitudes that not only allowed
the new product to survive but encouraged its The Determination of N eeds
further growth until it became a new norm. The relationship between a product and
J ACKSON (1964) describes how the great the needs which i t satisfies is two-directional.
difficulties of getting the Volta River Project The needs to be satisfied by a nouveau design
under way were overcome by keeping initial do not always exist even when the means of
capital costs low. This involved the sacrifice their satisfaction is under development, as
of some of the project's best features, but can now be seen in the case of space vehicles
allowed it to reach a stage when it could and of the Concord. There must have been a
elicit its own support. In such a case the time when this was true of any successful
purpose of the product is to restructure the innovation such as the ball-point pen. It is
situation rather than to satisfy an existing also true of any unsuccessful development.
demand. Are there some needs which exist before
Alexander has pointed out that the choice suitable products are designed and is there
of components from which a product or environ- any sure way of identifying the needs which a
ment is composed can limit its adaptability new product is capable of bringing into
to existing situations and to unforeseen existence?
changes. He proposes the mathematical mini- It is, of course, the sponsors and the
mizing of the network of dependencies between users of products who ultimately decide which
groups of functions so that local adaptations needs they are willing to pay to satisfy; but it
to change can be made, without the often un- is nevertheless a part of a designer's job to
acceptable penalty of having to redesign the re-assess these needs before spending his
whole system. Alexander has cast a new light sponsor's money on aims that may not be
on the fundamental difficulty of designing: realistic or well chosen. A difficulty here
that of choosing the components of which the comes of the orientation of users and sponsors
product is to be made and of relating these to to the present, and of designers to the future.
each other. He seems, however, to overlook It is doubtful if anybody considers a need or
the performance and cost penalties of mini- aim to be real unless he can envisage a means
mizing the dependence of one component on of satisfying it. Users and sponsors of design,
another. It has been suggested that an aero- being concerned with what already exists,
plane designed on this principle might be too will tend therefore to specify or approve of
heavy to get off the ground. Alexander's aims that can be satisfied with small departures
approach may be more applicable to the design from existing designs. They will not find it
of environments, or to systems of standard easy to believe in non-existent demands which
parts, than it is to the design of mechanical designers, with their antipathy to the present
DESIGN RESEARCH 301
and their power to envisage the future, claim constraints on the designer's search space.
will come into existence when the means of Type (3) will aim at the satisfaction of all
satisfaction becomes available. This conflict new demands which could be expected to
lies behind the professional designer's come into existence as a result of uncon-
traditional distrust of those forms of market strained development to the limit of the current
research which are based on the housewife's manufacturing expertise. Initial proposals for
immediate reactions to new designs of which all three types are prepared before the decision
she has no experience. is made to go ahead with one or more of the
One of the advantages of small organi- alternative sets of objectives and possible
zations is the possibility of bridging this solutions.
gulf of potential mistrust between sponsor and
designer by close ties of friendship, family Designers
relationship or mutual respect. When this Recently, much has been said about
occurs, there may be no need for the checks potentially creative persons, but not so much
and restraints which a more remote management about persons who have actually created
feels obliged to put between designers and something which is valued by others.
their interpretation of future needs. There are GETZELS and J ACKSON (1962) and others
signs that a new profession of design manage- have suggested the idea that creativity can be
ment is springing up to provide this bridge measured by the number of unlikely uses
in organizations where it does not already which a person can think of for an object such
exist (F ARR, 1965). as a brick or a top hat. It appears that persons
who score highly in such tests are not always
Some design theorists advocate the fixing
and weighting of objectives before design the most intelligent. It is pertinent to ask if
analysis begins. Others, with whom I agree, successful artists, designers or inventors
believe in exploring the design situation and would select, as assistants, persons who have
its possible solutions before deciding the the greatest number of ideas; or would they
Iook first for evidence of having created
design objectives, each of which should be
something, and second for the ability to
assigned initially on equal value. The argument
envisage clearly the physical conditions
of the first group is presumably that, without
which decide the feasibility of an idea?
clearly evaluated objectives, no progress will
Mann has drawn attention to what he calls
be made. The opposing view is that objectives
a 'bimodal attributes' of a successful
that are set without knowledge of the feasible
engineering designer. First he needs sound
solutions are likely to restriet a designer's knowledge of the physical principles that are
area of search. Furthermore, the weight which relevant to his area of work: engineering
one would ascribe to an objective is a function teaching has in the past been aimed at provi-
of the solution one has in mind. If one is
ding this knowledge. Second, he needs the
thinking of a Rolls Royce, one will not rate
ability to deal with ambiguity, incomplete
very highly the objective of easy parking.
knowledge and the absence of reliable theory.
If one is thinking of a minicar, one will put
Both of these seemingly Contradietory abilities
parking high on the list.
are needed to convert an unstructural design
A strategy which side-steps this difficulty problern into an accurate description of a
is to have three sets of objectives, and to physical assembly which can be relied upon
seek three sets of solutions. Type (1) will to solve it. The ability to toterate uncertainty
satisfy only the sponsor's most immediate while proionging the search for a sufficiently
aims and will involve the minimum modification reliable solution, and when committing !arge
of existing hardware and software. Type (2) resources to untried ideas, is also relevant.
will involve the assessment and satisfaction
of needs which are not so pressing and are Mental Processes
more expensive to satisfy, but which offer O'Doherty give prominence to the ability
greater advantages to users and impose fewer of gifted persons to carry out simultaneous
translations between different sensory modali- skilled person cannot hirnself explain. He
ties. He has also referred to the ease with suggests that 'kinaesthetic images' (the
which an artist can recall details of the memories of bodily sensations that accompany
appearance of something he saw months or skilled performance in percei ving and doing)
years ago, whereas a layman may be unable are the media through which a performer con-
to recall more than its general character after sciously organizes and deploys the skills at
only a few minutes. The size and contents his disposal. Once triggered off by internal
of the memory seem to be closely related to or external stimuli, the detailed components of
the Ievel of creative. achievement. a skill can be carried out without conscious
Mann describes the pattern of an engineer- control. Thus the skilled performer can direct
ing designer's work as long periods of routine his attention ahead of his actions. It is shown
analysis which he calls 'crank turning or later that skill is likely to become more promi-
grunge', relieved by 'creative peaks'. O'Doherty nent when designing is aided by quickly
implies that the onset of a 'leap of insight' responsive Computers.
is by no means accidental, but is consciously
induced by the undertaking of long periods of Procedures
immersion in details of the problem. BELLO Page, in his review of the papers presented
(1959) quotes Land as saying that for him at the Conference on Design Methods, noticed
sixty hours of continuous work on a problern that in the enormaus range of design procedures
is equal to a year of interrupted thought. discussed, from regional planning to the design
Mackworth suggests that a person capable of of scientific instruments, there was only one
originality has a Capacity for 'effective point of almost complete agreement. It is that
surprise' at small but, to him, significant designing is 'a three-stage process demanding
differences between expectations and reality. analysis, synthesis and evaluation'. He thought
It may be that this surprise is what induces that the cyclic nature of these stages should
the leap of insight to which O'Doherty refers, have been emphasized.
and in which the more deterministic design Mann suggests three major stages which
theorists appear to disbelieve (SIMON and he calls concept, analysis and specification.
SIMON, 1962). O'Doherty refers to 'two camps In beginning with a concept or solution, he
in respect of creativity: one holds the rather differs from the theorists of systematic design
Platonic mystical idea that one is visited by but appears to reflect current practice in
one's daimon and the creative act follows'. engineering and architecture. It may be that
At the other extreme is the Mill tradition that systematic design methods, being largely the
'all one has to do is to put things tagether means of formalizing what designers tradition-
and the result will be a newly created product'. ally do in their heads, comprise an additional
The first view is implied by 'claims that one predesign sequence of analysis, synthesis and
is seized by the unconscious, or by the evaluation, the outcome of which takes the
numinosum, or by inspiration, or by the medium place of \1ann's initial concept.
itself - the brush, or the pen, or the chisel, Asimow has suggested twenty-fi ve stages
which then is supposed to guide one's hand'. for the engin~ering design process. Each stage
The second view is implied by many theories is an iterative loop into which is fed the
'which would repudiate its explicit formulation. outcome of the last stage tagether with new
Thus it is implied by cybernetic models of the information or an appropriate mode of analysis.
creati ve process, by "brainstorming", and by Within each loop is a stage of synthesis
logical posi ti vist approaches.' followed by a stage of evaluation. In a case
O'Doherty makes the point that skill, history of chemical plant design GREGORY
which is a constituent of the creative act, is (1964) describes fifteen major decision stages.
learnt on the basis of an innate endowment. In this example there were three feedback
It is retained within the nervous system not as loops, the most troublesome of which invol ved
knowledge but as performance, which the the rather fruitless recycling over four stages.
DESIGN RESEARCH 303
Gregocy notes that it is impossible to break of changing demands, competitive products

out of such a loop without either feeding in and customer moti vations from which the new
new information or taking a decision that it design is expected to elicit a favourable
is not fully supported by the information already response.
available. The decision which he made was (2) The solution structure - The physical
to choose a heater much smaller than any structure of a possible design.
that had been found to work before, and which (3) The resources structure - Relation-
would operate near to the theoreticallimit. ships between the physical properlies of the
This decision, which was disputed at the available materials, the limitations of available
time, but which apparently led to a successful manufacturing processes and the costs of each.
new product, may be an example of the human
ability to shift from one approach to another The achievement of a good match between
on the basis of inadequate information. Mann these three structures may be a criterion of
takes this to be the major difference between good design. Any particular design procedure
human and computer performance. Mackworth might be judged by its effectiveness, first
believes it to be the main feature of 'problem- in making these structures apparent, second
finding', which he defines as the ability to in minimizing the structural mismatches of
detect the need for a new mental programme chosen solutions, and third in minimizing the
when the evidence is scanty. He points out total cost of the design process.
that humans can do this repeatedly without There are three further structures that
starting afresh each time. He assumes that this seem relevant:
is done by using previous experience as a
code from which the missing part of the input ( 4) The product life structure - The
is read when available information is pattern of growth and decay of demand for a
fragmentary. new product, the performance it achieves, its
failures and its history of modification or
The Recognition of Structure further development.
Inability to describe this coding procedure (5) The designer's mental structure - The
makes it difficult to programme computers to pattem of experience and ideas that is avail-
carry out recognition tasks without human aid. able to a designer when he examines a problern
Mackworth suggests an investigation of the or considers a solution.
methods by which unfamiliar items are classi- (6) The analogue structure - The pattems
fied by the human brain. The points at which a of external symbols or models which the
ciassification is abandoned and a new one is designer uses to represent the structures of
started may correspond to O'Doherty's leaps the situation, the solution or the resources.
of insight and Mann's creative peaks. On such
occasions, the designer may be thought of as Each of these last three structures is the major
a circuit that is capable of damping small constituent of a method of designing. The most
signals of mismatch between the structures of primitive of these, and also the most reliable,
the classification scheme and · the structure of is the method of trial and error. A failure which
the things classified (only some of which have occurs in the life of one product is corrected
yet been seen). He may . also be said to be in the next without re-assessment of the design
capable of resonance when there are signs of as a whole. The finest products of traditional
similarity between these two structures. The craftsmanship were reached in this way without
design process, with its orientation to what the intervention of any designers or other
does not yet exist, involves not two but three specialists. The developing of hand-built
structures that have to accord with one another. prototypes by trial and error is a modern
I find it useful to think of these as: version of this ancient technique.
The designer's mental structure plays the
(1) The situation structure - The pattem main part in the second method of designing -
that of simulating the product and exploring search for interpretations of the problern and
its properties in the imagination, aided by for solu tions to i t.
calculation, drawings and informed experience. The essentials of each of the systematic
This is of course many times faster than methods to be discussed are (i) to break the
evolution by trial and error, and permits the problern into pieces, (ii) to solve each piece ·
re-assessment of the design as a whole. by itself, and (iii) to combine the pieces into
Provided the designer's experience of the a new whole which may surprise everyone
design situation and of the resources is ade- including its designers.
quate, this method is very quick and reliable. This fragmentary treatment of a design
It is the method by which most industrial problern is opposed to the development of
products have been designed. Its weakness is overall concepts which Asimow and others
its dependence upon past experience. As Page take to be the basis of engineering design.
points out, experience is a douple-edged Ilowever, if innovations in design are compared
weapon - it saves time and it saves thinking. with what preceded them, it can be seen that
The greater one's experience the rnore difficult this kind of breakdown and reconstitution is
it becomes to restructure one's thoughts to in fact achieved. In each of the innovations
match the structure of new design situations, that is shown in Table 32.1, the novel solution
new solutions and new resources. is composed not of rearrangements of the
External analogues of the design situation, existing parts, but of parts which are new in
of possible solutions, or of the resources themselves. The divisions of reclassification
available, are dominant in the third method of of the problern into new functional components
designing. The aim here is to represent, which these parts satisfy appears to be the
outside the brain, the major part of the design creative step. The inventors concerned may
process, so that it becomes visible as a whole not have set out to systematically reclassify
(not piece-by-piece as it is in evolutionary existing functions but surely they were driven
trial and error). This externalized designing to take this step in moving forward from what-
is no Ionger tied to the experience of one ever were their conceptual starting-points.
person. Greater leaps forward in design are It may well be that the second task of an
possible because there is unlimited opportunity inventor, that of eliciting sufficient response
to restructure and test the relevant information to his invention, is dependent on the rate at
patterns before finalizing the design. This which his reclassification of functions can be
systematic restructuring of thought enables appreciated by those who could use the in-
a designer to explore rnore widely, and to test vention. Only when his reclassification is
more precisely with reference to experience understood, can the administrative reorgani-
that is not necessarily his own. His skill has zation that is necessary to exploit the invention
initially to be directed at the linguistic problern take place. It is not until an invention has
of creating a suitable problem-language which elicited this response, that it can be said to
implici tly defines boundaries within which a have become an innovation. Man-machine
host of alternative solutions lie. The design system design procedures include the formal
of flow systems is perhaps the only field in consideration of this problern of assimilation
which externalized designing has been suc- of 's()ftware' at the initial stage of design.
cessfully applied to problems that are too It may be that in the increasingly organized
large and too complicated to be formulated in world, the changing of software is becoming
any one brain. more difficult than the changing of machines.
Table 32.1 resembles the morphological chart
Methods of Viverging that is described by NORRIS (1963). He shows
The common feature of the so-called how the use of charts can oblige a designer to
systematic methods of designing (they are think of several solutions for each of the major
really rnethods of handling design information) functional requirements, and how these solutions
is that they permit a widening of the area of can be combined to form thousands and sorne-
DESIGN RESEARCH 305
Table 32.1
Existing Existing Existing classification New classification New New

whole parts of functions of functions parts whole
Cut-throat Blade Cutting Cutting edge Thin blade Safety razor

razor
Handle Control Stiffness of edge Blade clamp
Strop Sharpening Control of angle Edges of clamp
Control of strokes Handle
Maintenance of edge New blade
Retail Counter Buying and selling Selection of goods Self-service displays Supermarket
shop
Shelves Availabillty of goods Payment for goods Checkout area
Replacement of goods Service staff
----- ---- ·--- -
Aero Engine Power Admit air Inlet Jet engine

engine
and Propeller Thrust Aceeierate air Compressor
propellor
Fuel Power to engine Expand air Bumers
Expel air Outlet
Power to aceeierate air Turbine in expanded air
Power to expand air Fuel
times millions of alternative designs. A less problem. For some time at Manchester, an
formal method of achieving a wide range of elaboration of brainstorming had been used
ideas is the 'brainstorming' meeting proposed for collecting quantities of design information
by ÜSBORN (1963) and by other advocates of (]ON ES, 1964). Each person concerned is
creativity in design. Persons of varied ex- asked to read journals, examine existing
perience are asked to suggest any conceivable products, talk to users, and speculate privately.
way of tackling a design problem. The in- He writes each relevant thought or fact he
hibiting effect of criticism is avoided by a rule comes across on an index card. The cards are
that no idea is to be evaluated until the read out in random order at a meeting and many
meeting is over. TAYLOR, BERRY and BLOCK more suggestions are then made, each of which
(1958) have shown that group 'brainstorming' is written on a card. The structure of the
does not produce better ideas than does problern is examined by classifying and
solitary . thought. There is, however, little reclassifying the cards in different ways until
doubt that it is an extremely quick way of a structure that seems realistic to the designer
extracting information from the memories of emerges.
persons whose experience is relevant to the Page describes these techniques of
306 THE DESffiN METHOD
divergence as 'planned relearning in a frame- far unless there is evidenee of eonvergenee

work that forces divergent rather than con- on an optimal solution. MARPLES (1960) has
vergent thought'. The difficulty here is to pick shown how engineers direet their knowledge
out usable ideas or approaches from the mass and experienee to the avoidanee of blind-
of material which is generated. Unfortunately, alley design deeisions whieh are likely to
neither brainstorming nor morphological charting ereate difficulties at later stages.
includes a reliable way of doing this. Ex- In suggesting a fairly formal path of
perience of card sorting suggests that one may eonvergenee on one solution Matebett (1963)
nevertheless benefi t from the restrueturing of and Asimow both state that goals must be very
one's thoughts. clearly stated and fixed at the start. Asimow's
}ONES (1963) and Matebett both deseribe major stages in this convergenee are:
more controlled methods of widening the area
search. Matebett suggests beginning from a (1) Feasibility - A set of feasible con-
definite starting point, such as a tangible cepts or outline solutions.
weakness of an existing design, and then (2) Preliminary design - Selection and
permuting possible eauses and remedies in a development of the best concept.
formal way to expose the whole problem. My (3) Detailed desigR - Engineering des-
method is to write performance specifieations cription of the eoneept.
for eaeh of a num ber of eritieal requirements
( 4) Planning - Evaluating and altering the
and to eombine partial solutions into eompatible
sets. A diffieulty with these methods is that concept to suit the requirements of production,
eonflicts arise between the structure of the distribution, consumption and product retire-
analysis and the strueture of one's thoughts. ment.
MA TCHETT (1964) advocates the use of social
pressure to induee a designer to abandon his ROSENSTEIN (1960) put forward a system
old thinking proeess and to stick to his design procedure which is similar:
analysis. I favour the abandoning of an analysis
in whieh one has lost confidence, and the ldentification of need.
starting of another which is closer to one's Information collection.
present view of the problem. This diffieulty Identifieation and statement of system
variables:
of reconeiling formal methods of exploration
inputs
with the freedom to ehange one's mind, in
outputs
response to signals of mismatch between
transforming means
analogue strueture and mental strueture, is
constraints.
something to which more attention could be
Criteria for optimum design.
given. Synthesis for:
physical realizability
Methods of Convergin,g economic worthwhileness
At this stage in designing (which for financial feasibility.
Asimow, Mann, Moulton and others appears to Optimizing and sufficing.
eome very soon after the start) evaluati ve Test and evaluation.
rather than exploratory techniques are needed. Iteration.
PAGE (1964) diseusses the strategy of begin- Communication, implementation and presen-
ning with models of alternative solutions that tation.
are as rough as can be tolerated, and testing
them in an analogue of the design situation. The difference between these descriptions of
This intentional roughness avoids spending engineering design convergence and some of
design effort on detailed studies of designs the more divergent systematic techniques is
that may later be found to have major faults. that, in the latter, solutions to parts of the
A di verging idea must not be developed very problern are evaluated and converged upon
DESIGN RESEARCH 307
before the general form of the design has be quicker, and the in-out sequence is likely
emerged. In the former, a concept for the whole to be more reliable. I have not come across
design is chosen before detailed evaluation evidence that it is essential to use a combi-
takes place. nation of both sequences, or that there are
design situations in which one or other of the
Strategies
sequences, in its pure form, has a clear
Asimow suggests that the choice of tech-
advantage.
niques for tackling any problern is a unique
tactic. It is likely that any of several such Computer-aided Designing
tactics would reach an acceptable solution Questions of design procedure which are
provided there are plenty of iterations in each. of academic interest when design thinking
The route chosen may be influenced by the occurs in one person 's brain, become of
designer's temperament and preferences, but practical importance if that thinking is to be
the solution reached must be workable and shared with a digital computer. As Mackworth
must be shown to be so. Sirnon and Simon, in (1964) and Mann point out, it will one day be
their computer simulations of chess playing, possible to provide designers with individual
have shown how the best players have success- and speedy access to computers via keyboards
ful strategies which Iead from one to another and quickly-responding displays of the
without exploration of more than a fraction of 'sketchpad' type (SUTHERLAND, 1963). The
the alternatives and without needing to think major weakness of present Computers, that of
further ahead than the location and identity being unable to change programmes quickly
of the next strategy. Such behaviour seems (Page, 1963) will thereby be removed. As
alien to the following through of a pre- Mackworth (1964) has remarked, 'no other
established series of stages in design, but not devices in the world are quite so badly de-
to the seemingly impetuous changing of tech- signed from the point of view of ease of human
niques as the work proceeds. use'. Mann envisages programmes which are
Two strategical points can be mentioned. not fixed in advance but which are capable of,
The first is 'minimum commitment'. This is and require, human intervention at crucial
Asimow's term for the principle of taking only points. He suggests that the languages used
those decisions which are necessary to the should be graphical and verbal, as well as
stage that has been reached in the design mathematical, and that they should permit the
process. Other decisions should be deferred designer to explore complex situations that he
so as to leave maximum room for manoeuvre at hirnself does not fully understand. The time
later stages. This principle is difficult to of response should be equal to, or less than,
stick to if there are feedback loops over several the designer's 'cognition delay time', so that
stages ansmg from interactions between he can 'mold, shape, interrupt and redirect the
detailed and general aspects of the product, a computer's manipulations' in response to his
difficulty that is particularly noticeable in evaluations and judgments of a developing
architectural design. situations. This is the antithesis of the present
The second strategical point is the choice use of computers for the completely automatic
between in-out and out-in design sequences. exploration of those sections of design prob-
The design of a house can begin with an lems which are well understood.
exterior into which is fitted a plan and then BOSTON (1963) has successfully used a
rooms into which is fitted equipment. Alter- Computer programme which is dependent on the
natively the designer can begin with activities responses of an experienced designer. His
and equipment and move outwards to rooms, to experiments suggest that the effect of the
a plan and lastly to exteriors. In practice he computer aid is to permit the designer-computer
may use both sequences in turn. This is the combination to make better-informed leaps to a
familiar conflict between starting with overall new design possibility, when the designer
concepts or starting with solutions to parts of decides to reject the last design possibility
the problem. The out-in sequence is likely to which has been printed out. This is because
11
the machine can take account of details of the Computers have already been used to
designer's previous judgments which he him- generate manufacturing information for new
self cannot recall in detail. The effect of the products that can be made up of existing
designer's interventions i:;; to direct the process component types according to standard rules.
of automatic exploration away from unfrui tful The critical problern of capturing the design
searches which might otherwise overtax its logic of the engineering designers who formerly
storage capacity. It is thus possible to jointly did this work, has been solved by the use of
undertake problems of a very large mathematical decision tables for recording each item of
scale. design procedure (IBM, 1962).
Mann picks out speed, memory and reli-
ability as the predominant characteristics of a Organization of Designing
Computer, and suggests that it be applied to the Page (1963) suggests that the organizers
mechanization of all design experience that the of creative work have two main functions:
designer hirnself understands. What he does not First to provide talented persons with the
understand, but is nevertheless capable of best available tools, and second to provide a
dealing wi th, is left to the human intelligence. framework within which they can work with a
Anything that becomes understood in the course feeling of security rather than insecurity.
of joint exploration is thereafter the responsi- Mackworth (1964) notes the outstanding fact
bility of the machine. 1\ßann's creative peaks that the social and intellectual environment is
are thus crowded together because the machine of much greater importance to original work
takes over most of the 'grunge'. I imagine that than is the physical environment.
this change would greatly aceeierate innovation, I have previously suggested Qones, 1963)
and raise the intensity of a designer's work to that a new kind of design organization may be
something like that of a tightrope walker or a necessary to permit a complete change to
concert pianist. systematic work. The elaboration of the pre-
Mann also suggests that computers should liminary stages of design is likely to require
be used to store details of standard parts, the the setting up of specialist predesign sections,
properlies of materials, standard procedures, which are insulated from day-to-day contin-
and the history of previous designs, successful gencies and which operate on Ionger budgeting
and unsuccessful. The designer should be periods than are normal in design and develop-
presented with the possible alternatives from ment. The cost and time of this extra work
this store of information 'upon making an early in designing would be justified only if
absolute minimum query to the computer'. At the total development costs are lessened, and
intermediate stages the computer should be if the tendency to over-run deli very dates is
able to check the geometrical compatibility thereby kept under better control.
of parts, and to plot the repercussions of Mackworth (1964) concludes that a single
changing a particular detail. When a satis- brain is better than a group of brains for
factory design has been reached, the computer deciding, at the start, the general character of
'should generate drawings, parts lists, numbers an investigation. He also notes thatco-operation
of fasteners, etc., as well as prepare the from persons of diverse experience is of more
director tapes for the numerically controlled value before and during the development of an
machine-tools which will produce the product'. idea than in subsequent assessment of its
Mann does not consider the possibility that worth. Environments which encourage acci-
computer-aided designing would release the dental meetings of reasonable duration and
designer from his present need to think in frequency are valued in this respect.
terms of solutions to the problern as a whole. The assignment of problems to inter-
The prior exploration of solutions to parts of disciplinary or interprofessional groups seems
the problem, which is a feature of systematic to be essential if really novel problems are to
design as we now know it, might be a more be tackled successfully. Some of the diffi-
suitable basis for computer-aided designing. culties of doing this have been described by
DESIGN RESEARCH 309
LEWIS (1963) in his precise and involved groups are established, there remains the
description of the reasons why group members problern that the persans concerned feel cut
are either unaware of the misunderstandings off from their chances of promotion if they
that are almost certain to arise between them, stay too long away from their own fields.
or are unable to progress on a broad front once Despite these and other difficulties, some
they realize what is going on. He suggests ways of enabling mixed professional groups to
that means must be found to give each member deal with design situations will have to be
an influence commensurate with his knowledge found. Otherwise society will be overwhelmed
of each of the topics that are discussed. Page by large-scale problems such as traffic con-
(1963) has referred to the example of the gestion, road accidents, crime detection, food
building industry in which the different pro- shortage and housing shortage, each of which
fessional groups attempt to pursue incompatible seems incapable of solution by the unaided
strategies set by their different frames of efforts of the professions that are traditionally
reference. When effecti ve interprofessional concerned.
Chapter 33
TECHNOLOGIES AND VARIETIES OF DESIGN

W. E. Eder
lntroduction possible solutions for an appreciable length

Chapter 3 of this book has outlined the of time. The designer should avoid 'falling in
general design process in an industrial en- love' with any one solution, delaying a decision
vironment. This process seems to be generally on which solutions to drop until the decision
applicable to most technologies, and even can be made on objective criteria, or at least
finds some parallels in the design of research until the criteria are very much clearer than
apparatus and experiments. ~v1ostly, the pressure at the outset.
of the environment forces the engineering This state of mind, although difficult to
designer to work in the way that has been attain, should yield the best possible solution
seen to give quick results, and allows him in the time available. More time can be con-
little time to experiment with other methodolo- centrated on the earlier stages, and a better
gies. The methodologies outlined in Chapter 3 basis created for the final design of the pre-
were: ferred solution. The obvious disadvantage is
that the 'boss' does not see enough on paper
(1) Experience - developed by the indi- in the early stages, especially very few
vidual in his own way, du ring his working life, drawings.
to deliver new solutions. It does not seem necessary to work through
(2) Modification - redesign after ex- a problern whilst adhering rigidly to any of
perience to take new circumstances into these methodologies. Good designers use
account. routines when they consider them helpful,
(3) Check-lists - an attempt to list all and drop back on experience when this can
possible influences, in order to channel and lead to a quicker result. Some of the existing
free the designer's mind. technologies are now discussed in this light.
( 4) Design tree - a method of recording'
the steps in a decision process, in such a way
that the designer can review his progress and Technologies
recall the reasons behind his decisions. Mechanical Engineering
(5) Fully systematic methods - a form of I propose to start with the technology with
data processing to free the designer's mind which I am most familiar, and to which many
still further, and allow him to tackle more other technologies turn for at least a part of
complex problems with better records of their ultimate hardware - mechanical engineer-
previous work. ing. Most of the innovations in this field have
(6) System search methods - useful when been attributed to lone designers of no little
a system must be found, and the components or genius using methodology (1) listed above.
elements are already in existence or can be Many names spring to mind immediately - Watt,
made with existing technology. Stephenson, Brunel, Kaplan, da Vinci,
Issigonis, Whittle, etc. Their work has been
In methodologies (3) to (6), the purpose is to continued and improved by nameless hosts who
allow the designer to follow two or more have relied mainly on methodologies (1) and
311
(2), although in some cases methodologies (3) the same production organization. Check-
and ( 4) have also been used. The salien t lists and checking personnel can reduce the
feature of most artefacts produced in this way number of errors introduced in detail design,
is that they have been regarded as separate and should be used frequently. See, for further
'machines' with a specific job to do, but detail, Chapter 22.
largely independent of the more subtle inputs, The position in jig and tool work is very
and with little regard to the eventual output; similar and depends to a large extent on crafts-
the drawing is extremely important both during manship and its thorough appreciation. There-
the thinking (incubation) stages, and for fore methodologies (1) and (2) are the most
communication (see also CLAUSEN, 1958). frequently used, although much of the work is
Only very recently has this fraternity (among done with standards and recommendations
which I count myself) started to realize that a (check-lists in an extended sense) in constant
diesei engine, for example, is a system with a use. This situation is not likely to change
feedback loop (through the governor to the fuel much in the foreseeable future, except where
pump), and it can therefore be investigated for automation rears its head.
dynamic response by using mathematical In the machine-tool field, much traditional
models; also, that a machine-tool is a part of a work is still done - methodologies (1), (2) and
production system, connected by a transport (3). This is being augmented by the automatic
system with other machine-tools, and therefore control of machine-tools, by both interchange-
queueing theory and operational research able cams, as in the repetition automatic
methods are applicable to it - which implies lathes, and by numerical control programmes
using statistics. (position and continuous path) which require
Mechanical engineering designers are considerations of the response of the machine-
frequently concemed with mechanisms and tool and its control apparatus. This is the
their components. The optimum synthesis of system engineering approach (6). Numerical
mechanisms has also recei ved much attention control is tending to make the detail
recently, resulting in works by BEYER (1965), (component) drawing obsolete, since a punched
TAO (1964), FREUDENSTEIN and SANDDR or magnetic instruction tape is needed and the
(1961) and others. This requires an investi- 'draughtsman' or detail designer must learn
gation of the need for a mechanism (e.g. the this new 'language of communication'. Some
bucket movement of an overhead loader) and notes on trends appear in the informal dis-
the motion it should perform, and this synthesis cussion published by the institution of
can then give the principal (functional) dimen- Mechanical Engineers (1965).
sion of the components needed to produce the
motion. The positions, velocities, accelerations Powered Manual Equi.pmont
and forces in each component must be deter- Hand-tools, although basically mechanical
mined. Their material dimensions must be engineering artefacts, usually with strong
calculated to prevent static failure under electrical Connections, have some extra
maximum load, with checks on other failure problems of aesthetics and ergonomics super-
modes such as elastic buckling (overall in imposed. The shape of an electric hand-drill
bending or shear, and local), creep, and fatigue is determined partly by the electrical and
(in stress-concentrating steps and fillets, or mechanical interior. But it is especially deter-
due to rolling or sliding contact). This latter mined by the need to apply pressure by hand
procedure must take due account of the in a well-specified direction with reasonable
materials, method of production, etc., to yield comfort, to avoid muscle fatigue and cramp, in
a working specification (drawing) of the the usual conditions (oily hands?). The drill
component. The parallels to the general design must also appear pleasing.
process of Chapter 3 are, I think, reasonably Domestic appliances are dominated even
obvious; the sponsor (the system designer) more by aesthetics, which must be included in
and the customer (the workshop) are here in the design process in a prominent position
DESIGN RESEARCH 313
during problem-solving, tagether. with the equipment. However, these considerations are
consideration of noise. The market is very being forced by the application of tariffs -
'looks' conscious. In both these cases the although the results may only be a 'hit and
usual approach is via methodologies (1) to ( 4), miss' approach to power factor corrections.
although (6) is sometimes required where the
'automatics' are not merely set-time switched, Hydraulic Engineering
but controlled by some process variables. Hydraulic engineers are in a much more
favourable condition since they must consider
Electronics the system. Hydropower works on the environ-
Electrortics has been one of the traditional mental system of rainfall collection and
hot-beds of the systems approach, using drainage. In spite of this, methodologies (1)
standard components like valves, transistors, and (2) play the largest roles in design work,
resistors, capacitors, etc. Methodology (6) is coupled with model tests in an attempt to
almost universally used for design of equip- eliminate the imponderables. Hydraulic trans-
ment; the components are more frequently mission work is very system-conscious and the
developed from results of pure research by use of methodology (6) is frequent, even though
trial and error technology on the shop floor - the number of available elements is probably
methodology (1) or (2). This industry is larger than that of the electronic engineers
discussed in more detail in Chapter 34. (mainly because a frequent transfer of
Data transmission has some peculiar electrical/mechanical-to-hydraulic and hydrau-
problems of its own, especially as it is con- lic-to-mech anical/ electrical f u n c t i o n s is
cerned in one of its methods with microwave required to design an effective and useful
electronics, guided in cavities that are system).
surrounded by mechanical engineering problems
of materials, accuracies and production Chemical Engineering
methods. Therefore all the methodologies find Chemical engineering also has some
some use. peculiar problems, in particular regarding the
chemical behaviour of matter (reaction
Heavy Electrical Engineering dynamics) and the artefacts needed to contain
Design work in electric power is distri- the media. This technology has a great interest
buted between at least three different branches. in continuous working. Methodology (6) should
One branch is concerned with the distribution be found very useful to determine the possible
of electric power, and therefore uses the strings of reactions, separations, heating and
system concept with all its analytical forms, cooling, distillation, etc. (termed 'unit opera-
as indicated by MORTLOCK (1964). Design is tions' and 'unit processes'). It should also be
restricted, according as to whether the need useful in attempting transformations into
for expansion is progressive, and whether hardware units (distillation columns, heat
additions to an existing system to increase the exchangers, reactor vessels, filters, etc.) in
magnitude of input and output can be covered the most economical way. At present, method·
by basically analytical techniques. Power ologies (2), (3) and ( 4) tend to be used. Once
generation technology provides the means of the chemical process and the flow design
adding to the input. It is concerned with (determining the space required for the re-
systems for control, with mechanical and civil action) have been achieved, the mechanical
engineering (the separate machine idea) for the design of the separate hardware units can
generators and prime mover equipment and with proceed methodologies (1) to ( 4). Their layout
the design needs outlined by TURNER (1964). can be determined using all available tricks,
The way in which the output is absorbed, including models and analytical calculations
does not yet show much consciousness of the with due regard to cost, heat losses, etc. This
systems idea, in particular regarding the wider and conscious picture of chemical
'upstream perturbation' effects of connected engineering design is still rare within the
technolo gy. Much of the hardware design mechanism suggested by the sponsor, and
concerns pressure vessels and involves inter- certain minimum (and/or maximum) dimensions
pretation of the relevant codes. detennined. Limiting conditions will be set
by frictional slip (where forces must be trans-
Camplex Problems mitted from one contacting face to another),
1\'lethodology (5), hardly mentioned so far, contact stresses, bearing loads and load
has been developed to cope with far more carrying capacity, lubrication, wear, life, or
complicated problems, such as occur in similar considerations for a mechanism; a
municipal engineering, many architectural and structure is limited by deflection, rigidity,
planning situations, transport system design, collapse loads, etc. Certain assumptions must
and nuclear power generation. The problems already be made to solve the equations.
are aggravated by the large number of vari- Evaluation criteria derived from this study
ables, and the scarcity of definite information. are set in terms of the sponsor's requirements,
As this methodology becomes more universally cost, overall size, and possibly aesthetic,
known and accepted, its range of use will ergonomic and other considerations. The
probably be extended. It requires conscious, sponsor should again approve the work up to
deliberate and careful application and the first this point.
few times will probably prove a disappointment- The problem-solving step is concerned
'I could have clone it easier by guesswork'; but with preventing failure of structural· integrity
I feel it is worth persevering, especially as in the component. The effects of load, the
Britain's position in the world today depends actual (and assumed) distribution of stress in
not only on innovation and research, but even the component, and the failure modes of
more on the application of these in profitable yielding, fracture, fatigue, creep, structural
industries. instability (buckling), etc., must be dealt with;
also the effects of surface finish, stress
Components concentration, corrosion, etc. All of these are
Once the system or the basic function of amenable to calculation in an analytical
the mechanism has been decided, it must be sense, but design (the determination of the
broken down into the elements and components necessary dimensions and geometry) demands
from which it is to be assembled. Elements an inverse approach which is not always
and components are frequently of a mechanical possible. Problem-solving is therefore again
nature, although a parallel may be found in a cyclic process which takes all possible
purely electrical components, and therefore failure modes and inaccuracies of available
intend to show that the sequence of design data into account.
for a mechanical component fits reasonably Only two failure modes mentioned above
into the pattern described in Chapter 3. show a direct relationship between load and
The sponsor is invariably the designer the properties of a cross-section: yielding,
who decided on the overall scheme of function. and buckling of a perfect column. In each case
His requirements are stated in terms of power, the stress is assumed to be evenly distributed
speed, flow rate, pressure, force, etc., and the throughout the section, which is sufficiently
variation with time. In order that the system close to reality for ductile materials. Minimum
shall not fail, the component must be designed sizes may therefore be obtained for the cross-
in such a way that it will not fail either in section to carry the maximum load (tension,
function or in structural integrity. Therefore compression, torsion, bending) without failure
the analysis of the sponsor's requirements by overall yielding or buckling. If necessary
must be based on the modes of failure. simplified assumptions, with increased
As the sponsor is concerned mainly to safety factors may be used, followed by
prevent functional failure, this must be the checking with the full load. The other failure
subject of the feasibility study. The laws of modes depend in some way on the shape of the
mechanics, etc., must be applied to the outline component and the local geography. It is there-
DESIGN RESEARCH 315
fore important to decide what steps, enlarge- component and their effects on the other
ments, fixtures, etc. are required for the components, or between the sizes of different
component to function properly. The component parts of a component, sometimes even between
must then be checked to establish whether the operational principles. Once more, the
other modes of failure can occur with the advantages can be seen of dealing with more
proposed geometry, and whether the geometry than one solution until a clear and objective
can be modified to prevent this failure mode ranking of the solutions can be obtained.
(by reducing the elastic stress concentration, The general design methodology covering
improving surface finish, possibly by increasing this sequence is the check-list methodology
the basic dimension, or altering the force (3), with some use of the design tree (4). It
application). Examples of this type of approach contains a large number of calculations, some
may be found in EDER and GOSLING (1965), of doubtful accuracy, and requires wide ex-
and in a paper by EDER (1964). If this approach perience to cut this work down to a minimum.
is used, the safety factors need only cover Calculations should be made wherever reason-
inaccuracies in data. An attempt to cover able doubt about structural integrity exists:
other failure modes must fail due to their 'running redesign' on the developmen t test-bed
dependence on local geometry. is no substitute for careful work in detail
In the real engineering situation the design.
sequence of events is usually not so Straight-
forward. The shape and size of any one Acknowledgments
component can and does react on other com- I wish to express my sincere thanks to
ponents in the same assembly. The problern Mr. S.A. Gregory and l'v1r. W. Gosling for their
then exhibits an extra cyclic loop, feeding help in preparing this Chapter and Chapter 3,
back from the failure considerations of the and the University College of Swansea for the
individual components to the overall function. opportunity to indulge in work of this nature.
In effect, the dimensions of the assembly and The opinions expressed, in particular regarding
its components must be optimized to produce the uses and usefulness of techniques and
an economic compromise solution. This requires methodologies, are my own, and I take full
some skilful juggling with sizes of each responsibility for them.
Chapter 34
PRELIMINARY RESEARCH INTO ELECTRONICS DESIGN
H. V. Beck
lntroduction of design can now be seen ernerging as a

An ad hoc Committee on Electronics discipline, but up to two or three years ago
Design has been set up within the Electronics most work on these lines was of a speculati ve
Division of the Institution of Electrical nature (BECK, 1961). No formal papers had
Engineers to study the basic design processes been published on design processes in the
and organizational procedures used in the electronics field, and the few pioneering design
electronics field, and to determine if any gaps philosophy papers in other engineering disci-
exist in the information available to the plines (MARPLES, 1961) tended to concentrate
electronics designer. The work of the committee on design problems in which there was a well-
in preparing to carry out its mandate may be of specified objective and on the purely logical
interest to corresponding bodies in other aspects of design, both of which relate to only
fields and is outlined here. a small part of the electronics design field.
The flexibility of electronic techniques makes
Ad Hoc Committee on Electronics Design electronics design problems particularly open-
Most members of the committee were ended and, because of the complex behaviour
invited to serve because of their interest in of electronic circuits and systems, logical
the manner in which design is carried out. design forms only a part of the electronics
All are professional engineers and most are designer's acti vities.
electrical engineers, at least four of whom
have had extensive experience of design in the Conference on Electronics Design
electronics field. Specialist experience of In view of the lack of published material,
members of the committee covers such topics various concepts and proposals were evolved
as aesthetics, sociology, management and a priori by members of the committee before it
teaching. One member is the nominated repre- held its ·first meeting in September 1964, and
sentative of the Electronic Engineering Asso- it was decided to hold a conference in February
ciation and another of the Naval Training 1965 to obtain reaction to these ideas. The
Department of the lVIinistry of Defence. conference itself was something of an open-
ended design problem. The committee was
Publi shed Material acting as an inexperienced design team for
One of the principal problems facing the unspecified customers' unknown needs, using
committee was the lack of published work slender and unfamiliar resources. In a manner
about the manner in which design is carried appropriate to the electronics field, it was
out in the electronics field. There are numerous decided to organize the conference on a
papers covering the design of devices, cir- 'breadboard' basis.
cuits and systems, but almost without exception Speakers were sought and invited to cover
these concentrate on the purely technical such topics as characteristics of the equip-
aspects and on the end-result of a design ment; the emphasis given to particular
exercise, rather than on how the particular characteristics in various fields of application;
configuration has been evolved. The philosophy studies of problem-solving activities; various
317
aspects of industrial design; information or convenient notation for relating each facet
required by the electronics designer; and the of design in a given field to a common set
teaching of electronics design. of concepts, it is difficult to see how an
In arranging the conference, the topic of applicable generalized notion of design can
circuit design was deliberately avoided although arise. It was therefore decided to concentrate
it did in the event appear in the teaching on studying electronics design in depth, taking
session. Considerable interest in the con- as much account as possible of work in other
ference was expressed by staff of uni versities fields. Concepts and models can be compared
and colleges of technology concerned with later with those of other fields and appropriate
introducing more design work into their elec- adjustments made to fit more general models
trical engineering courses, and by admini- of design. Specializing in electronics design
strators, design managers and chief engineers can also be justified by the greater ease with
professionally concerned about the efficiency which the knowledge generated may be applied
of design. It was important that the views of to the training of engineers to design electronic
designers themselves should be expressed at equipment.
the conference, and a special effort was made
to arouse their interest. An article by BECK Hierarchical Considerations
(1 %4) was published in the Journal of the The electronics design field is itself
Institution of Electrical Engineers gi ving the large enough to make a choice of area within
background to the conference. the field necessary for initial study. There is
The conference proved very satisfactory. a hierarchy of complexity extending from
Some 300 engineers, including designers, supersystems (as required for global satellite
teachers and managers attended the conference communications) down to the parts and materials
and there was a useful exchange of views. from which devices (such as transistors) and
components (such as resistors, capacitors or
Electronics Design inductors) are manufactured. Table 34.1 gives
Enquiries into design processes and some examples at various levels in the
practice are now being carried out in several hierarchy of complexi ty.
engineering disciplines. Useful concepts and By examining, in a general way, the
results of possible general validity are being attention given to various characteristics at
promulgated and one approach to the study of different levels in the hierarchy, one comes to
electronics design would be to examine these the conclusion that the greatest di versity of
results and seek to apply them to the elec- characteristics occurs at about the equipment
tronics design field. There was a considerable level, (see Figure 34.1). Size, weight and
incentive to do this in the Institution of appearance, for example, are characteristics
Electrical Engineers, which covers both the of a television receiver but not of a television
purely electrical and electronic fields, with network. The electronic expertise required for
examples of design ranging from power stations design at the supersystem level is practically
to transistor radios. Concepts of engineering none, and increases towards the systems and
design covering all fields are, however, likely equipment levels and reduces again at the
to be too general to apply to particular disci- devices and components level where the know-
plines. For example, words have quite different ledge required is predominantly of physics.
meanings and significance in different Considerations such as these have led to the
engineering disciplines; the word 'drawing' confinement of the field of study to the equip-
means much more to a mechanical designer who mentlevel.
works most of his ideas out on a drawing
board, than to an electronics designer whose Specialized Aspects
mode of expression is via the circuit diagram Electronic techniques have made an
and for whom drawing is a rather routine impact in most branches of human acti vi ty;
operation. Until there is a common language the use of the proton resonance detector in
DESIGN RESEARCH 319
Table 34.1
Supersystems National defence radar networks; Eurovision
Systems Ship's radar; a television network
Equipments Radio and television receivers; oscilloscopes
Functional modules Decade scalers; analogue/digital converters
Circuits Binaries; cathode followers
Devices and components Transistors, triodes, resistors, capacitors
Parts and materials Silicon chips; ceramic tubes
archaeology, computers in business, and manufacture and Operation. This detail

electrocardiographs in medical diagnosis, are includes unseen 'components' such as stray
indications of their extreme flexibility. The capacitance which could have a marked in-
flexibility is to a !arge extent due to the wide fluence on performance.
range of functions that may be obtained by Thus, the designer of electronic equipment
minor changes in circuit configuration or has to make a very !arge number of small-value
values of components, and this gives rise to decisions. A typical electronic equipment
certain unusual problems in electronics design. might contain 1,000 components and devices,
Considerable attention to fine detail is re- and associated with each are several decisions
quired to ensure that the necessary function relating to circuit function, cost, quality,
is obtained under all normal conditions of reliability, safety, and so on. In a circuit, for
Charact eri st ics
if . ;jJ z !Tl
...
0 l> l>
...., .
::0
."'
(") ~ (/l ~
0 .., .., 0
<
111 ~- <
a. "0 N ~
!e. ?.
PI "0
"'
~
111 ö.., !!.
0
"0
p;-
10
;:r
- s. ·DI
-
~
3 '< :;· "0
[
Dl
- ..
111
..
Dl QJ QJ
--·
'< '< :J c ~ ~ :J
n
"'
."
.:;c
'< n
.:<
Supersystems + + .. .. + • +
.
~
Systems + • + + + • + +
Equipments • • + + + + . + + + + •
Modules + + • + + + + + +
Circuits + . + + + + + +
Devices and components + + • + .. + +
Piece parts and materials + + + + + +
Figure 34.1. Characteristics at different hierarchy Ievels

example, a resistor costing a few pence has own view of design. Attempts have been marle
to be defined in terms of electrical value, to define 'design' rigorously to take account
tolerance, power, type of construction, position of one or more of these points of view. Ross
and wiring arrangement to take account of (1964) puts forward the definition:
amplification, frequency response, d. c. condi·
tions, breakdown conditions, supply voltages 'Design is the process of conceiving,
and temperature changes. refining and recording plans, which if
Another particular feature of electronics carried out would lead to a high proba-
design is the wide range of effects and para- bility of effective accomplishment of an
meters which the designer may encounter. interrelated set of desired goals without
Frequency ranges from 10 15 to 1 are possible, the occurrence of penalties such as would
taking the limits of present-day achievements. be regarded by informed contemporary
Frequency ran ges from 10 6 or 10 7 to 1 are quite opinion in the con text as unnecessarily,
common in a single equipment - a television or substantially, offsetting the desir-
receiver, for example, incorporates SO c/s ability.'
techniques in supply and frame generator
Another approach is to formulate a definition
circuits and SO Mc/s in the receiver. Voltages
which ernborlies those aspects of design which
are quit~ commonly in the range from 10' to 1,
need particular stress at any given moment.
as is indicated by overall amplification of this
KANTOROWITZ (19S8) points out that for at
order.
least two thousand years attempts have been
The next point is that in the electronics
marle to define 'law' rigorously, with rather
field there is probably more design of complex
unsatisfactory results. It is preferable to adopt
general-purpose equipment or equipment with
a conceptual pragmatic approach, i. e. a
open-ended specifications, than in other
definition based on 'what should be' rather
fields. A general-purpose oscilloscope has its
than 'what is ', and which is fruitful for the
counterpart in a lathe, but the choices avail-
purpose of the particular exercise.
able to the designer of an oscilloscope are
very much wider than those for a lathe. The definition of the design of electronic
equipment, called a working definition, adopted
Finally, the rate of change of techniques,
by the committee was: 'Design is the process
devices and components is probably greater
of establishing relationships between all
in the electronics field than in most others.
relevant characteristics of an equipment.'
It is unlikely, for example, that such a radical
Some of the characteristics by which
change has taken place as that due to the
equipments can be described are given in
introduction of the transistor. Other techno-
Figure 34.1. At the equipment level of the
logical developments, such as the monolithic
construction of circuits, are likely to have as hierarchy, design consists of finding a set of
significant an effect. relationships between the characteristics;
For these reasons also it is desirable to just as at the circuit level, design consists of
study design processes for electronic equip- establishing relationships between, say,
ment in particular, leaving comparison with amplification, bandwidth, output voltage,
other fields to be marle at a later stage. reliability and cost. In establishing relation-
ships, account must be taken of the use to
Definitions which the equipment is to be put and of the
There are many different interpretations objectives, facilities, and restraints of the
put on the word 1design' To some it signifies organization in which, or for which, the design
'appearance', to othets it is that part of the is undertaken - this is an essential condition
evolutionary process of an equipment that for successful design.
takes place in a drawing office. Many regard
it as an activity which is by nature purposeful Conceptual Models
rather than accidental. Everybody, according In order to study the practice of design in
to his own experience or organization, has his the electronic equipment field, a picture of the
DESIGN RESEARCH 321
design activity as a whole is needed so that outside the organization. The design nucleus
the relationships between the various aspects concept is important when considering the
can be seen and understood. However, design manner in which the set of relationships is
is such a complex operation that the picture established.
needs to be a composite one, built up from a
series of models reflecting different points Materials-handling Model
of vi~w. Thus, for example, one model may be Design can be regarded as a major part of
evolved on the basis of information flow, a process which results in instructions or
another on materials processing, and another information for the conversion of materials to
on human relationships. equipment. Materials come into an organization
In dividing up the whole design field into which manufactures electronic equipment as
various study areas it is neither necessary nor componerits, devices, sheet metal, etc. The
desirable to follow the existing conventions same materials come out of the organization
or demarcations. As with definitions, the modified, adorned and combined in the form of
models may be based on a conceptual pragmatic electronic equipment. The simple model shown
approach, i.e. established on a higher level in Figure 34.2(b) identifies three areas:
of abstraction than present practice and in
groupings and terms which may be readily
understood and easily applied by those directly
concemed. The models should at first be (a)
extremely simple so that fundamental areas
of enquiry may be indicated.
The Design Nucleus Concept

The individual designer is a rarity, at
least on the equipment level. Design more often EqUipment
lb! Matenals,
takes place by teams, but even this is some- components Manufactunng
thing of a limited picture when consideration and dev1ces process
....... {nformalion ftow
is given to the manner in which a given set of -Material tloN
relationships is established between the
characteristics. Figure 34.2. (a) Information required by
A user stating the requirements for each design nucleus; (b) handling of materials
characteristic, taking into account what can
be achieved, has carried out the basic design Characteristics of the Equipment- The
of the equipment. A circuit engineer, salesman, characteristics need to be identified and
production engineer and user corporately defined. The units in which they may be
drawing up an achievable specification of all measured and their relationships should be
the characteristics have taken a major step in examined.
designing the equipment. A product policy l.faterials, Components and Devices - In
committee deciding on the emphasis to be addition to normal constructional and finishing
given to the various characteristics are materials (e.g. plating), electronic equipment
exerctsmg a design function. There is in any requires components such as resistors, capaci-
organization a section or committee or informal tors and inductors, together with devices such
association of people which plays a leading as diodes, transistors and cathode ray tubes
part in determining the set of relationships of in considerable variety and quantity. A list
characteristics. This may be called a design needs to be made of these items and their
nucleus rather than design department or team. characteristics, related to the type of elec-
The design nucleus may vary from product to tronic equipment produced. The effect of
product according to the knowledge, experience, availability and standardization on design
status and motivation of persons within and should also be determined.
Manufacturing Processes for Electronic Identification of the starting point of the

Equipment- Manufacturing processes particu- design process.
larly applicable to the electronic equipment Judgment, decision, decision trees,
field have been devised or adopted, and these iteration.
have a marked effect on design. Conducting Synthesis, optimization, reconciliation
proteelive finishes, dip soldering, printed processes.
wiring, simultaneous wiring and testing, and Random, creative and logical elements.
thin film and integrated circuit construction Thought processes particularly relevant
are among many aspects of manufacture that to design.
have to be taken into account. These processes Psychology of designer.
should be listed and described. Creativity in design.
Model. Based on the Information Required by Communications Between the Design Nucleus
the Design Nucleus and the Design Organization
If design can be regarded as a major part Communication between the design nucleus
of a process resulting in information which and the organization takes place in a variety
enables materials, components and devices to of ways and for many purposes. Forms of
be converted to equipment, what is the infor- communication are the instruction to proceed ·
mation required at the input? The model shown with the design, the design appraisal meeting,
in Figure 34.2(a) indicates four areas:
and prototype evaluation. An organization's
design procedure and many other manifestations
Materials, Component and Device Inform-
of the administration of design are of great.
ation.
importance in providing contact between
Techniques Information.
designers and their organizations. What is the
Characteristics Required by the User -
present practice?
Users can be divided into various categories,
for instance the single user, the multiple user
with identical purposes, and the multiple user Conclusions
with varied purposes. How can information be The understanding of design is at an early
collected from the various classes of user? stage, and a great deal of research needs to
What are the market requirements? What set be carried out. At present, studies should be
of emphases is given to the various characteris- in particular fields of engineering and genera-
tics in a given field of application. lized design concepts covering many disci-
Manufacturing Processes - See above. plines should be built up later by the com-
parison of results. Within one field, models
The Design Process or concepts are needed to define areas of
In the design process, the information study. These may be drawn up on a heuristic
received by the design nucleus is used and basis and tested by reference to appropriate
supplemented by, for example, an idea or by committees, conferences and discussion
deduction. What does the designer do with the meetings. Decisions to limit research to
incoming information? Topics that should be particular regions or levels in the field may
covered are: be necessary.
Chapter 35
DESIGN SCIENCE
S.A. Gregory
lntroduction ' .. .it seems not unreasonable to hope

The aim of this chapter is to promote the that the whole discipline of system engineering
concept of design science. This is introduced may serve as a paradigm for a rational theory
by a brief definition which is orientated of design. By a rather serious restriction on
towards the practical values of the subject. the range of competence that it professes by
The principal contributing sources are then concerning itself primarily with assembly-job
briefly mentioned. The characteristics of a flow systems in a formalized design situation
science are detailed, and the scope of design it has been possible to develop a family of
is reviewed in a general manner for the purpose useful synthesis techniques. Some (such as
of revealing what appear to be fundamental the feasibility study and failure design) can
concepts. The design process is then con- be taken over directly into a more general
sidered in terms of some of the possible context, others, (such as the heuristic theories)
models. The scope of research in design may need fairly radical modification, yet
science is outlined in the Appendix to this others (topological models, perturbation
chapter. methods) may have little application in fields
Design methods have been in existence outside system engineering. As system
in various fields at least from the time of the synthesis and analysis techniques, grow in
Greeks, and in modern times may readily be potential it becomes profitable to try where
traced back to Leonardo da Vinci. In recent possible to re-express other kinds of design
years, particularly in the last decade, attention problern in system terms ... '
has been devoted progressively to the study
of general methods of design, as opposed to Design science is concemed with the
the study of methods of design within parti- study, investigation and accumulation of
cular fiei.ds of technology or other relevant knowledge about the design process and its
subject. The latter, according to the maturity constituent Operations. lt aims to collect,
of the technology concerned, might have organize and improve those aspects of thought
received anything from two centuries of study, and information which are available conceming
as in the case of civil engineering, down to design and to specify and carry out research
half a century for chemical engineering, with in those areas of design which are likely to
perhaps somewhat less for electronics. be of value to practical designers and design
The development of specific approaches organizations.
of a general nature, such as the morphological
method of ZWICKY (1948) and the broad claims For these reasons, design science has to
of the several varieties of system engineering, develop its full interdisciplinary potential and
have emphasized the arrival of a well-defined take what is available, not only from the
and important branch of a field of study, itself fields of systematic design and system
ready for treatment by the accepted methods of engineering, but also from management science
science. It is ptobably this which caused and those aspects of the behavioural sciences
GosLING (1963) to write: which have the possibility of throwing light
12 323
on design, designers, design organizations, to a fragmentation of human experience and

and the social implications of design. potentiality, although accompanied by a
deepened knowledge. Today the need is for
Can there be a Science of Design? the development of sciences covering the key
First, it is necessary to recapitulate the interdisciplinary themes, of which design is
characteristics by which a science may be but one.
recognized. There is no absolute criterion: Design science is concerned with the
practical scientists within a given area tend study of design as shown by the evidence of
to agree at a particular time upon the features material objects, in the behaviour of indi-
which distinguish the scientific from the non- viduals engaged in design, and in the behaviour
scientific; practice sees li ttle of the cri teria of design groups. It is concerned with the
proposed by POPPER (1959, 1963) which are effects and fate of the end-products of design.
primarily of a prophylactic nature. Design science is concerned to treat these
According to the maturity of a science, phenomena in a manner as fully in accord as
a gradation of behaviour reckoned as scien- possible with the normal understanding of
tific may be expected. This may range through: scientific method, and to make available the
resulting knowledge and theories for the better
(1) Description of phenomena (natural accomplishment of practical design.
history phase).
(2) Categorization in terms of apparently What is Design? The Product and its Behaviour
significant concepts. The material and inanimate evidence for
(3) Ordered categorization whose pattern the study of design, available for all to see,
may be deemed a model (the evolutionary lies in the accumulation of artefacts, an
taxonomy or periodic table phase). inclusive category for the totality of products
( 4) Isolation and test of phenomena, with of human art and workmanship.
implied reproducibility by independent Within technology, groups of operations
observers (foundation of 'research'). may be seen which embrace substantially the
(5) Quantification (classical physics phase). material and physical side of economic
endeavour:
Accompanying these will be the progressive
development of models of all types, initiated (1) Materials- their mining, extraction,
by speculation, conditioned by the prevailing preparation.
conceptual climate, and valued in terms of (2) Energy - its preparation, distribution,
their abili ty to explain and then predict utilization.
phenomena. Quantitative prediction provides a (3) Shaped products - capital equipment,
major influence in the reduction of uncertainty consumer durables, consumer goods, etc.
concerning the validity of a model. This is (4) TransportaUon - land, water, air,
reinforced by the number of occasions of space, etc.
success. (5) Information transfer - printing, tele-
As far as desi gn is concerned, the occur- communication, computers, replacement of
rence of all these kinds of reputedly scientific man, etc.
behaviour may be expected. This is so because
the area of study is interdisciplinary in the A brief picture of the way in which these great
present academic sense. Up to this time, branches of industry are linked tagether to
sciences have largely developed around topics give man power over nature is presented in
of a restricted character, definable in terms of Figure 35.1. This depicts the processes which
some material level in the physical world. eventually deliver material, energy, information,
The early study of medical science is, however, or some combination of these, to the consumer
an example to the contrary. In general the for his satisfaction. The processes required
development of the classical sciences has led are designed, or have come tagether in the
DESIGN RESEARCH 325
T~cnnology
(t~chn•cal and Sc~Pnce and
manag•mf'nt gpneral engineenng
r•sponsibil•t••s)
{
Petrole-um Meta I
rf'fmtng, ~melt•ng,
/
Chemtcal c.hPmtcal
gas•f~eatton,
engtn~rtng manufacturt>,
coktng polymer
manufacturp
E1Ktrtca1 Syst.,ms
engn"''e•rmg eng1neoering
Ptoductton
engineoerino {
Figure 35.1. Same linkages between the great branches of industry, the technologies
and the sciences
course of time, to form systems. They operate of the living process. Apart from providing a
through equipment, now commonly termed suitable shape for the process, means have to
hardware, which is the material embodiment be provided to protect the process from external
of the design. The expression of the process in terference, particularly from the effects of
design is through a hardware design. weather. In chemical engineering a process
The processes themselves require an is designed which is carried through in inter-
abstract design and the formal statement of connected pieces of plant of suitable shapes.
these comprises software. Although the obvious These plant items are usually constructed in
evidence for the work of technology is the such a way as to prevent their contents from
accumulation of material artefacts, accompanied affecting the surroundings, and to avoid the
by the delivery of satisfactions, the basic influence of the environment upon the process.
and most important artefacts are abstract. In telecommunication the process takes place
As far as processes are concerned, the inside suitable components, and often in
material embodiments may be viewed as selected parts of the environment, avoiding
vessels or shapes within which the processes where possible uncontrolled leakage.
take place. In architecture the concem is with From these examples, material embodiment
the construction of vessels for the operation may be seen to incorporate at least two
functions: the facilitation of the process by a with subjecti ve values, one may see the
particular shape; and the protection of the objecti ve material functions, perhaps rather
process from interference or loss. simply, in the following ways:
That some distinction needs to be made
between the desi gn of the process proper and (1) Foodstuffs - stores of materials and
the design of the embodiment, and between energy.
the process function of the embodiment and (2) Clothing - containers for life-process
its support or Containment function, is not units.
essentially new, although probably unfamiliar. (3) Furniture - Containers for stores such
There is part of such a distinction to be found as food, books, etc., or containers for life-
in an article on the steam engine, published in process units for specific duties.
1853 by TOMLINSON: ( 4) Semi-durables - e.g. r e f ri gerat o r s,
television sets, which comprise subsidiary
'For abou t 60 years, from 1710 to 1770, the systems needed for the life-process.
engine remained almost in the state to which (5) Durables - e.g. houses, which con-
Newcomen had brought it, although occupying stitute and provide space for the life-process,
for the latter part of that time the attention of and which contain subsidiary systems such as
Smeaton, indisputably the greatest statical water-supply, heating, etc. These may-' be
architect of modern times. It is truly astanish- directly in support of the main process or may
ing that a designer so unrivalled in the
serve to prevent interference from outside, etc.
mechanism of all fixed structures, fulfilling
in them the newest and boldest requirements in Material embodiments are, by their nature,
the very simplest ways (and moreover so composed of materials of construction. These
ingenious and conclusive observer and experi- need manufacturing specifications in order
menter on the mechanical powers, including that they may be correctly shaped. Static
steam itself,) should be employed on this embodiments involve shape and magnitude; for
engine. . . without seeing the great improve- dynamic embodiments motion must be added.
ments for which it was now fully ready ... It Materials may also be used as feedstocks for
seems, however, that even between Statics chemical transformation, or as sources of
and Dynamics, which are regarded as divi- energy. Materials may be held between stages
sions of the same science, or at least between of transformation, whether in shaping, chemical
the inventive applications of each, Architecture processes, or other system Operation. This
(or Engineering as it has been called since constitutes a store stage. For such a process
Smeaton's time) and Machinery, the connection an input and an output have to be specified;
is not close enough for the same persons to the condition of functioning is that no change
excel in both ... ' shall occur. Storage is a zero-process but not
Given such a separation in thought, a non-process.
hardware design can be seen as broadly con- For containers and supports, the essential
cerned either with systems involving transport feature is non-process. In general, at the
or transformation phenomena, or with structures interface between the process and the environ-
or groups of structures concerned with non- ment, there must be no process transaction,
transport or non-transformation. The material but because of the limitations of many materials
artefacts involved in these may be built up of construction this Iimitation may have to be
from components which themselves may be con verted into an acceptance of an economi-
systems or static units. cally small leakage or loss or transformation.
Since an important part of design is con- It is a general characteristic of artefacts
cerned with the delivery of satisfaction to the that the nearer they come to the consumer or
consumer, it is worthwhile to look more closely user (this allows for machine-tools, etc.) the
at specifically consumer products. Bearing in more consideration needs to be given to the
mind that, since they come into close contact interaction at the interface of the artefact or
with the consumer, these must be impregnated system with the human beings concerned.
DESIGN RESEARCH 327
Varieties of Design in a specific case has been reported by

Students of technological design already GREGORY (1964b). This refers to process
recognize different versions of design activity design, process shape, container design and
which may have a hierarchical relationship materials selection, and also invol ves an
within the design-space of a system. For aspect of production method design.
example, it may be a system or a component that
is designed; it may be a process or a container. Design ltself a Process - Possible Models of
Not all major practical design activity is Designer Behaviour
necessarily overtly concerned with system Much of what has been considered so far,
design. Systems have the advantage that the may now be exploited in the consideration of
design problern may be formulated in terms of the design process itself. Design is normally
an input-output equation: taken for granted as a process and its
operations described in the form of a block
Output = f (!) diagram. It is also readily assumed that design
is a process which has an input and an output.
The transfer functions symbolized by f con- Although many models have been suggested
stitute an important part of the study of the for the design process it would seem likely,
given technology (in so far as it is based upon in the light of the earlier discussion, that a
transformations) and their determination is the system model is perhaps the least sophisticated
foundation of a design of the type concerned. model with a possibility of giving useful
Such a system may consist of a single line; results. Any simpler model is likely only to
or of several systems in parallel where one have literary interest. A simple example of the
may be dominant; or there may be interactions latter would be to describe design as a crystal-
between parts of one system, or interactions lization.
between systems in parallel. Clearly, where According to the system model, concern is
interaction occurs, the design becomes more mainly with the specification of the transfer
difficult. functions. If design is seen as an information
For the case of non-process design, the transforming process, then the individual
input-output approach cannot be immediately information transformations must be specified.
used. In the simplest cases concerning retention As far as is known, by direct inspection or by
of shape or position, the essence of design psychological investigation, the number of
may be expressed by an inequality in which classes of transformations is small. Scientific
the force tending to change must not overcome exploration is therefore a possibility. Design
the resistance. If the required materials which is an unsteady-state process (see Chapter 31)
give the condition of stability cannot be found although occurring within almost constant
then some version of the input-output equation designer facilities.
may have to be used. The design would nor- Together with these functions, concern
mally involve a feasibility study and an must also be with storage and with the pre-
optimizing study, with a possible interaction. vention of loss or interference during the
Design for shape may be dominated by process: with the conversion of abstract know-
process or stability considerations under ledge into embodied knowledge.
economic constraints, or by direct geometrical In the light of information theory it should
constraints, or by human requirements. Some be possible to expand the details of this model
aspects of the geometry involved in design substaritially. This remains as a task to be
are extremely complex and design techniques clone. In this task, the designer is regarded
or strategies are ripe for mathematical investi- as in the same class of machines as the digital
gation, probably exploiting the computer computer and, given this similarity, there
advantageously. should be no reason why a computer should not
A detailed study of some of the inter- carry through a design of the type which this
actions between successive stages of design model represents. For a closer approach to the
practice of design, models with greater sophis- 12, looks at the creati ve search for infor-
tication are needed. Watts, in Chapter 11, mation. SHACKLE (1961, and earlier) empha-
provided a model of an autonomic information sizes the creativity of decision.
process. The ambition to discover the mechanisms
whereby these operations take place will be
powerful. Yet the type of human behaviour
Design as a Psychological Process discussed is likely to be well em bedded in the
The study ofthe behaviour of the individual 'black box' class, and to resist detailed
designer shows a number of characteristics analysis. What is likely to happen is that
not revealed in the simple information process some more techniques of a practical nature
model proposed. The human designer is a very will be discovered and finally it will become
complex system with a number of lines Opera- possible to set up some simple models by which
ting in parallel, and possessing enough parts of the operation may be carried out. The
inherent flexibility to accommodate a variety kind of investigation relevant to a system will
of design programmes and to invent new ones. be performed in which an input with suitable
The designer provides something analogous variations is provided and the effect on the
to the function of a pump in a material flow output noted. Thence it will be possible to
process: he has 'head' and 'capacity'. He argue to the nature of the transformation
displays that ability to provide new solutions devices. Since the input and the output will
which is called creativity. He takes decisions consist of information of various kinds, in any
on values, and under uncertainty both in of the possible communication languages, it
respect of information potentiality available will perhaps be assumed (as in Gregory 1964b)
now and in the future. The designer is prepared that comparable transformations, whatever the
to gamble on the unknown and on his own language of communication, are produced by
ability. He is vulnerable to his environment. equivalent creativity functions. In order to
Attitudes are important. obtain such a result, the use of a suitable
To discuss just one of these differences, black box may be considered, such as that
namely creativity, brings out the greater com- described by ASHBY (1956), fitted with trans-
plexity of the model and, at the same time, ducers to convert the input communication into
suggests some of its values. It is possible to common black box language, and to convert
see the phenomenon of creativity as an inter- the output suitably. Ashby comments on the
action between the life-process and the design process.
mechanical information transformation process. This model has many interesting impli-
It is this interaction which provides the possi- cations and, like the mechanical information
bility of a creative transformation at every transformation process, still requires elaboration
stage of the design process, and not just at and detailed study.
the stage commonly called synthesis. If the
stages of design are set out in the form of a
block diagram as a first approach to the speci- Design as a Sociological Process
fication of the transformations, then creativity The mechanical informaHon model and the
may occur at every stage. A new perception of psychological model have been looked at, and
need or a new appreciation of value might be some of the possible points for further research,
obtained (GREGORY,1965). based on questions stimulated by these models,
This aspect of the model provides further have been noted, but attention has been re-
opportunities for research. Already some stricted to the individual designer. A model is
people have developed practical approaches needed which would enable a useful investi-
which are exploiting some of the possible gation of the execution of a design process
areas of specific function creativity. For within a working group to be made. Indeed, it is
example, the 'Synectics' group has worked on possible to look forward to composite or hybrid
the task of stating a problern in the most fruitful systems involv~ng a number of designers and
way (GORDON, 1961). Farradane, in Chapter one or more computers, with othet facilities.
DESIGN RESEARCH 329
Such a system would need a model representing Research in Design Science

a micro-sociological process, as suggested by Some areas for research have already been
GREGORY and BURDIS (1965). indicated and these are developed in the
It would be reasonable to expect that the Appendix. These tend to be concerned with
original model of the mechanical information large, or general, or abstract topics. Principal
transformation system could be subsumed in stress has been placed upon the nature of the
some form within the psychological model and design process and upon the development of
that this, in turn, might be subsumed within the broad concepts which might assist the analysis
micro-sociological model. Some of the first of design procedures. But consideration must
implications are being explored elsewhere, also be given to research which promises to be
in terms of creativity and of decisions. more immediately productive. This is likely to
deal with the techniques of present design,
Design Situation mechanical and theoretical, and with the day-
The design si tuation incorporates the to-day strategies practised in most areas of
opportunities for and the constraints upon the technological design. Any organized design
practice of design. The opportunities are those research must achieve some balance between
for the provision of satisfaction to the consumer the immediately useful and the fundamental,
and user. They are also the opportunities for each fertilizing the other.
the satisfaction of the designer organization.
This may be in terms of the contemporary
economic values, or in terms of challenge, APPENDIX: Outline of Suggested Research
or according to the lapped hierarchy of need- Fields in Design Science
fulfilment of JlvtASLOW (1954). The outiine is tntended to be concerned
Unless the design situation is studied with all major aspects of design and its con-
scientifically and the results acted upon, all stituent operations, in theory and in practice.
efforts at the study of design in products, or First, possible topics are listed in terms of
of the process of design, are likely to be the design process; then follows a suggested
sterile. A systematic survey of opportunities allocation of such problems among existing
for design in growth areas needs to be de- academic disciplines.
veloped - generalized marketing for design -
taking due regard of such pointers as the Design - The Product, Process and Situation
National Plan, and of critical survival require- (1) The Design Process- Adequate and
ments. The necessary profile of design capa- plausible accounts of design under all con-
bilities should be specified, and the changes ditions .. Comparison of various kinds of design.
needed with time should be reviewed. One Generation and exploration of concepts charac-
should therefore look more closely at the terizing design. Generation of models. Evalu-
topics touched upon in Chapters 4-9, and ation of models.
repeatedly ask how existing products might be (2) Establißhment of Design Need- Market
improved to meet the needs of the situation, research. Horne and export market characteris-
or whether radically new approaches should tics. Short-term and long-range forecasting.
be used. Comparison of mass-production and one-off
Constraints will always be present. Our outlets, and the designs for them. The possible
effort must be turned to dissolving, eroding, applications of innovation research to
or overturning unnecessary constraints. These marketing.
are largely in the mind, as opposed to natural (3) Values and Design - Social effects
constraints, and come either from idleness or of new designs. The individual and new design.
misplaced enthusiasm for outworn doctrines. Methods of quantifying values.
The scientific analysis of innovation problems, ( 4) Methods of Approaching Design
along the lines suggested by ROGERS (1962), Problems - Collection of procedures and
is our hope, provided that it is followed by rational analysis thereof. Development of new
action. approaches. Comparison of methods used in
various technologies and their branches. ing general management techniques. Measures
Establishment of techniques of approach of design output and efficiency. Passihle new
appraisal. Practical comparison of alternative methods of tncreasing productivity. Effects of
techniques on selected problems. the design environment. Methods of stimulating
(5) lnformation-handling for Design - The creativity.
nature of design information. Retrieval methods.
Methods of generating information. The value Possible Allocation among Disci]}lines
of information and its economics. Mechanical (1) Philosophy - Study of the key con-
methods of information transformation. cepts of design. Problems of value. The nature
(6) The Production ol ldeas- Collection of optimality, novelty, uncertainty. lnvesti-
of existing heuristic techniques. Rational gation of areas reputedly deficient in con-
analysis. Development of new heuristics. cepts. Conceptual change.
Application of computer methods. Creati vity. (2) Mathematics and Computer Science-
Analysis of creati ve behaviour. Generation of Optimization techniques. Sturlies of spatlal
new creativity techniques. arrangement. Studies of heuristic rlevices and
(7) The C&mmunication of ldeas- In- heuristic programming. Development of COm-
formation theory and i ts applications in design. puter methorls of executing tasks now under-
The channels of perception. Symbols, models taken manually. Communication investigations.
and their varieties. The specification as (3) Physics and Electronics - Optical
communication. Mechanical methods of repro- methorls of reproduction. Electronic methorls of
duction and transmission. Optical methods. reproduction. Hybrid methods of reproduction.
Electronic methods. Economics of alternative Optical anrl other methods of shape manipu-
methods of reproduction and communication. lation.
Ext>loration of new techniques. ( 4) Materials Science- See earlier hearling
(8) The Testing of New ldeas- Abstract Materials anrl Design.
models. Drawings. Concrete models. Analogues. (5) Engineering Technologies - Develop-
Pilot plants. Prototypes. Levels of certainty. ment of specific design techniques. Develop-
(9) Decision-making and Decision ment of general system rlesign techniques.
Strategies - Adequate accounts of decision- (6) Economics - Marketing and design.
making in design under all conditions. Gen- The economic values of design. The economics
eration of suitable models. Psychological tests. of information in design. Economic techniques
Passihle selection procedures. in design.
(10) Materials and Design - Relationship (7) Psychology- Psyc'hology of the
of materials to design, including transformation designer. lnvotvement and motivation. Attitude
properties, protection and support, prorluction anrl design. Creati vity. Selection procedures
implications. 1Vlaterials selection procerlures, for designer choice. Psychology of the user.
their collection and rational analysis. Develop- Ergonomics. Psychology of the purchaser.
ment of computer methorls of materials selec- Problems of innovation. Behaviour in un-
tion. Exploration of material possibilities and certainty.
probable effects on design. Sturly of major (8) Sociology - Behaviour of design
materials limitations in existing rlesign. ReH- groups. Innovation problems of rlesign. Sturly
ability and its relationship with material of social rlemands for rlesign.
properties. (9) Management Science - Exploration of
(11) Man-Machine lnteroction - Ergonomics specific management techniques as applierl to
anrl design. Safety. Aesthetics. design activity. Sturlies of companies anrl their
(12) General System Design Problems. rlesign procerlures in relationship to successful
(13) Management and Design - Methods of company operation. Sturlies of design prorluct-
improving design prorluctivity. Effect of exist- ivity.
GLOSSAR V
Algorithm. A routine or device for calculating. General systems theory. A theory developed
The algorithmic approach becomes more over the last decade in an attempt to com-
general in methodical procedures for design. prehend the significant aspects of all
The present-day significance of the algorithm classes of material objects having structure.
lies in its potential use in a computer General systems theory is essentially con-
programme. cemed with Ievels or hierarchies of behaviour
Artefact (artifact). A product of human art and which are related in some way to the com-
workmanship; the material result of de- plexity and machinery of the objects con-
signing; anything which has been designed cemed. For example, ·fundamental particles
and manufactured. come together in atomic nuclei. Upon these,
Autonomie. Having the right of self-government, suitable development of electron shells
of making decisions. A designer who is not provides the chemical basis of atomic
autonomic to some degree is not a designer. behaviour. Atoms combine to form molecules.
Concept. A general notion. In normal parlance Molecules of sufficient complexity and
this conveys the idea of a class of obj"ects, adequate chemical properlies come together
whereas for a designer it implies the first to form living cells. For the designer there
outline of an artefact, whether in the mind is a hierarchical series of classes of
or as a sketch, which needs further details artefact, starting with a static mechanical
for a satisfactory specification. piece. Such pieces may be formed into a
Constraint. A limitation, usually of a com- static structure. A structure may be caused
pulsory nature. In facing constraints a to deflect or vibrate. Other static pieces may
designer is limited or confined in respect of be formed into a structure capable of trans-
choice. Some constraints are natural, e.g. the mitting motion by rotary or other movement.
limit in strength of a material; others are The addition of other pieces makes it
social; still others spring from the individual. possible to control the speed or motion of
Creativity. The ability to respond in a novel the structure. Structures may be developed
and useful way to a problem. Creativity can to convey or transform electrical impulses,
only be seen after the event. Tests of or the structures may be used to house
creati vity are concemed with disclosing chemical reactions. By the suitable
what is believed to be potential creativity accretion of devices, complex artefacts
in an individual regardless of circumstance. may be constructed which will contain
Decision. A choice between alternatives which living material, or which will carry out
settles a course of action. sophisticated logical Operations, or simulate
Ergonomics. The study of the relation between self adaptive behaviour. The aim of general
man and his occupation in terms of anatomy, systems theory is to characterize the prin-
physiology and psychology, and their inter- cipal pattems of systems structure and to
action with the characteristics of working provide an account of the underlying logic.
equipment and environment. Graph theory. See Network.
Function design. The central part of design Heuristic. Having to do with finding; related
which is followed (if it is not a mechanical to improving problem-solving performance;
function) by mechanical and production relying upon problem-solving devices which
design. appear to be particularly relevant to the
331
circumstances and, if need be, devised for be represented by a network of lines and
the occasion; a mode of procedure held to points drawn on paper or constructed in
be nfi!~essary in the absence of satisfactory three dimensions. Such a network may be
algorithms. Heuristic procedures move from random or ordered, may be richly connected,
the use of known methods to novel methods. or may divide into successive branches
Creati vity techniques are attempts to like a tree.
stimulate heuristic behaviour. Such a network may represent material or
lconic. Of the nature of a representation. abstract relationships. A static network may
Innovation. Making changes or bringing in represent relationships in a general manner
novelties. In the sociolo gical sense inno- or imply some sequence in value or time.
vation concems the processes of communi- Also, the lines may imply some specific
cation and adoption of products or tech- orientations in space and, in addition to
niques. direction, may indicate length and other
Methodical procedure. 0 r der 1 y or regular factors such as capacity. Flow of some
arrangement to proeure a result. An algorithm kind may occur and such flow may be steady
is a methodical procedure for calculation. or unsteady. Portions of the network may
In design a methodical procedure normally move relative to each other. Their inter-
conveys the notion of a prograrnme which connections may remain the same or change.
may be written down and subsequently used The study of structures of this type, the
by someone unskilled. Most design in- study of relationships, is clearly a general
volves some further element and the methodi- way of looking at some of the fundamentals
cal procedure may be seen as a device for of systems.
making skill more available to the difficult Objective. The point or goal towards which
areas of design. This kind of procedure may operations are directed. Policy is the set of
involve a sequence of strategies. principles underlying a course of action,
Methodology. A study or extended development declared or implicit, determined to achieve
of method in some sphere of activity. an objective.
Model. A representation of some past, present, Opportunity. Opening for the exercise of enter-
or future obj ect or event, used for communi- prise in any kind of activity; with reference
cation or for dealing with problems. There to design, changes in the market, technical
may be a considerable degree of trans- skill, or produclive capacity, provide
formation between the representation and openings.
that which it represents. Optimization. Improvement in performance,
Morphological. Having to do with the study of preferably to the useful limit, as gauged by
form. In the field of design the word refers some agreed relevant standard of value.
particularly to the approach of Zwicky Paradigm. Example or pattem; a normative
(1948) to analysis and construction. If some model.
well-defined characteristic is involved, a Precedent. Previous design taken as example
schematic representation of the totality of for subsequent work. A successful precedent
possible occurrences of the characteristic carries a level of justification.
is arranged in a 'morphological' box or Product. That which is produced either in the
manifold. Completeness of the manifold is course of nature or by manufacture. Today it
achieved if no compartment contains more is normally used to represent something
than one occurrence. Some compartrnents designed, made, and offered to a consumer.
may be empty. The object in each compart- Within design there are also special mean-
ment is evaluated in the light of the original ings.
requirement. The morphological method is Realization. Conversion of a plan into fact.
primarily a search procedure. The realization of a design involves the
N etwork. An interconnecting set of points or completion of mechanical design, its detail-
items in two or three dimensions (n dimen- ing or specification in terms of fabrication
sions are not ruled out). This network may in some selected material, the production of
GLOSSARY 333
these items, their assembly, test, and and ends at successive stages in the
commissioning, and the overall organization operation.
of these operations. System. A word having a number of meanings
Restraint. See Constraint. relevant to design. The first dictionacy
Satisfaction. The fulfilment of a want or need definition usually suggests: complex whole,
by sufficient supply. or set of connected parts or things; organi-
Set theory. A species of logic dealing with the zation of material or immaterial things. It is
relationship between discrete objects or this meaning which is stressed since it has
points in terms of the groupings to which a double relevance to design. Design itself
they may belong. The theory is considered is concerned with the conception, production,
fundamental to modern mathematics. and arrangement of parts or things in some
Situation. The total circumstances in which a relationship having value. Furthermore,
design is prepared and a product made, some members of the class of systems, as
marketed and used. The designer attempts seen in general systems theory, have con-
to bring about a match between the needs of siderable engineering and economic signifi-
the situation and the characteristics of the cance. The behaviour and mode of design
product The function of marketing is to of such systems have been studied in some
gauge the opportunity presented by the detail.
situation. System design and engineering. The design of
Strategy. A pattem of action aimed to achieve systems of the type mentioned above. Such
an objective. Design strategies are usually systems are concemed with the transfor-
concerned with making the most effective mation or transport of mechanical energy,
use of design ßkill in complex situations. electrical energy, chemical materials, or
A strategy has to be directed generally information. In many practical systems
towards the objective and must lie within there may be considerable interaction with
the policy bounds. Because of unknown humans.
territocy ahead, a strategy may not be suc- Systematic procedure. See Methodical procedure.
cessful. A strategy may contain a suc-
cession of sub-strategies developed by For further definitions, see also Chapters 3
study of the relationship between means and 32.
REFERENCES AND BIBLIOGRAPHY
A.I.CH.E.-I.CHEM.E. Symposium Series (1966) Application BAXTER, P.A. {1961) 'Work study in the pre-production
of Mathematical Models in Chemical Engineering field' in Proc, Productivity Techniques Conf., A.B.C.M.,
Research, Design and Procluction, No.4, Institution London
of Chemical Engineers, London BEAMENT, 1.W.L. (1960) Models lllld Analogues in Biology,
ACKOFF, R.L. (1961) Progress in Operations Research, Cambridge University Press, London
Vol. 1, Wiley, New York BECK, II. V. (1 %1) 'Design processes for electronic e!pip-
ADELSON, RJ.!. (1965) 'Criteria for capital investment: an ment' (abstract of Chairman's address), Proc. lnstn
approach through decision theory', Opl Res.Q. 16, 19-50 elect. F:ngrs 108, ß38
ADMIRALTY (1959) Code of Practice for Designers of RECK, II.V. (1%4) 'F.lectronics design', E/ectronics &
Naval (Seabome) F.:lectronic Equipment, BR6002, Power, Oecember, 443-5
Weapons Dept. BELLMAN, R. and RROCK, P. (1960) 'On the concepts of
ALEXANDER, C.W.1. (1964) Notes on the Synthesis of a problem and problem-so! ving', Am.math.Moo. 67, 119-33
Form, Oxford University Press, London BELLO, F. (1959) 'The magic that made polaroid', Fortune
ALEXAN DER, T. (1965) 'Inventing by the madness method', 59, 124-39
Fortune 65, 165-8; 190; 1Q3-4 BEY ER, R. (1963) Kinematic Synthesis of Mechanisms,
ALEXANDER, W.O. (1965) 'Rivalries between metals and McGraw-llill, New York
with other materials' in Metallurgical Achievements BINSTED, D.S. (1960) 'Work study in project planning of
(ed. Alexander, W.O.), Pergamon, Oxford a chemical process, Part 2', Chemy [nd. 16th 1anuary,
ALGER, 1.R.M. and HAYS, C. V. (1964) Creative Synthesis 59-65
in Design,c Prentice-Hall, New York BIRCHALL, H. (1960) 'Work study in project planning of a
ALLPORT, G.W., VERNON, P.E. and LINDZEY, G. (1931) chemical process, Part 1 ', Chemy lnd. 16th 1anuary, 56-59
A Study of Values, Houghton Mifflin, Boston, Mass. BOOKER, P.1. (1962) Principles and Precedents in
ARCHER, L.B. (1965) Systematic Method for Designers, Engineering Design, Institution of Engineering
Council of lndustrial Design, London Designers, London
ARISTOTLE, (variottS editions) Nicomachean Ethics BoOKER, P.1. (1963) A History of Engineering Drawing,
AsHBY, W.R. (1956) An lntroduction to Cybemetics, Chatto &. Windus, London
Chapman & Hall, London BooKER, P.1. (1964) Conf. Teaching of Engineering
ASIMOW, M. (1962) lntroduction to Design, Prentice-Hall, Design, Institution of Engineering Designers, London
Booz, ALLEN and HAMILTON, Management of New
New York
AYER, A.1. (1936) Language, Truth and Logic, Gollancz, Products, 135 South LaSalle Street, Chicago 3
BosTON, D.C. (1963) Adaptive Proßrammin/1 for Design
London
Applications (M.A. Thesis), University of Bristol
BOUCHER, D.F. and AL VES, G. E. (1959) 'Dimensionless
BAKER, R.1.S. (1962) The Management of Capital Projects, numbers', Chem.Engng Prog. 55,55-64
Bell, London BOUCHER, D.F. and ALVES, G.E. (1963) 'Dimensionless
BAKER, S.S. (1962) Your Key to Creative Thinking, Harper numbers, 2', Chem.Engng Prog. 59, 75-83
and Row, New York BOULDIN G, K. E. (1956) 'General systems theory - the
BARR, A.H. (1933) Picasso:· Filty Years of his Art, New skeleton of science', Mgmt Sei. 1, 197-208
York Museum of Modern Art BOULDING, K.E. (1964) 'Foundations for a general
BARTLETT, Sir F. (1961) Remembering, Cambridge Univer- systems theory' in Views on General Systems Theory
·sity Press, London (ed. Mesarovic, M.D.), Wiley, New York
BARTLEY, S.ll. (1958) Principles of Perception, Harper, BRAITHWAITE, R.B. (1953) Theory of Games as a Tool
New York for the Moral Philosopher, Cambridge University Press,
BARTLEY, W.W. (1962) Retreat to Commitment, Knopf, London
New York BRATT, E.C. (1958) Business Forecasting, McGraw-Hill,
SATTELLE MEMORIAL INSTITUTE (1965) Mechanical New York
Reliability Concepts. See also: BRECH, E.F.L. (1957) Organisation: The Framework of
Cress and Cheaney (1963) Management, Longmans Green, London
McCrory (1963) BRECH, R. (1963) Britain 1984, Darton, Longman & Todd,
McCrory, Wilkinson and Frink ( 1963) London
Mitchell and Creswick {1964) BRECH, R. (1964) Planning Prosperity, Oarton, Longman
Trainer and Glasgow (1965) & Todd, London
335
BRIGHT, J.R. (1962) Technolo~ical Plannin~ on the CONCEICAO,P.H. and PARTON, K.C.(l963)TheEconomic
Corporate Level, llarvard Graduate School of Business and Practical Application of a Di!1ital Computer in
Administration, Cambridge, l'/lass. Electric Machine Desi~n Offices, I.E.E.E. Paper No.
BRIGHT, J .R. (1964) Research, Development and Techno- CP 63-266
lo~icallnnovation, lrwin, Homewood,lll. COOKLIN, L. (1964) 'New product development',Federation
BRILLOUIN, L. (1%2) Science and Information Theory, of British lnclustries Review, August- December
Academic Press, London COONS, S.A. (1963) 'An outline of the requirements for a
BRITISH STANDARD 436 (1940) Machine-cut Gears: computer-aided design system', AFIPS Conf. Proc. 23,
Helical and Strai!1ht Spur, British Standards Institution, (Spring Joint Computer Conf.), Cleaver-Hume, London
London COOPER, D. (1956) Georges Braque: The Evolution of a
BRITISH STANDARD 545 (1949) Bevel Gears, Machine- Vision, Arts Council of Great Britain, London
cut, British Standards Institution, London COOPER, W.W., LEAVITT, H.J. and SHELLY, M.W .. (1964)
BRODBECK, M. (1960) 'Models, meanings and theories' New Perspectives in Organization Research, Wiley,
in Dccisions, Values and Groups (ed. Willner, D.), New York
Pergamon, Oxford CORNFORD, E.C. (1964) 'Formulation of the requirement'
BROGDEN, H.E. and SPRECHER, T.B. (1%4) 'Criteria of in Management of Large Technological Projects,
creativity' in Creativity: Pro~ress and Potential 6th Symp. - Key Problems in Aeronautics
(ed. Taylor, C. W.), McGraw-Hill, New York Cox, Sir H.R. (1964) The Shape ol Thin~s that Work,
BROSS, l.l).j. (1953) Desi~n for Decision, Macmillan, Third Annual Lanchester College Lecture
New York CRAWFORD, R.P. (1954) The Techniques of Creative
ßROWN, R.H. (1962) 'On the selection of economical Thinking, Hawthorn Books, New York
machining rates', lnt.]. Prod. Res. I, 1-22 CRESS, H.A. and CHEANEY, E.S. (1963) Determinin~
BROWN, R.R. (1963) Explanation in Social Science, Design Feasibility, A.S.M.E. Paper No. 63-MD-38
Routledge & Kegan Paul, London CURRIE, R.M. (1960) Work Study, Pitman, London
BROWN, W. (1953) Epicyclic Gearin~, Yellow Book No.33,
Machinery Publishing Co., London DAVIES, Lord Llewelyn, (1965) 'Disciplines before data',
BucK, C.ll. (1963) Problems of Product Development and Desi~, February, 45-8
Desi~n, Pergamon, Oxford DAVlES, O.L. (1%2) 'Some statistical considerations of
BUNGE, M. (1%3) Myth of Simplicity, Prentice-Hall, New research projects in the pharmaceutical industry',
York 4ppl. Statist. II, 170-83
BURNS, T. and STALKER, G.M. (1961) The Mana~ement DAVlS, M. (1958) Computability and Unsolvability,
of Innovation, Tavistock, London McGraw-llill, New York
BYRNE, A. and VAN KOOTEN, L.C. (1965) 'Controlling DAVIS, R. (1963) 'What will happen to your good ideas?',
process by remote computer', New Scientist 28, 866-7 Engineering, 27th September, 388-9
DAVIS, R. (1964) 'The function of aeativity', Chart. Engr
II, 368-73; 386
CANGUlLHEM, G. (1963) 'The role of analogues and DE VRIES, M.G. (1964) 'The dynamic effects of planning
models in biological disoovery' in Scientific Chan~e horizons on the selection of optimal product strategies',
(ed. Crombie, A.C.), lleinemann, London Mgmt Sei. 10, 524-44 ·
CARRITT, E.F. (1949) An lntrocluction to Aesthetics, DESCARTES, R. 'Rules for the direction of the mind' in
Hutchinson, London The Philosophical Works by Descartes (translated by
CARTER, C.F. and WILLIAMS, B.R. (1959) Science in Haldane, E.S. and Ross, G.R. T., 1911-1912), Cambridge
lnclustry, Oxford University Press, London University Press; London
CHASE MANHATTAN BANK (1964) The European Markets: DEWEY, J. (1910) How We Think, Heath, Boston
a Guide for Businessmen DRUCKER, P. (1%4) Mooaging for Results, Heinemann,
CHOMSKY, A.N. (1959) Review of B.F. Skinner's 'Verbal London
behaviour', Lan~uage 35, 26-58 DUCKWORTH, J.C. (1965) 'lncentives to innovation and
invention' (Eleventh Graham Clark Lecture), Chart.
CHURCHMAN, C. W. (1961) Prediction and Optimal Decision,
mech. Engr 12, 201-5
Prentice-Hall, New York
DUHEM, P.M. (1906) La Theorie Physique: Son Objet, Sa
CHURCHMAN, C.W. and RATOOSH, P. (1959) Measure-
Structure, Chevalier et Riviere, Paris
ment: Definition and Theories, Wiley, New York
CHURCHMAN, C.W. and RATOOSH, P. (1%1) 'Innovation ECKMAN, D.P. (1961) Systems: Research ood Design,
in group behaviour' in Proc.2nd lnt. Conf. Oparational Wiley, New York
Research, English Universities Press, London EDER, W.E. (1964) in Conl. Teachin~ of Engineering
CHURCHMAN, C.W. and SCHAINBLATT, A.ll. (1965) 'The Design (ed. Booker, P.J.), Institution of Engineering
researcher and the manager: a dialectic of implemen- Designers, London
tation', M~mt Sei. II, B69-B87 EDER, W.E. (1964) 'Fatigue design, Parts 1 and 2',
CLAUSEN.' H. (1958) En~ineerin~ Desi~n: the Back~round En~g Mater. Des. 7, 528; 618
and Basis of Contemporary Life, Institution of Engineer- EDER, W.E. (1965) 'Fatigue design' (correspondence),
ing Designers, London Enflng Mater. Des. 8, 121; 191
CLAUSER, II.R., FAßlAN, R.J, and MOCK, J.A. (1965) EDER, W.E. and GOSLING, W. (1965) Mechooical System
'How materials are selected', Des.F:n~n~, July·, 109-28 Design, Pergamon, Oxford
REFERENCES AND BIBLIOGRAPHY 337
EDINBURGH, Duke of, (1961) The Engineer in Common- GARNER, W.R. and CLEMENT, D.E. (1963) 'Goodness of
wealth De•·elopmcnt (Seventh Graham Clark Lecture), pattern and pattern uncertainty', J. verbl Leamg verbl
Institution of Mechanical Engineers, London Behav. 2, 446-52
EDWARDS, P. (1955) The Logic of Moral Discourse, Free GARNER, W.R. and CREELMAN, C.D. (1964) 'Effect of
Press of G!encoe, Chicago, Illinois redundancy and duration on absolute judgments of visual
ELLIS, 0.0. and LUDWIG, F.J. (1962) Systems Philosophy, stimuli', ]. exp. Psychol. 67, 168-7-2
Prentice-llall, New York GETZELS, j.W. and }ACKSON, P.W. (1962) Creativity and
ELMAGIIRABY, S.E. (1964) 'An algebra for the analysis of lntelligence, Wiley, New York
generalised activity networks', .\fgmt Sei. 10, 494-514 GHISELIN, A. (1952) The Creative Process, University of
ENGELUART, D.C. (1962) Augmenting Human /ntellect: Califomia Press, Berkeley
A Conceptual Framework, Report SR 3223, Engineering GIBSON, Q. (1960) The Logic ol Social Enquiry, Routledge
Seiences Division, Stanford Research Institute & Kegan Paul, London
ESHERICK, E. (1963) 'Problems of the design of a design GITTER, D., GORDON, W.J. and PRINCE, G. (1964) The
system' in ConM (ed. Jones, J.C. and Thornley, D.G.), Operational Mechanisms of Synectics (draft), Synectics,
Pergamon, Oxford Cambridge, Mass.
EVANS, C.R. and NEWMAN, E.A. (1%4) 'Dreaming: an GOMBRICH, E.ll. (1960) Art and lllusion, Phaidon Press,
analogy from computers', New Seienfist 24, 577-9 London
Gooo, l.J. (1%2) 'Rational decisions', J. R. statist. Soc.
FAIRES, V.M. (1955) Design ol Machine Elements, 814, 107-14
Macmillan, London GOODE, H.H. and MACHOL, R.E. (1957) System Engineer-
F ARR, M. (in press) Design Management, Hutchinson, London ing, McGraw-llill, New York
FARRADANE, J. (1963) lnf. Stor. Ret. I, 3-11 GORDON, W.J.J. (1%1) Synec~ics: The Development ol
FEARN, \\!. -(1962) 'Some further thoughts on its (work Creative Capacity, Harper, New York
study) application and ~xamples as applied to plant GORE, W.J. and nvsoN. J.W. (1964) The Makinil ol
Iayout and drawing office techniques and procedures Decisions, Free Press of G!encoe, Chicago, lllinois
with partial reference to construction', Chem. Engr, GORE, W.J. and SILANDER, F.S. (1959) 'A bibliographical
October, A98-A104 essay on decision-making', Admve Sei. Q. 4, 97
FEDERATION OF BRITISH IN DUSTRIES (1961) lndustrial GosLING. W. (1962) The Design of Engineering Systems,
Research in Manufacturing lndustry, 1959-60, F.B.I., lleywood, London
London GOSLING, W. (1%3) 'The relevance of system engineering'
FEIGL, H. (1952) 'Validation and vindication' in Readings in CoDM(ed.Jones, J.C. and Thomley, D.G.), Pergamon,
in Ethical Theory (ed. Sellars, W. and llospers, J .), Oxford
Appleton-Century-Crofts, New York GRA YSON, C.J. (1960) Decisions under Uncertainty:
FEILDEN,G.B.R. (1963) Engineerin/1 Design, Her \1ajesty's Drilling Decisions by Oil and Gas Operators, Harvard
Stationery Office, London University Press, Cambridge, Mass.
FERBER, R. and WALES, II.G. (1958) Motivation and GREGORY, S.A. (1963) 'Creativity in chemical engineering
Market Behaviour, lrwin, llowewood, 111. research' in Symp. Productivity in Research (ed. Pirie,
FINKELSTEIN, L. (1963) 'The principles of measurement', J .M.), Institution of Chemical Engineers, London
Trans,Soc.lnstrum. Techno/. 15, 181 GREGORY, S.A. (1%4a) 'The development of an automatic
FIRTH, D. (1965) 'On being a designer' in Engineering adsorption drier, Parts 1 and 2', Chem. Engr, December,
Materials and Design Con/;, Engineering Materials and CE293-302
Design, London GREGORY, S.A. (1964b) 'Art, technology and creativity'
FISHBURN, P.C. (1964) Decision and Value Theory, Wiley, in Creativity Symp., Design & Innovation Group,
New York Birmingham
FORRESTER, J.W. (1961) lndustria/ Dynamics, Wiley, GREGORY, S.A. (1965) 'Creativity and the total environ-
New York ment', in Seminar on lndustrial Creativity, Foundation
FREUDENSTEIN, F. and SANDOR, G.N. (1%1) 'On the for Research on Human Behaviour - DYNAMAR,
Surrnester points of a plane', Trans. Am. Soc. Mech. 8azain ville & Sunbury
Engrs 83, 107 GREGORY, S.A. (1%6) 'Basic strategies in process
FREUDENTHAL, H. (1963) The Concept and the Rote of development' in Symp. European F ederation of Chemical
the Model in Mathematics and Natural and Social Engineering (to be published by Institution of Chemical
Sciences, Gordon & Breach, New York Engineers, London)
FRIEDMANN, G. (1%1) The Anatomy ol Work, Heinemann, GREGORY, S.A. and BURDIS, A. de V. (1965) Discussion
London at Seminar on lndustrial Creativity, Foundation for
Research on Human Behaviour DYNAMAR,
GADDIS, P.O.(l959) 'The project manager', Harv. business Bazainville & Sunbury
Rev. May-June, !19-1}7 GUILFORD, J.P. (1959) 'Traits of creativity' in Creativity
GADDIS, P.O. (1%1) 'The age of massive engineering', and its Cultivation (ed. Anderson, ll.H.), Harper, New
Harv. business Rev. J anuary-February, 138-45 York
GARGIULO, G.R., IIANNOCH, j., HERTZ, D.B. and
ZANG, T. (1961) 'Developing systematic procedures JIADAMARD, J. (1949) The Psychology ol Invention in the
for directing research programs', l. R. E. Trans. Engng Mathematical Field, Princeton University Press,
Mgmt EM8, 24-9 Princeton, N.J.
HALL, A.D. (1962) A Methodology for Systems F:ngineering, JOHNSTONE, R.E. and THRING, M.W. (1957) Pilot Plants,
Van Nostrand, Princeton, N.J. Models, and Scale-up Methods in Chemical Engineering,
HALL, E.W. (1952) What is Value? An Essay in Philo- McGraw-Hill, New York
sophical Analysis, Jlumanities, New York JONES, J.C. (1963) 'A method of systematic design' in
HALL, M.F. (1965) 'Communication within organisations', CoDM (ed. Jones, J.C. and Thornley, D.G.), Pergamon,
]. Mgmt Stud. 2. 54-69 Oxford
HAPPEL, J. (1958) Chemical Process Economics, p.27; JONES, J .C. (1964) Methods of Assessing Car Seal Comfort,
179; 187 el seq., Wiley, New York Industrie! Section, Ergonomics Research Society, London
HARE, R.M. (1952) The Language of Morals, Oxford Uni- JONES, J .C. (in press) 'Systematic design methods and
versity Press, London the building design process' in Proc. 3rd Congr. Inter-
HARRIS, J.S. (1961) 'New product profile chart', Chem. national Council for Building Research Sturlies and
Engng News, 17th April, 110-18 Documenlation, 1965
HARTMAN, R.S. (1958) 'General theory of value' in JONES, J.C. and THORNLEY, D.G. (1963) Conference on
Philosophy in the Mid-cenlury (ed. Klibansky, R.) Design Methods (CoDM), Pergamon, Oxford
La Nuova ltalia, Florence
JONES, M.H. (1962) Executive Decision Making, lrwin,
IIARVARD CASEBOOKS IN PRODUCTION MANAGEMENT
Homewood, Ill.
(1964) 'The management implications of PERT, in
Operations Planning md Control, Wiley, New York
HARVEY, J. (1950) The Gothic World, Batsford, London KANTOROWICZ, K. (1958) The Definition of Law,
HAYEK, F.C. (1953) Counter-revolution of Science, Free Cambridge University Press, London
Press of Glencoe, Chicago, Illinois KARGER, D. W. (1960) The New Product, The lndustrial
HESSE, M.B. (1961) Forces and Fields, Nelson, London Press, New York
HESSE, M.B. (1963) Models a~d Analogies in Science, KARPLUS, W.J. and SOROKA, W.W. (1959) Analogue
Sheed and Ward, London Methods in Compulation and Simulation, McGraw-Hill,
HETENYI, M.I. (1950) Handbook of Experimental Stress New York
Analysis, Wiley, New York KAST, F.E. and ROSEN ZWEIG, J .E. (1963) Science,
HIGGIN. G. and JESSOP, N. (1965) Communications in the Technology and Management, McGraw-Hill, New York
Building lndustry, Tavistock, London KLINE, S.J. (1965) Simililude and Approximation Theory,
HlTCH, C.J. (1957) 'Operations research and national McGraw-Hill, New York
planning: a dissent', Opl Res. r;, 718-23 KOFFKA, K. (1935) Principles of Gestalt Psychology,
JIODNETT, E. (1955) The Art of Problem So/ving, Harper, Hercourt Brace, New York
New York KRAMERS, H. and WESTERTERP, K.R. (1963) Elements of
HOGMANDER, O.C. (1962) lmproving Coordination between Chemical Reactor Design and Operation, p.58, Chapman
the Design Engineer and Marketing, A.S.-M. E.Paper 62- & Hall, London
MD-15 KRICK, E. V. (1965) An lntroduction to 'F.:ngineering and
HOLLIDA Y, L. and MANN, J. (1964) 'The structure and Engineering Design, Wiley, New York
properties of complex and heterophase materials' in
Proc. Symp. Advances in Materials, Pergamon, Oxford LANGHAAR, H.L. (1951) Dimensional Analysis and
HOPKINS, R.C. (1961) 'Possible applications of infor- Theory of Models, Chapman & Hall, London
mation theory to management control', I.R. E. Trans. LARRABEE, H.A. (1945) Reliable Knowledge, Houghton
F:ngng Mgmt EMB, 40-8 Mifflin, Boston, IVIass.
IIRONES, J.A. (1960) 'On automatic control' Qames LA THAM, R.L. (1965) 'Problem analysis by the logical
Clayton Lecture) in Proc. Symp. Recent Mechanical approach system' in Design F:ngineering and Manage-
Engineering Developmenls in Automatie Control, ment Conf., P.E.R.A., Melton Mowbray, Leics.
Institution of Mechanical Engineers, London
HUSSERL, E. (1965) 'Philosophy and the crisis of European LATHAM, R.L., TAYLOR, H.K. and TERRY, G.J. (1965)
man' in Phenomenology and the Crisis of Philosophy, 'The problern analysis by logical approach system' in
Engineering Materials and Design Conf., Engineering
llarper, New York
HYMAN, R. (1963) 'Creativity', lnt. Sei. Techno[, August, Materials and Design, London
LE BRETON, P.P. and HENNING, D.A. (1961) Planning
51-2; 54-6; 58
Theory, Prentice-Hall, New York
I.B.M. (1962) General Information Manual: Automated LEVINE, L. (1964) Methods for Solving 'F.:ngineering
Design Engineering Problems Using Analog Computers, McGraw-Hill, New
I.MECH.E. (1965) 'Trends in machine tool design', Charl. York
Mech. Engr 12, (3) 163 LEVY, H. (1938) Philosophy for a Modem Man, Gollancz,
London
LEWIS, B.N. (1963) 'Communication in problem-solving
JACKSON, Sir R. (1964)'The Volta River project', Progress
50, 146-61 groups' in CoDM (ed. Jones, J.C. and Thornley, D.G.),
J AQUES, E. (1956) Measurement of Responsibility, Pergamon, Oxford
LICHFIELD, N. (1964) 'Implementing Buchanan: problern
Tavistock, London
of social choice', Traft. F:ngng Control 12, 498
jOHNSON, R.A., KAST, F.E. and ROSENZWEIG, J.E.
(1963) The Theory and Management of Systems, McGraw- LOCKYER, K.G. (1964) An lntroduction to Critical Path
llill, New York Analysis, Pitman, London
LOUGHBOROUGH COLLEGE OF ADVANCED TECHNO- MILLER, D.W. and STARR, M.K. (1960) Executive Deci-
LOGY (1965) Symp. Process and Project F.:ngineering: sions and Operations Research, PrenHce-Hall, New York
Jones, D.W. 'The work of the project engineer' MIN SKY, M. (1963) 'Steps toward artificial intelligence'
Ransome, J .A. 'The commissioning engineer' in Computers and Thoughts (ed. Feigenbaum, E.A. and
Smith, F.A. 'Controlling constructional work' Feldman, J .), McGraw-llill, New York
Terry, D.P. 'Estimating and cost control' MITCHELL, R.K. and CRESWICK. F.A. (1964) The Syn-
Wuhrman, C.M. 'Procurement/purchasing' thesis and Use of Mathematical Models· as Aids to
LUCE, R.D. and RAIFFA, H. (1957) Gamesand Decisions, Design, A.S.M. E. Paper No. 64-MD-45
Wiley, New York MODER, j .J. and PHILLIPS, C. R. (1964) 'Project Manage-
ment with CPM and PERT, Reinhold, New York
MOORE, G.E. (1912) Ethics, Oxford University Press,
McCRORY, R.J. (1963) The Design Method: A Scientific
London
Approach to Valid Designs, A.S.M.E. Paper No.63-
MORLAND, G.W., ATKIN, K. and GANULES, S. (1965)
MD-4
'Structurol model techniques and their application to
MCCRORY, R.J. (1965) 'Elements of realism in mechanical
oil engine design', Chart. Mech. Engr 11, 161-2
reliability', in Mechanica/ Reliability Concepts Paper,
MORTLOCK, J.R. (1964) 'Computers and engineers'
A.S.M.E.
(Chairman's address, Science and General Division)
McCRORY, R.J., WILKINSON, W.H. and FRINK, D.W.
Proc. lnstn elect. Engrs III (1) 153-6
(1963) Synthesis of Concepts in the Design Method,
MOULTON, A. (1965) The Case for Innovation in Design,
A.S.M.E. Paper No. 63-MD-37
Council of Industrial Design, London
MCPHERSON, J .H. (1964) 'Environment and training for
MUMFORD, L.S. (1963) 'The loneliness of the long-range
creativity' in Creativity: Progress and Potential (~.
forecaster', Chemy lnd. 9th November, 1788-96
Taylor, C.W.), McGraw-Hill, New York
MCPHERSON, J.H. (1965) 'llow to manage creative
NEWELL, A., SHAW, J.C. and SIMON, H.A. (1963) 'Chess-
engineers', Mech. Engng, February, 32-6
playing programs and the problern of complexity' in
MACKWORTH, N.H. (1964) 'Originality', Am. Psychol. Computers and Thought (ed. Feigenbaum, E.A. and
20, 51-66
Feldman, J.), McGraw-llill, New York
MALINOWSKI, 8.(1922) Argonauts of the Western Pacific,
NEWELL, A. and SIMON, !I.A. (1963) 'GPS, a program that
Dutton, New York
simulates human thought' in Computer and Thought
MALLYA, R.V., I(ING, J.R. and EILON, S. (1964) 'A study
(ed. Feigenbaum, E.A. and Feldman, J.), McGraw-Hill,
of optimal dimensions of cylindrical cans', lnt.].Prod.
New York
Res. 3, 35-9
NEWELL, A. and SIMON, H.A. (1964) 'Problem-solving
MANN, R.W. (1963) 'Engineering specifications for a man-
machin es', lnt. Sei. Techno/. 12, 48-9; 53-62
computer system for design', AFIPS Conf. Proc. 23
NEWMAN, W.ll. and SUMMER, C.E. (1961) The Process o(
(Spring Joint Computer Conference), Cleaver-Hume,
Management, McGraw-Hill, New York
London NIEBEL, B.W. and ßALDWIN, E.N. (1963) DesibJing for
MARGEN, P.ll. (1960) /Vuclear Reactor Optimisation,
Production, Irwin, Homewood, Ill.
Temple Press, London
NORBERG-SCHULZ, C. (1963) Intentions in Architecture,
MARPLES, D.L. (1960) The Decisions of Engineering
Allen & Unwin, London
Design, Institution of Engineering Designers, London
NORDEN, P.V. (1963) 'Resource usage and network plan-
MARPLES, D.L. (1961) 'The decisions of engineering ning techniques' in Operations Research in Research
design' I.R. F.:. Trans. F.:ngng Mgmt EM8, 55-71 and Dcvelopment (ed. Dean, ß.V.), Wiley, New York
MASLOW, A.ll. (1954) Motivation and Personality, Barper NORRIS, K.W. (1963) 'The morphological approach to
& Row, New York engineering design' in CoDM (ed. Jones, J.C. and
MASLOW, A.ll. (1962) Toward a Psychology of Being, Van Thornley, D.G.), Pergamon, Oxford
Nostrand, Princeton, N.J. NOWELL-SMITH, P.ll. (1954) Ethics, Penguin, Harmonds-
MATCHETT, E. (1963) 'The controlled evolution of an worth, Middlesex
engineering design' in Conf. Systematic Design Methods,
Engineering Employers' West of England Association O'BRIEN, J.J. (1965) CPM in Construction Management,
MATCHETT, E. (1964) 'Design training in post-graduate McGraw-llill, New York
courses' in Conf. Teaching of Engineering Design (ed. O'DOHERTY, E.F. (1963) 'Psychological aspects of the
Booker, P.j.), Institution of Engineering Designers, creative act' in Co DM (ed. jones, J .C. and Thomley,
London D.G.), Pergamon, Oxford
MATOUSEK, R. (1963) Engineering Design: A Systematic OLSEN, II.F. (1958) Dynamical Analo~ies, Van Nostrand,
Approach, Blackie, Glasgow Princeton, N.J.
MERLEAU-PONTY, M. (1963) Les Seiences de l'Homme OSBORN, A.F. (1963) Applied lmaaination, Scribner,
et Ia Phenom&nologie, Paris New York
MESAROVIC, M.D. (1964) Views on General Systems
Theory, Wiley, New York PACKARD, V. (1957) The Hidden Persuaders, Longmans
METCALF, H.C. and ÜRWICK, L. (1941) Dynamic Admini- Green, London
stration, Management Publications, London PAGE, J.l(. (1963) 'A review of the papers presented at the
MILES, L.D. (1961) Techniqucs of Value Analysis, conference' in Co DM (ed. Jones, J .C. and Thornley,
McGraw-Hill, New York D.G.), Pergamon, Oxford
PAGE, J.K. (1964) 'Environmental research using models', RHENMAN, E. (1964) 'Research planning' in Proc. 3rd
Architects' ]. 139, 587-93 lnt. Conf. Operational Research (ed. l{reweras, G.
P ANKIIU RST, R. C. (1964) Dimensional Analysis and and Morlat, G.), English Universities Press, London
Scale Factors, Institute of Physics and the Physical ROBERTS, E.B. (1964) The Dynamics of Research and
Society, London Development, Harper & Row, New York
PARKINSON, C.N. (1958) Parkinson's Law: or the Pursuit RODGER, L. W. (1965) Marketing in a Competitive Economy,
of Progress, Murray, London Hutchinson, London
PARTON, K.C. (1964) The Digital Computer, Pergamon, ROGERS, A.E. and CONNELLY, T.W. (1960) Analog
Oxfonl Computers in Engineering Design, rvJcGraw-Hill, New
PEARSON, Sir D. (1965) 'Working with new materials' in York
The World in 1984, Vo!. 1 (ed. Calder, N.), Penguin, ROGERS, E.M. (1962) Diffusion of Innovations, Macmillan,
llarmondsworth, Middlesex New York
PENNY, A.M. (in press) Study of the Predesign Planning RüGET, P.rvl. (1852) (rev.ed.1953) Thesaurus of F.nglish
of a Teaching Hospital, Royal Institute of British Words and Phrases, Penguin, Ilarmondsworth, Middlesex
Architects, London ROSENSTEIN, A.B. (1960) 'Report of the working group on
PERLMUTTER, H.K. (1965) The Building of Indispensable the design process', Proc. 1st Conf. Engineering Design
lnstitutions, Tavistock, London Education, Case Institute of Technology, Cleveland,
PlAGET, J. (1953) Logic and Psychology, Manchester Ohio
University Press, Manchester ROSENSTEIN, A.B., RATHBONE, R.R. and SCHNEERER,
PILDITCII, J. (1964) 'Good design helps to sell any W.F. (1964) Engineering Communications, Prentice-Hall,
industrial product', lnd. Advertng Mrktng I, Tl-9 New York
Ross, D. T. and RODRIGUEZ, J .E. (1963) 'Theoretical
PILDITCH, J. and SCOTT, D. (1965) The Business of
foundations for the computer-aided design system',
Product Design, Business Publications, London AFIPS Conf. Proc. 23(Spring Joint Computer Conference),
PIRIE, J .M. (1957) Scaling-up of Chemical Plants and Cleaver-Hume, London
Processes, Institution of Chemical Engineers, London Ross, I.M. (1964) The Definition of Design (communication
POLY A, G. (1945) How to Solve lt, Princeton University to the ad hoc Committee on Electronics Design),
Press, Princeton, N.J. Institution of Electrical Engineers, London
POLY A, G. (1954) 'Patterns of plausible inference' ROSSMAN, J. (1964) lndustrial Creativity: the Psychology
(Vol. 2 of Mathematics and Plausible Reasoning), of the Inventar, University Rooks, New York
Princeton University Press, Princeton, N.J. ROYAL INSTITUTE OF 8RITISH ARCHITECTS (1965)
POPPER, Sir K.R. (1959) Logic of Seienlilie Discovery, Handbook of Architectural Practice and Management,
Hutchinson, London R.I.B.A., London
POPPER, Sir K.R. (1960) Poverty of Historicism, Routledge RUBINSTEIN, A.H. and HABERSTROH, C.j. (1960) Some
& Kegan Paul, London Theories of Organization, Irwin, Howewood, Ill.
POPPER, Sir K.R. {1962) The Open Society and its RUGG, H. (1963) Imagination, Barper & Row, New York
Enemies, Routledge & Kegan Paul, London
POPPER, Sir K.R. (1963) Conjectures and Refutations, SALOMON, M. and BROWN, R. V. (1964) 'Commercial fore-
Routledge & Kegan Paul, London casting problems', Opl Res. Q. 15, 239-47
POSTAN, M.M., HAY, D. and SCOTT, J.D. (1965) Design SALZ ER, J .M. (1961) 'Evaluating design of complex
and Development of Weapons, Her Majesty's Stationery systems' in Systems: Research and Design (ed.
Office, London Eckman, D.P.), Wiley, New York
Product Engineering (1965) series on 'Value engineering' SCHER, P. (1965) 'Theory of design: report on a Birmingham
symposium', Architects' ]. 140, 1223-8
PUTN AM, H. (1964) 'The compleat conversationalist: a
SCHOENBERG, A. (1950) Style and ldea, Williams and
"systems approach" to the philosophy of language' in
Norgate, London
Views on General Systems Theory (ed. Mesarovic, M.D.),
SCHUETZ, A. (1951) 'Choosing among projects of action',
Wiley, New York
Philosophy phenom. Res. 12, 161-84
PYE, D. (1964) The Nature of Design, Studio Vista, London
SELWYN, R.S. (1965) 'Materials selection: a new approach',
Engng Mater. Des. November, 834-6
QUINN, J .8. (1961) 'Lang-range planning of industrial SHACKLE, G.L.S. (1961) Decision, Order and Time,
research', Harv. business Rev. July-August, 88-102 Cambridge University Press, London
QUIRK, G.G. (1961) 'Predict product performance', Prod. SHAHBENDERIAN, A.P. (1963) 'The economic life and
Engng, April, 24 size of process plant in underdeveloped countries',
Chem. Engr, May, 168
RAPOPORT, L. and DREWS, W.P. (1962) 'Mathematical SHAHBENDERIAN, A.P. (1964) 'lnfluence of profitability
approach to Iang-range planning', Harv. business Rev. criterion on chemical reactor systems', Chem. Process
May-June, 75-87 Engng 45, 647
RESWICK, J.B. (1965) Prospectus for the Engineering SHELLY, M.W. and 8RYAN, G.L. (1964) Human judgments
Design Center, Case Institute of Technology, Cleveland, andOptimality, Wiley, New York
Ohio SHONE, Sir R. (1965) 'NEDC projections' in Models for
RE V ANS, R. W. (1964) Standards for Morale: Cause and Decision (ed. Bemers-Lee, C.M.), English Universities
Effect in Hospitals, Oxford IJniversity Press, London Press, London
SHUBIK, M. (1958) 'Studies and theories of decision THOMPSON, Sir D'A.W. (1917, 1942) On Growth and Form
making', Admve Sei. Q. December, 289-306 (2 vols.), Cambridge University Press, London
SHUBIK, M. (1964) 'Behaviouristic or normative decision TOMLINSON, C. (1853) Cyclopaedia of Uselul Arts and
criteria' in Proc. 3rd lnt. Conf. Opl Res. (ed. Kreweras, Manulactures, Virtue, London
G. and Morlat, G.), English Universities Press, London TOULMIN, S. (1950) The Place of Reason in F:thics,
SIMON, li.A. and NEWELL, A. (1958) 'Heuristic problern Cambridge University Press, London
solving: the next advance in operations research', TOYNBEE, A.J. (1960, abridged) A Study ol History,
Opl Res. January-February, 1-10 Oxford University Press, London
SIMON, li.A. and SIMON, P.A. (1962) 'Trial and error TRAINER, T.M. and GLASGOW, J.S. (1965) 'Materials
search in solving difficult prob\ems: evidence from the selection in the design method', Mech. F:ngng 87, 41-3
game of chess', Behav. Sei. 7, 425-9 TRAYSER, D.A. and CRESWICK, F.A. (1962) 'Using a
SITTING, J. (1963) 'Defining quality costs', 7th Conl. mathematical model as a design tool', A.S.M.I':. Design
European Organisation for Quality Control, Copenhagen Engineering Conl., Preprint No. 62-MD-1
SMILES, S. (1874) Lives of the Engineers, VoLl, Murray, TURNER, B.T. (1964) 'The effective use of engineers in
London the changing industrial environment' in Engineering
SMITH, G.ll. (1954) Motivation Research in Advertising and Education lllld Career Patterns, Institution of
Marketing, McGraw-Hill, New York Mechanical Engineers, London
SMITH, I.M. (1964) Spatial Ability, University of London TWEEDDALE, J .G. (1962) Metallurgical Principles for
Press, London Engineers, Iliffe, London
SMITH, R.A. (1965) 'Designing the materials we need' in
The World in 1984 (ed. Calder, N.), Penguin, Harmonds-
worth, Middlesex VALENTINE, C.W. (1962) The F:xperimental Psychology
SPORN, P. (1964) Foundations ol Engineering, Pergamon, of Beauty, lldethuen, London
Oxford VICKERY, B.C. (1960) Faceted Classification, ASLIB,
STACEY, N.A.Il. and WILSON, A. (1963) lndustrial Market- London
ing Research: Management and Technique, llutchinson, VON FANGE, E.K. (1959) Professional Creativity, Prentice-
London llall, New York
STARKEY, C.V. (1964) 'Electron gun assembly', Engineer, VON NEUMANN, J. and \IORGENSTERN, 0. (1953) Theory
14th August, 246 of Games and Economic Behaviour, Princeton University
STARR, M.K. (1963) Product Design and Decision Theory, Press, Princeton, N.J.
Prentice-Hall, New York
STEIN, E. (1964) Amold Schoenberg: Leiters, Faber & WADE, G.P. (1960) in Work Study (ed. Currie, R.M.), Pitman,
Faber, London London
STEVENSON, C.L. (1944) Ethics and Language, Yale WALLACE, P.j. (1952) The Technique of Design, Pitman,
University Press, New !Iaven, Conn. London
STONE, R. (1965) 'Social accounting matrix models: a WALLAS, G. (1926, reissue 1931) The Art of Thought,
framewolk for economic decisions' in Models for Cape, London
Decision (ed. Bemers-Lee, C.M.), English University WALTER, W.G. (1953) The Living Brain, Duckworth, London
Press, London WARD, E.P. (1965) 'The dynamics of product planning',
STUCKENSCHMIDT, H.ll. (1959) Amold Schoenberg, Chart. Mech. Engr 12. 29-34
John Calder, London WERKMEISTER, W.ll. (1952) 'Problems of value theory',
STURT, G. (1923) The Wheelwright's Shop, Cambridge Philosophy phenom. Res. 12, 495-512
University Press, London WERKMEISTr:R, W.H. (1<154) 'Prolegomena to value theory',
SUTHERLAND, l.E. (1963) 'Sketchpad: a man-machine Philosophy phenom. Res. 14, 293-308
communication system', J\FIPS Conf. Proc. 23 (Spring
WERTHEIMER, M. (1945) Produclive Thinking, llarper,
Joint Computer Conference), C!eaver-llume, London
New York
SZUPROWICZ, B.O. (1%3) 'Network planning and economic
WILLIAMS, R.R. and ~:coTT, W.ll. (1965) Investment
development', New Seienlist 18, 728-30
Proposals and Decisio1s, Allen & Unwin, London
TAO, D.C. (1964) Applied Linkage Synthesis, Addison- WILLIAMS, L.A. and FINLAYSON, R.J. (1963) 'Engineer-
Wesley, Reading, Mass. ing market research: t1eory and practice', Proc. lnstn
TARSKI, A. (1954-5) 'Contributions to the theory of models, mech. Engrs 177 (32) 877-91; 892-5
Parts 1-3', lndag. math. Ned. Akad. Wet. 16 WILLSMORE, A.W. (1956) l'rocluct De,·elopment and Design,
TAYLOR, C.W. (1964) Creativity: Progress and Potential, Pitman, London
lldcGraw-llill, New York WILSON, A. (1 <J66) The Marketing ol lndustrial Products,
TAYLOR, D.W., BERRY, P.C. and BLOCK, C.H. (1958) llutchinson, London
'Does group participation when using brainstorming WILSON, W.E. (l'l65) Concepts ol T<:ngineering System
facilitate or inhibit creative thinking?', Admve Sei. Q. Design, McGraw-llill, New York
3, 22-47 WOLSTENIIOLME, A.W. (1<)62) 'A generat appraisal of the
TAYLOR, F.A. (1962) Colour Technology, Oxford Univer- roJe of work study in the design process', Chem. Engr,
sity Press, London October, A!I8-A97
TAYLOR, J.W. (1961) How to Create New ldeas, Prentice- WOODGATE, li.S. (1%4) Planning by Network, ßusiness
llall, New York Publications, London
WOODWARD, J. (1965) lndustrial Organisation: Theory and MCCORMICK, E.J. (1964) Human Factars Engineering,
Practice, Oxford TJniversity Press, London McGraw-Jiill, New York
MORGAN, C.T., COOK,J.S., CHAPANIS, A. and LUND, M.W.
YOULE, P.V. (1963) 'Mathematical models: an aid to (1963) Human 1\ngineering Guide to 1\quipment Design,
industrial chemistry', Chemy lnd. 10th August, 1336-8 McGraw-llill, New York
MUCKLER, F.A. (editor) Human Factars (scientific journal
ZUCKERMAN, SirS. (1961) Report of the Committee on the published bi-monthly), The Johns Hopkins Press,
Management and Control of Research and Development Aaltimore
(7.uckennan-Gibb Report), ller Majesty's Stationery MURRELL, K.F.H. (1965) 1\rgonomics- Moo in His Warking
Office, London 1\nvironment, Chapman & Hall, New York
ZWICKY, F. (l!J48) 'The morphological method of analysis WELFORD, A. T. (editor) 1\rgonomics (scientific journal
and construction' in Studies and l':ssays (Courant published quarterly), Taylor & Francis, London
Anniversary Volume), Interscience, New York WOODSON, W.E. and CONOVER, D.W. (1964)//uman En-
gineering Guide for Equipment Designers, University
Ergonomics of Califomia Press and Cambridge University Pre~,
Berkeley and London
References
CONRAD, R. (1960) 'Experimental psychology in the Workspace
field of telecommunications', Ergonami es 3, 289-96
FLOYD, W.F. and ROBERTS, D.F. (1958) 'Anatomical and BRITISH STANDARD 853044 (1958) Anatomical, Physio-
physiological principles in chair and table design', logical and Anthropometrie Principles in the Design of
Ergonomics 2, 1-16 Office Chairs and Tab/es, British Standards Institu-
LEHMANN, G. (1958) 'Physiological measurements as a tion, London
basis of work organisation in industry', 1\rgonomics I, BRITISH STANDARD BS3079 (1959) Anthropometrie
328-44 Recommoodations for Dimensions of Non-adjustable
MACKWORTH, tl.ll. (1955) Work Design and Training for Office Chairs, Desks and Tab/es, British Standards
Future lndustrial Skills (Sir Alfred Herbert Lecture), Institution, London
Institution of Production Engineers, London · BRITISH STANDARD BS3404(1961)Anthropometric Recom-
RONCO, P.G. and SAWYER, H.L. (1962) 'A survey of mendations for Dimoosions of Office Machine Operators'
Russian Iiterature related to human factors engineering', Chairs and nesks, British Standards Institution, London
Human Factars 4, 107-11 JONES, J.C. (1960) 'Fitting for action, parts I and II',
WELFORD, A. T. (1960) Ergonomics of Automation, D.S.I.R. Design 135, 38-42; 137, 49-52
Series No. 8, Her Majesty's Stationery Office, London ]ONES, J.C. (1963) 'Anthropometrie data and fitting
trials', Architects' ]. 6th February, 317-25; 13th Febru-
ary, Info. Sheets 1185 and 1186
General Texts
(Various tables of data on body size may be found in this
CIIAPANIS, A., GARNER, W.R. and MORGAN, C.T. (1949) last reference and in the generat reference books gi ven
1\pp/ied 1\xperimental Psycho/ogy, Wiley and Chapman above.)
& Hall, New York and London
CHAPANIS, A. · (1965) Man-Machine Engineering, Environment
ßehavioural Science in Industry Series, Tavistock,
London BEDFORD, T. (1948) Basic Principle of Ventilation and
D.S.l.R. BOOKLET SERIES: Ergonomics for /ndustry Heating, II.K. Lewis, London
(free from D.S.I.R., Information Division, State llouse, BRITISH LIGHTING COUNCIL (1965) Guide on How to
London WC1) 1\void Discorniort Glare, R.L.C., 16 Lancaster Place,
No.l. SINGLETON, W.T. (1962) The lndustrial Use of London W2
F:rgonornics, ller Majesty's Stationery Office, London D.S.I.R. BOOKLET SERIES: Factory Building Studies
No.2. SHACKEL, 8. and WHITFIELD, D. (1963) No.l. KEYTE, M.J. and GLOAG, II.L. (1960) The
Instruments and People, Her Majesty's Stationery Lighting of Factories, ller Majesty's Stationery
Office, London Office, London
No.3. GRIEW, S. (1963) Design of Work Iot the Disabled, No.2. .ALDERSEY-WILLIAMS, A.G. (1960) Noise in
ller Majesty's Stationery Office, London Factories, Jler Majesty's Stationery Office, London
No.4. ßELßlN, R.M. (1963) lnspection and Human ILLUMIN ATING ENGINEERING SOCIETY (1961) The /ES
Eflicioocy, Her Majesty's Stationery Office, London Code: Recommendations for Good lnterior Lighting,
No.5. SELL, R.G. (1964) Ergonomics versus Accidents, I.E.S., 32 Victoria Street, London S W1
ller Majesty's Stationery Office, London (The general references given above also deal with most
No.6. ßROADUeNT, D.E. (1964) Noise in lndustry, aspects of the environment in reasonable detail.)
Her Majesty's Stationery Office, London
No.7. PROVINS, l;{,A. (1965) Men, Machines and
Controls, Her Majesty's Stationery ()ffice, London •
NAME INDEX
See also References and Bibliography
Ackoff, R.L. 111-2 Bross, I.D.J. 134

Adelson, R.M. 2<)0-1 Brown, R.H. 162
Alexander, C. 2%, 300 Brown, R.V. 36
Alexander, T. 117 Brown, W. 207
Alexander, W.O. 235 Brunel, I.K. 117
Alger, J.R.M. 24, 85, 88, 93 ßryan, G.L. 134
Allen, 65 Bryant, Sir A. 45
Allport, G. W. 68 Buchanan, C. 39
Alves, G.E. 14!"1 Buck, C.H. 279, 285
Archer, L.B. 77-8, 89, 95, 154-5 Burcis, A. de V. 81, 115, 118, 3~
Aristotle, 124 Bums, T. 67, 284-5
Ashby, W.R. 328 Byme, A. 153
Asimow, M. 7, 78, 85, 95, 147, 151, ~1. ~S-6, 298, 302,
304, 306 Canguilhem, G. 143
Atkin, 1<. 149 Carothers, W. 237
Carritt, F..F. 37
Baker, R.J.S. 288 Carter, C.F. 67
Baker, S.S. 113 Cezanne, P. 113
Barr, A.H. 113 Cheaney, I':.S. 14
Bartlett, Sir F. 114 Chomsky, A.N. 80, 105
Bartley, S.H. 145 Churchman, C.W. 65,95
Bartley, W.W. 63 Clausen, II. 312
Baxter, P.A. 78 Clauser, II. R. 216
Bayes, T. 138 Conceicao, P.H. 173
Beament,J.W.L.143 Connelly, T.W. 146
Beck, H.V. 317-8 Conrad, R. 52
Bellman, R. 150 Cooklin, L. 36
Bello, F. 302 Coons, S.A. 79
Bennett, H. 246 Cooper, D. 113
Bemoulli, J. 131 Cooper, W.W. 134
Berry, P.C. 308 Comey, C. T. 219
Beyer, R. 312 Comford, E.C. 250
Binsted, D.S. 78 Crawford, R.P. 116
Birchall, H. 78 Cress, II. A. 14
Block, C.H. 308 Creswick, F.A. 151
Booker, P.J. 23, 26, 134, 138, 143, 147, 295 Currie, R.M. 287
Booz, 65
Boston, D.C. 307 da Vinci, L. 323
Boucher, D.F. 149 Dante, 3
Boulding, K. E. 78, 148, 152 Davies, A.L. 211
Braque, G. 113 Davies, Lord L. 60
Brech, E.F.L. 287 Davies, O.L. 290
Brech, R. 217 Davis, M. 111
Briggs, A.H. 77, 183, 279 Davis, R. 259-60, 262, 267
Bright, J .R. 288 De Vries, M.G. 36
Brillouin, L. 148 Debussy, C.A. 113
Brindley, J. 4 Descartes, R. 152
Broadbent, G.H. 79, 81, 111, 154 Dewey, J. 112-4
Brack, P. 150 Drucker, P. 36
Brogden, H:E. 81 Duckworth, j.C. 249
343
Duhem, P.M. 127 Hay, 0. 289

Oyson, J.W. 134 Hays, C. V. 24, 85, R8, 93
Head, Sir H. 114
F.ames, 300 Heisenberg, W. 148
Eder, W.E. 19, 23-4, 79, 138, 203, 311, 315 Hesse, M. B. 143
F.dinburgh, Duke of, 20 Higgin, G. 292
F.dison, T.A. 69 Hitch, C.J. 139
F.ilon, S. 162 Hodnett, E. 6
Einstein, A. 113, 118, 148 Hogmander, O.C. 250
Ellis, F.L. 201 Holliday, L. 237
F.imaghraby, S. E. 2q() Jlrones,J.A.19
F.ngelbart, O.C. 80 Hurwicz, L. 138
F.sherick, E. 295 Husserl, E. 126
Evans, C.R. 154 Hyman, R. 114
Fabian, R.J. 216 Jackson, P. W. 114, 301

Faires, V.M. 203 Jackson, Sir R. 300
Fan, M. 301 J aques, E. 138, 262
Fanadane, J. 79, 97, 101, 279, 328 J essop, N. 292
Fanand, R.G. 246 Johnstone, R. E. 146, 150
Feam, W. 287 Jones, J.C. 23, 26, 28, 115, 88-9, 97-8, 295, 300, 305-fi, 308
Feil den, G.B. R. 19, 259, 298 Jung, C. 107
Finkelstein, L. 95
Finlayson, R.J. 36, 251 l{antorowitz, 1{. 320
Firth, 0. 150 Karger, 0. W. 249
Fishburn, P.C. 82, 95, 291 Kast, F.E. 28B
Floyd, W.F. 51 King, J.R. 162
Follett, M.P. 38 Kipling, R. 45
Forrester, J. W. 288 T<line, S.J. 149
Freud, S. 128 T<offka, 1{. 114
Freudenstein, F. 312 Kramers, H. 159
Freudenthal, H. 143 l{rick, F.. V. 7
Friedmann, G. 262
Frink, 0. W. 13, 97 Langhaar, H.L. 146
Laplace, P.S. 138
Galileo, 148 Lanabee, II.A. 89
Ganules, S. 149 Latham, R.L. 77, 88-9
Gargiulo, G.R. 82 Le Corbusier, 85
Garner, W. R. 38 Leavi tt, H.J. 134
Gauss, J.F.C. 112 Lehmann, G. 51
Getzels, J. W. 114, 301 Levy, H. 78
Ghiselin, B. 112 Lewis, ß.N. 309
Gitter, n. 117 Lichfield, N. 82, 135
Glasgow, J .S. 214 Lindzey, G. 68
Gombrich, E.H. 38 Lockyer, T<.G. 289
Goode, II.H. 30, 92, 95 Long, A. 116
Goodman, N. 121
Gordon, W.J.J. 117, 328 McCrory, R.J. 11-4, 711, R5, 97, 221, 292
Gore, W.J. 131, 134 McMullen, P. 241, 279
Gosling, W. 23-4, 30, 79, 89, 95, 111, 138, 147, 203, 291, McPherson, J .H. 261
315, 323 Macho!, R.E. 30, 92, 95
Gravenor, R.B. 290 Mackworth, N.H. 52, 153, 296, 302-3, 307-8
Grayson, C.J. 130 Malinowski, B. 60
Gregory, S.A. 3, 8, 26, 28, 35, 77, 81, 115, 118, 131, 133, Mallya, R. V. 162
135-6, 143, 155, 235, 287, 302-3, 315, 323, 327-8, 329 Mann, J. 237
Guilford, J.P. 7, 79, 154 Mann, R. W. 296, 298, 301-8
Margen, P.H. 162
Hadamard, J. 112, 118 Marples, O.L. 25, 133, 306, 317
Hall, A.O. 7, 30, 36, 65, 131, 134-5, 288 Maslow, A.H. 35, fi7, 119, 326
Hamilton, 65 Masterton, J. 246
llappel, J. 159, 161-2 Matchett, F.. 26, 77, 183, 279, 296, 306
llarris, J. S. 36, 288, 291 Matousek, R. 24, 115, 133, 202-3
Harvey, J. 19 Mayall, W.ll. 35, 37, 39
NAME INDEX 345
Mesarovic, \1. D. 78, 85, 88, qs Rogers, E.M. 37, 65, 68, 73, 329
Metcalf, H.C. 38 Roget, P.M. 115
Miles, L.D. q Ronco, P.G. 51
Mill, j.S. 304 Rosenstein, A.B. 147, 281, 306
Minsky, M. 153 Rosenzwei g, J. E. 288
Mitchell, R.l{. 151 Ross, D. T. 80
Mock, J.A. 216 Ross, I.M. 26q, 320
Moder, ] .] . 289 Rubens, P.P. 27q
Monk, A.S. 246 Rugg, II. 114, 118
Morland, G. W. 14q
Mortlock, ] . R. 152, 312 Salomon, M. 36
Moulton, A. 2q8, 306 Salzer, ] .M. 290
Mozart, W.A. 112 Sandor, G.N. 312
Mumford, L.S. 36 Savage, L.J. 138
Sawyer, H.L. 51
Needham, A.M. 175 Scher, P. 78
Newell, A. 25, 153 Schneerer, W.F. 147, 281
Newman, A.n. 37, 80, 105, 145, 154 Schoenberg, A. 113
Newman, E.A. 154 Schuetz, A. 125
Norberg-Schulz, C. 3B Scott, J. D. 289
Norden, P. V. 288 Scott, W. H. 36, 82, 288
Norris, K.W. 28, 78, 93, 97, 116, 304 Selwyn, R.S. 203
Shackel, A. 49
O'Brien, ].]. 289 Shackle, G.L.S. 140, 328
O'Doherty, F..F. 2q5, 301-2 Shahbenderian, A.P. 157-60
Osbom, A.F. 115-7, 305 Shannon, C.E. 148
Shelly, ~.W. 134
Packard, V. 113 Shone, Sir 'R. 290
Page, J.K. 134, 150, 2q6-7, 302-6, 308 Shubik, M. 82, 153
Pankhurst, R.C. 146 Silander, F.S. 131
Parkinson, C. N. 265 Simon, H.A. 25, 153, 302
Parton, K.C. 167, 173 Simon, P.A. 302
Pearson, Sir n. 235 Si ttin g. J . 250
Penny, A.~l. 59 Smiles, S. 4
Peplow, M.E. 37, 65, 279 Smith, I.'ß. 81
l'erlmutter, H. K. 38 Smith, R.A. 235
Phillips, C.R. 289 <;now, Sir C. 59
Piaget, ] . 79 Sporn, P. 7
Picasso, P. 113 Sprecher, T.B. 81
Pilditch, ] . 255 Stacey, N.A.H. 36
Pirie, ] .'-!. 150 Stalker, G.'ß. 67, 284-5
Plato, 151, 302 Starkey, C. V. 164
Pleydell-Pearce, A.G. 82, 121 Starr, '-!.K. 85, 95, 135, 138, 147, 151
Poincare, H. 112 Stein, E. 113
Polya, G. 152 Stobart, A. F. 35, 45
Popper, Sir K. R. 64, 324 Stone, R. 290
Postan, M.M. 289 Stuckenschmidt, H.H. 113
Prince, G. 117 Sturt, G. 296
Putnam, H. 80 Sutherland, l.E. 153, 307
Smprowicz, ß.O. 290
Quirk, J .B. 29
Tao, D.C. 312
Rathbone, R.R. 147, 281 Tarski, A. 143
Ratoosh, P. 95 Taylor, C.W. 81
Rennie, J. 4 Taylor, D.W. 308
Repucci, L. 119 Taylor, H.K. 17, 88-9
Reswick, J.B. 295 Taylor, J. W. 115-6
Rhenman, F.. 291 Telford, T. 4
Roberts, D.F. 51 Terry, G.J. 77, 8~9
Roberts, F.. B. 288 Thompson, Sir D'A.W. 37
Rodger, L.W. 36 Thomley, D.G. 23, 26, 85, 295
Rodriguez, }.E. 80 Thring, M. W. 146, 150
Rogers, A. E. 146 Tomlinson, C. 326
Toynbee, A.j. 81, 287 Wa!ter, W.G. 112

Trainer, T.M. 214 Ward, E.P. 3fi
Trayser, D.A. 151 Watts, R.D. 77-9, 81, 85, 97, 152, 328
Tredgold, T. 39 Wedmore, D. B. 233
Turner, B. T. 36, 79, 249, 287, 313 Welford, A. T. 52
Tweeddale, J. G. 203 Wertheimer, M. 113-4
Westerterp, K.R. 159
Valentine, C. W. 37 Whiting, C. 116
van Kooten, L.C. 153 Whitworth, Sir j. 4
Vernon, P.E. 6!1 Wi!kinson, W.ll. 13, 97
Vickery, B.C. 100 Williams, ß. R. 36, fi7, 82, 288
von Fange, E.l{. 115-fi Williams, L.A. 36, 251
von Helmholtz, H.L.F. 112 Wilson, A. 3fi
von Karman, T. 239 Wolstenholme, A. W. 78
Woodgate, II.S. 289
Wade, G.P. 77 Woodward, J. 287
Wagner, W. R. 113
Wald, A. 138 Youle, P. V. 153
Wallace, P .J. 25
Wallas, G. 112 7.uckerman, Sir S. 70
Wallis, B.N. 266 7.wicky, F. 78-9, 116, 323, 332
SUBJECT INDEX
A~ceptance (see Design product) Aspiration (see also Motivation) 40, 270-1
Action (see also Value), 121-2, 331 Assembly
and end, 124 as articulation, linkage (see also Aröculation), 21, 88,
and influences, 126 201
arbitrary, 124 as construction (see also Production engineer), 5, 29,
covert, 122 201, 204, 206-7, 297, 323
instrumental, 121-2 Associationist (see also Psychology), 114
justifiable, 121, 121! Attitude (see also Motivation), 9, 136
most effective, 122 Attribute (see also Artefact; Characteristic), 26, 85, 88,
negative, 122 94, 116
overt, 122 Automobile (see Desir,n product; Morphology)
Activity Autonomie (see also Design agent), 77, 82, 85, 108, 152,
of network, 105-8 328, 331
sampling (see also Work study), 287 Awareness (see also Need; Perception; Problem-finding),
Adoption ( .•ee Design product, general, acceptance) 46
Advertising (see also Market), 251, 258, 281, 284
Aesthetics (see also Psychology), 11, 37-8, 41-2, 80, Balance (see also Reconciliation), 140, 251
105, 108-ll, 154, 157, 255, 312 Beauty (see also Aesthetics), 109
Algorithm (see also Methodical procedure), 111, ll3, 331 Behavioural sciences (see also Anthropology; Economics;
Analogue (see also Model), 151 Ergonomics; Management science; Psychology;
computer, 146, 150, 152 Science; Sociology), 36, 41, 131, 134, 143, 269,
of thought process, 105-9, 185 'rl1, 185, 323, 330
Analysis, 26, 78, 85, 'l7, 103, 115, 167, 302, 323 Biography, utility of, 'rl7
by computer, 167-9 'Black box' (see also System), 13, 112, 118, 328
cos t ( see Cost) 'Brainstorming' (see also Creativity), 11, 26, 67, 116-8,
economics (see F.conomics) 302, 305-6
failure (see Failure) 'Breadboard', 30, 284, 317
f easi bili ty ( see F easi bili ty) Bridging (see also Network; Strategy; Tree), 29-30
sensitivity (see Sensitivity analysis) Brief (see also Need; Sponsor), 62-3, 154, 250
value (see Value analysis) British Standard Specification (see also Standard), 148,
Anastomosis (see also Network), 80, 292 207, 250, 314
Anatomy (see also Ergonomics), 51 Builder, building (see also Technology), 70-1, 201, 295
Anthropology (see also Behavioural sciences), 37, 59, 60
Anthropometry ( see also Ergonomics), 42, 51 Capability (see also Material; Resources), 12-3, 26, 225-6
Appearance (see also Aesthetics; lndustrial design), 3, Capital equipment (see Design product)
43, 298, 320 Case history (see also Precedent), 88, 95, 133-4, 155,
Approval of sponsor, 22 175-81, 183-99, 302, 322, 3'rl, 329
Architect, architecture (see also Technology), 4, 38, 41, Certainty (see also Decision; Information; Uncertainty),
43, 59, 63, 70-1, 82, 105, 150, 242, 21l5, 302, 314, 85, 135-7, 148-Q
325-6 Challenge (see also Motivation; Opportunity), 7, 16, 81,
Arrangement, 30, 146, 242-5 115, 'rl1, 'rl3-5, 291, 329
'breadboard', 30 Characteristic (see also Artefact; Attribute; System),
Artefact (see also Attribute; Characteristic; Product; 319-20, 322
System), 20-1, 31, 78, 85,. 92, 146, 324-6, 331 Chart (see Diagram)
characteristics, 85, 88, 91-2 Checking (see also Design agent, computer-aided; Model;
environmental connections, q 1 Work study), 171, 242
fate, 89 · Check-list (see also Methodical procedure), 7, 24, 111,
life-cycle (see also Life-cycle), 26, 252-3, 288 115, 133, 312
situation adaptability, 299-300 Chemical engineer, engineering (see also Technology),
ArtiCidation (see also Assembly; System), 237 5-6, 77, 138, 151, 153, 158-61, 202, 207, 213-4, 239,
Artist (see also Industrial designer; Musician; Painter; 242-3, 313-4, 3~3. 325, 331
Poet; Sculptor), 4, 7, 20, 37-8, 41, 112-3, 'rl9, 283, Choice (see also Decision; Selection; Value), 39-40, 81,
295, 302 88, 121-Q, 331
347
Circuit (see also Electronics engineer; Systehl), 319 10, 24, 49, 52, 54, 61, 62, 77-8, 82-3, 85-6, 121,
Civil engineer, engineering (see also Technology), 4, 7, 125, 129, 131-40, 176, 186-7, 193, 199, 249, 269,
201, 297. 313, 323, 325-6 272, 275, 289-91, 296, 308, 322, 328-31
Client (see also Customer; Sponsor), 59 Definitions, 19-23, 295-309, 331-3
Colour (see also Perception), 4, 35, 37, 31), 42, 53, 113 decision, 132, 149, 331
Commissioning, 89, 131, 21)0 design, 3-6, 11, 19, 25, 85, 197, 259, 295-6, 320, 324-9,
Committee (see Management science; Team) 333
Communication (see also Information; Model; Transporta- morphological, 91-3, 332
tion), 5, 8-10, 21), 56, 59-60, 64, 69, 85, 145-8, 153, need, 67-8
241, 245, 256, 259, 279-85, 322, 328, 330 policy, 253, 332
Company (see also Managemenf science; Organization), 5, problem-solving, 6, 145
9, 15, 136, 138, 157-R, 184, 197, 241), 253, 258, reliability, 219
266-7. 290-1, 330 responsibility, 10-1, 149
Comparison (see also F.valuation; Ranking), 17 science, scientific method, 6, 61, 131, 149, 295, 324
Compatibility (see also Interaction; Reconciliation), 28, technology, 19-20, 61, 239, 324-6
53-4, 86, 103, 132, 191, 298, 306, 308 Dependency (see Hierarchy; lnteraction; Network)
Complaints (see also Feedback; Service), 21-2, 21), 66-7, Design
219, 231, 250 activity
Complexity (see also Network; System), 5-6, 40, 155, 201, analysis (see Analysis)
207, 220-1, 228, 290, 21)6, 298, 317 appearance (see also Appearance), 3
Component (see also System), 21, 88, 159, 202, 204, 221, approach, 20, 35-6, 88-9
314, 318-9 book-keeping, 88, 186
Computer (see also Design agent, computer-aided; Model), communication (see also f:ommunication), 85, 88,
6, 99, 102, 105-6, 151-2, 183, 233, 300, 327 241-6, 256, 281-5, 330
Concept (see also Design activity; Incubation; Psycho- concept ( see Concept)
logy), 4, 12-4, 16-R, 28, 66-7, 71, 78-CJ, 85-6, 91, convergence (see Convergence)
97, 107, 144, 148, 154, 184, 198-1), 214-5, 295, 302, critical mistakes, 14
306, 318, 320-2, 330-1 decision (see also Decision), 85, 131-40, 296
Connectivity (see also Network; Pattern), 80, 105-7 definition (see Definitions)
Constraint (see also lnteraction; ()ptimization; Restraint), divergence (see Divergence)
20, 77, 91, 1)4, 131-2, 237, 277, 288, 327-9, 331 evaluation (see Evaluation)
Consumer (see also Need; Satisfaction; User), 7-8, 10, 29, heuristic (see Heuristic)
35, 39, 48, 65-73, 81, 326 incentive (see Incentive)
goods (see also Design product), 36, 31) intellectual demands, 11, 18, 330
Containment (see also Design product; Function), 202, method (see Heuristic; Methodical procedure)
236, 326-7 motivation (see Motivation)
Control (see also Cybernetics; Management science; optimization (see Optimization)
Objective; Optimization), 5, 17, 30, 42, 53, 21Q, process, 3-4, 12, 21, 78-CJ, RS-95, 131, 302, 327-9
255, 258, 276, 287 science, 323-30
Convergence, convergent thinking (see also Divergence; strategy ( see Strategy)
Psychology), 7, 79, 296, 287 structure, 303
Cost (see also Economics; Product profile; Va1ue; Value study, 6, 133, 199
analysis), 11, 14, 20, 30, 37, 47-8, 56, 91, 132, 167, synthesis, (see Synthesis)
170, 173, 179, 198, 204, 206, 211-2, 214, 216, 220, varieties, S-6, 132, 134, 138-CJ, 158, 1CJ9, 202-3,
236, 249-50, 253-5, 279, 287, 298, 319-20 287-8, 2Q7-CJ, 311-13, 317-20, 327
Craftsmanship (see also Practice; Production engineer), agent
4, 204, 206, 261, 303 attributes, 18, 183, 261, 301
Creativity (see also 'Brainstorming'; Jleuristic; Incubation; computer-aided, 47, 7CJ-80, 112, llR, 133, 143, 147,
Psychology), 9, 23, 79-81, 98, 103, 108, 111-9, 125, 153, 167-73, 187-8, 207, 27'l, 28'1, 302, :107-8,
135, 137, 140, 154, 187, 259, 273-4, 301-2, 304, 327-8, 330
322, 328, 330-2 control (see also Contra!), 256-8, 276
Critical path (see also Network; Technical planning), involvement (see also Motivation), 264
15-6, 207, 241, 28CJ management (see also Management science), 249-85,
Critique (see also Evaluation; Judgment), 29, 61, 63, 322 288-9, 308-9, 330
Cubism, 113 nucleus (see also Team), 321
Customer (see also Client; Market; User), 7, 26, 36, 48, policy (see Policy)
69-73, 197, 219-21, 230, 250-1, 258, 330 responsibility (see also Responsibility), 11, 19, 40,
Cybernetics (see also Control; Management science; 132, 272
System), 295, 302 role, 21
specialist, 298
Decision (see also Certainty; Choice; Design activity; team (see Team)
Management science; Responsibility; Uncertainty), training, 183-5, 255-6, 261, 272
SUBJECT INDEX 349
pmduct (see also Artefact; Product) hospitals, 59, 62-4

general houses, 87, 201, 295, 300, 326
acceptsnce, 12, 65-73, 299-300 hydraulic machines, 151, 313
adaptability, 299-300 insulation, 162
appearance (see Aesthetics; Appearance) jigs and tools, 206, 312
batch, 205, 236 laboratories, 59
conventional, 5, 158 machines, 202-3, 259, 312
detail, 260 machine-tools, 5, 21, 56, 312, 326
development, 204 man-machine systems, 49-57, 298, 304
embodiment, 5, 325-6 materials, 235-9
evolutionary, 5, 15~. 162-4 mechanisms, 312
exploratory, 2Q1 mine-car couplers, 185-98
flow, 205 nuclear plant, 162, 201, 214, 298, 314
function, 5, 9, 14, 21-2, 37, 201, 236, 312, 314, petroleum refineries, 3, 201
326, 331 pilot plsnts, 150
improving, 216, 260 pipework, 243-4
innovative, 9-10, 24, 88, 133-4, 259-60 power plants, 6, 202, 318
'nouveau', 298, 300 pressure vessels, 161, 206, 314
one-off prime movers, 202, 255, 312
complex, >fi, 119, 206-7, 235-6, 21!8, 297, 32Q prosthetic devices, 298
simple, 206 railways, 220, 300
open-ended (see also Standard), 236, 297, 299, razors, 305
317, 320 refrigerators, 204, 206
pattern, 105 regions, 37, 295, 297, 302
pioneering (see also Sen.sitivity snalysis), 236 scientific instruments, 87, 90-2
process (see also Function), 326-7 shafts, 169, 202
radical, fi, 24, 134, 137, 158, 11i4-5, 285 ships, 203
realization, 9 structures, 203, 326
response (see Acceptance) supports, 202, 236, 326
routine, 260 telecommunication systems, 203, 318-9, 325
speci al purpose, 236, 299 tools
system (see also System), 134-5, 323 hand, 116
specific powered, 312
aerospace systems, 202-3, 211-2, 214, 296, 299- transistor radios, 318
300, 305 transportation, 21, 46
automobil!!s, 5, 21, 40, 56, 92, 157, 201, 203-4, tyres, 239
211, 220-1, 228, 280 weapons, 289
bearings, 151, 202, 223 relationship
boilers, 5 aesthetics (see also Aesthetics), 37-8, 42-3
bumers, 17>8 anatomy (see also Anatomy), 51
capital equipment, 20-1, 35, 40, 43, 201, 207, 255 anthropometry (see also Anthropometry), 51
cathode ray tube, 164 assembly (see also Assembly), 201, 289
chemical plant, 5, 151, 153, 159-61, 201, 207, customer (see also Customer), 36, 41-3, 67-8, 73
213-4, 239, 241-2, 313-4, 325 economics (see also Economics), 157-65
clothing, 326 environment (see also Environment), 135, 158, 269-
compressors, 206 77
computers, 6, 105, 300 ergonomics (see also Ergonomics), 52-6
consumer goods, 6, 43, 280, 326 fabrication (see also Fabrication), 204-7, 214
contact members, 202-3 information (see also Information}, 97-9, 136-7
COntainment, 202, 236, 32>7 innovation (see also Innovation), 67-73, 299, 332
domestic equipment, 5, 25, 27-8, 56, 93, 211, market (see also Market), 36, 250-1
280, 297, 312-3, 326 materials (see also Materials), 202, 211-6, 236-7, 330
electrical machines, 167-73, 202-4, 223, 253-4, physiology (see also Physiology), 35, 49
313 production ( see Assembly; Fabrication)
electronic equipment, 204, 313, 318-22 psychology (see also Creativity; Design activity;
envi ronments, 300 Ergonomics; Motivation; Psychology}, 330
farm wagons, 296 science (see also Science), 6, 12, 14, 19, 61, 111-2,
food, 326 143, 149, 323-4, 329-30
furnaces, 239 situation, 3, 38, 59, 61, 77, 89, 296-7, ~3. 329, 333
fumiture, 2S!, 41, 51, 116, 300, 326 sociology (see also Sociology), 59
gears, 151, 202-3, 207-10 symbolism, 37-8, 43
heat exchangers, 162, 203 Designer (see Desi&n agent)
Determinist (see also Creativity; Psychology), 114, 291 Function (see Design product)
Development, 24, 255, 281, 290 Functional module, 319
Device (see also System), 319
Diagram (see also Chart; Communication; Model), 1!, 28- Gantt chart (see also Critical path), 289
30, 144-6, 242-3, 318 General systems theory (see also Model; System), 77-8,
Discountlog (see also Economics), 137, 159 88, 152-4, 331
Discriminatory ability (see also Customer; Evaluation; Gestalt (see also Psychology), 42, 114
Judgment; Perception), 40 Goal (see also Objective; System), 52, 88, 122-4, 132,
Di vergence, divergent thinking ( see also Concept; Con- 272, 306
vergence, Methodical procedure), 7, 79, 91-3, 304-E · Golden section (see also Aesthetics; Shape), 37
Drawing (see also Communication; Model), 4-5, 133, 147, Graph (see aloo Model), 105
225-6, 241-6, 253, 259-60, 295, 312, 318 theory (see also Network), 80, 105, 331
office, 147, 201, 204, 241-6, 320 Graphics (see aloo Artist), 295
Guarantee (see also Warranty), 221
Economics (see also Behavioural sciences; Cost; Dis- Gyroscape (see also Model), 270
counting; Investment; Licensing; Market; Market
research; Objective; Optimization; Profit; Resources; Hardware (see also Embodiment; Realization), 5, 22-3,
Utility; Value; Value analysis), 36, 66, 91, 131, 133, 241. 325-6
133, 147, 157-65, 173, 204-5, 211, 215, 235, 290, Heuristic (see also Creativity; Search), 23-6, 82, 111,
330 133, 153, 291-2, 323, 330-2
Electrical engineer, engineering (see also Technology), Hierarchy (see also System), 80, 105, 107, 133, 149, 318,
4-6, 167-73, 202-4, 253-4, 313, 325, 331 320, 327, 329, 331
Electronics engineer, engin~ing (see also Electrical Human
engineer; Technology), 313, 317-22, 323 engineering (see also Ergonomics), SO
Element (see also System), 14, 21, 29-30, 314 factors (see also Ergonomics), SO
Embodiment (see also Fabrication; Mechanical engineering; Iconic (see also Model), 77, 85, 332
Production engineering; Realization), 5, 85, 88, 133, ldentification (see also Need), 97
251, 325-6 Imagination (see also Creativy; Psychology), 18, 23, 140,
End (see Objective) 185, 261, 304
Engineer (see Design agent; technologist) Incentive (see also Motivation; Objective), 249, 263
Enterprise (see also Management science), 249 Incubation (see also Concept; Creativity), 20, 28, 112,
Environment (see also Containment design; Design re- 154, 302, 312
lationship; Er&Qnomics; System; Test), 91 Industrial
Equipment (see also System), 318-20 design, designer (see also Artist), 20, 41, 283, 295, 317
Ergonomics (see also Behavioural sciences; Environment; dynamics (see also Project), 288
Man-machine; Physiology; Psychology; System), 6, Information ( see also Certainty; Communication; Computer;
31, 41-2, 49-57, 255, 270, 284, 298, 312, 331 Decision; Design; Model; Science; Search; Un-
Ethics (see also Product profile; Sociology; Value), 132 certainty), 52, 56, 64, 78-80, 88, 97-103, 136-7, 145-
Evaluation (see also Choice; Comparison; Critique; 9, 157-8, 241, 248, 322, 330
Feasibility; Judgment; Justification; Match; Optimi- theory (see aloo Feedback; Management science; Model;
zation; Ranking; Selection; Test; Value; Verifi- Redundancy; Uncertainty), 37-8, 145, 148, 281-3,
cation), 15, 17, 22, 24, 28-9, 47, 66, 78, 85, 94-5, 327
102, 198, 202, 302, 314 Innovation (see also Design product), 24, 65, 70-1, 111,
Exploration (see also Search), 79 134, 249-50, 252, 299, 302, 308, 331
Export (see also Market), 249-50 Inspection (see also Checking; Quality), 225
Externalization (see also Communication; Critique; Model), Installation (see also Assembly), 255
4, 131, 206, 304 Intention {see also Objective), 64
Interaction {see also Compatibility; Constraint; Depen-
Fabrication (see also Embodiment; Production engineering; dency; Network; Optimization; Reconciliation;
Realization), 88, 201, 203-7 Sociology; System), 19, 28, 31, 53-4, 64, 78, 89,
Facet ( see also Information), 99 92, 94-5, 162, 296, 300, 327, 333
Failure (see also Persistence; Success; Test), 15, 65-6, Interchangeability (see also Standard), 188
216, 231-3 Interconnection (see also Morphology: Network), 88
Fatigue (see also Materials), 213 Investment {see also Economics), 6, 17, 40, 47, 68, 72,
Feasibility (see also Evaluation), 7, 13-4, 70, 72, 85, 82-3, 159, 250, 266, 288, 290
134, 314, 323, 327 Involvement (see also Motivation), 81, 118, 135, 137,
Feedback (see also Communication; Complaints; Contra!; 262, 264, 267, 296
Cybernetics; Information theory; Iteration; Loop), 'lsolate' (see also System), 78, 80
21, 30, 56, 87, 157, 221, 282 Iteration (see also Loop), 23, 25-6, 87, 307, 322
Flexibility (see also Strategy), 66, 255
Forecasting (see also Market), 36, 61-2, 64, 249-52, 281!, Judgment (see also Critique; Evaluation; Subjective), 7,
290 11, 15, 17, 29, 79, 164, 185, 291, 322
Form (see also Shape), 37, 42-3 Justification (see also Evaluation), 63, 128-9, 332
SUBJECT INDEX 351
Kinaesthetic image (see also Perception), 302 Microminiaturization (see also Fabrication), 204
Model (see also Analogue; Communication; Computer;
Layout (see also Arrangement), 243 Diagram; Drawing; General systems theory; Gyro-
Licensing (see also Economics), 47, 136-7, 251-2 scope; Graph; Iconic; Information; Information theory;
Life-cycle (see also Product; Project), 19, 26, 89, 220, Management science; Mathematics; Optimization;
231, 252-3, 287-8, 299-30, 303 Paradigm; Pilot plant; Problem-solving; Prototype),
Loop (see also Feedback; Iteration), 12, 14, 21. 25, 30-1. 143-55, 330
302-3, 307, 315 abstract, 146-7
analogue, 4, 8-9, 152
Machine-tools (see Design product; Fabrication) autonomic designer, 77, ll5, 152, 328
Maintenance (see also Product profile), 29, 95, 220, 255 checking, 143
Man-machine (see also Ergonomics; System), 6, 37, 42, communication, 8, 88, 145-7, 153, 242, 283-4, 321, 330
so. 111, 135, 137, 298, 304, 330 computer, 152-3, 327
Management science (see also Behavioural sciences; creativity, 81, 108, 112, 118-9, 154, 302, 328
Committee; Communication; Company; Contra!; decision, 83, 134, 137, 292, 330
Decision-making; Oesign agent; Enterprise; Infor- design activity, 8, 12, 78, 184, 186, 270, 292, 320-1,
mation theory; Involvement; Model; ~loti vation; 327-9
Objecti ve; Operational research; Opportunity; Opti- evaluation, 29, 143, 195, 330
mization; Organization; Policy; Problem-solving; iconic, 85, 152
Procedure; Project, Responsibility; Satisfaction; information processing, 81, 145, 148-9, 328
Team; Training; Value engineering; Work study), 9, logical, 147-8
131, 136, 143, 249, 253, 263, 288-92, 301, 323, 330 management, 143, 146
Market (see also Advertising; Challenge; Client; Customer; market, 36
Economics; Forecasting; Need; Opportunity; Possi- material, 146, 149-50
bility), 7, 12, 15-6, 22, 36, 45-8, 87, 137, 157, 250-1, mathematical, l3, 87, 143, 146-7, 149-51
258, 27fi, 301, 332-3 problem-so! ving, 9, 118-9, 133, 145-6, 242, 2ll3-4, 303
Match (see also Evaluation), 132, 303 science, 143
Materials ( see also Design 1elationship; Embodiment; semantic, 146
Fabrication; Selection), 9, 13, 20, 42, 201, 203-4, stimulus-response, 80-1
211-7, 235, 319, 330 strategy, 154-5
technologist, technology (see also Technology), 211, symbolic, 146
217, 239 test (see also Pilot plant; Prototype; Test), 306
MathemaUes (see also Computer; Decision; Graph theory; theory, 143
Model; Operational research; Optimization; Problem- topological, '323
solving; Set theory; Statistics), 13, 20-1, 29, 36, training, 184, 256
82, 94-5, 132-3, 135, 143, 150-1, 153, 330 transformation, 147-9
Meaning (see also Information theory; Semantics), 80, 107 work study, 253
Mechanical engineer, engineering (see also Technology), Module (see also Electronics engineer; System), 319
5, 202-3, 311-3, 325 Morphology, morphological (see also Interconnection;
Mechanism (see Design product) Methodical procedure; Network; System), ;!8, 78-9,
Memory (see also Psychology), 7, 79, 112, 119, 152, 302, 89, 91-3, 97, 100, 108, 134, 189-90, 194, 323, 332
308 associati ve, localized or non-localized, 92
Metallurgist, metallurgy (see also Technology), 4, 216, 235 interaction (see also lnteraction), 327
Methodical procedure, methodology (see also Algorithm; sequerice, fixed or variable
Creativity; nesign activity; Heuristic; Problem- threadlike,
solving; Procedure; Strategy), 6-9, 11-2, 153, 155, multiple, 327
202, 302, 329-30, 332 single, 92, 327
ATOM (analysis, theorizing, delineating, modifying Motivation (see also Aspiration; Attitude; Challenge;
cycle), 25-6 Incentive; Involvement; Management science; Need-
check-list, 23-5, 311 fuUilment; Objective; Psychology; Sociology), 37,
cri tical examination, 7ll 51, 53, 67-8, 83, 118-9, 157-8, 269-73; 321, 330
design tree, 23-5, 79, 311 Musician ( see also Artist), 113
experience, 23-4, 303, 311
FDM (fundamental design method), 77, 183-99, 306 National Plan, 6
modification, 23-4, 304, 306, 311 Need (see also Brief; Consumer; Identification; Market;
morphological, 78-9, 91-3, 97, 304, 306 Opportunity; Product profile; Satisfaction; Sponsor;
PABLA (problem analysis by logical approach), 77, User), 12, 22, 35, 37, 39, 41-2, 45, 59, 63-4, 66-72,
89, 97, 199 85-7, 89, 102, 188, 222-3, 300-1, 328-9
precedent, 26, 134 Need-fulfilment (see also Motivation), 67-8, 329
system search, 23, 29-31, 79, 306, 311, 327 Network (see also Activity; Anastomosis; Bridging; Com-
systematic, 23, 26-9, 97, 155, 304-6 plexity; Connectivity; Critical path; Dependency;
trial and error, 66, 72, 303 Graph theory; Interaction; Morphology; Node; Recur-
work study, 77-8 sive; System; Tree), 80, 10f,-8, 151, 288-90, 332
Node (see also Network), 94, 105-9, 289 Policy (see also Management science; Objective), 9-10,
'Noise' (see also Communication; Infonnation theory), 78, 15, 67, 139, 157, 215, 249, 253, 262, 321, 332
2ll2 Possibility (see also Opportunity), 12, 36-7, 183, 265,
Numerical control (see also Contra!; Drawing; Machine- 290-1
tools), 5, 147, 206, 312 Practice (see also Communication; Craftsmanship), 136,
204, 242
Precedent (see also Case history; Methodical procedure),
Objective (see also Aim; Company; Design agent; F.nd;
26, 134, 136, 138, 332
Goal; Intention; Perfonnance; Policy; Product
Problem-finding (see also Need; Perception), 296, 303
profile; System; Target), 11, 15-6, 38, 48, 52, 64,
Problem-solving (see also Algorithm; Analogue; Creativity;
83, 132, 183-4, 250, 253, 264, 276, 301, 317, 320, Decision; Evaluation; Heuristic; Methodical proce-
332 dure; Model; Optimization; Search), 3, 5-7, 9, 22-3,
function (see also Economics; Optimization), 159, 290 59, 79, 111, 119, 145-53, 185, 198, 264, 271, 296,
Observation (see also Perception), 39, 261 314, 317
One-off (see also Design product), S-6, 139, 206-7, 235-6, Procerlure (see also Management science; Organization),
288, 297, 329 107, 264, 269-77, 291
Open-ended (see also Design product; Specification), 317, Product (see also Artefact; Design product), 3, 5, 36,
320 40-4, 332
Operational research, 111, 131, 139, 143, 291, 312 profile (see also Aesthetics; Commissioning; Cost;
Operator (see also Ergonomics), 49-57, 111, 220, 255-6,
Ethics; Ergonomics; Function; Investment; Main-
291 tenance; Need; Objective; Perfonnance; Quality;
Opportunity (see also Challenge; Market; Need; Possi-
Reliability; Safety; Service; Symbolism; Value),
bility; Resources; State-of-the-art), 7-10, 15, 17-8,
5, 16-7, 36, 132, 280, 288, 291
35-6, 38, 82-3, 136, 250-2, 289, 329, 332-3
Production engineer, engineering (see also Assembly,
Optimization (see also Constraint; Heuristic; Mathematics;
Fabrication; Materials; Technology), 20, 22, 201,
Model; Obj ecti ve function; Perturbation; Restraint;
205-7, 282, 284-5, 321-2, 325, 330
Sensitivity analysis; Sub-optimization), 17-8, 30,
Profit (see also Economics; Objective function), 47-8, 85,
41, so, 52, 57, 82, 85-7, 94-5, 97, 111, 132, 134,
159-61, 164-5, 252, 259, 266
146-7, 151, 153, 159, 167, 173, 197, 205, 207, 239, Project (see also Design product, one-off, complex;
258-9, 290, 296, 298, 300, 303, 312, 315, 332 Industrial dynamics; Management science), 5, 19,
Orgwüzation (see also Communication; Company; Manage- 138, 207, 288-9, 291-2
ment science; Procedure; System; Team), 5, 9-10, Prototype (see also Model; Test), 22, 136, 155, 188, 205,
67, 83, 131, 135-9, 259, 264, 269-77, 279, 284-5, 224, 285, 303
288, 291-2, 301, 308, 321-2 Psychology (see also Aesthetics; Associationist; Be-
Overhaul (see also Maintenance; Service), 289 havioural sciences; Concept; Convergent thinking;
Overlap (see also Loop), 23, 26 Creativity; Decision; Divergent thinking; Ergono-
mics; Gestalt; Imagination; Information; Involvement;
PABLA (see Methodical procedure) Judgment; Memory; Motivation; Need; Need-fulfil-
Painter, painting (see also Artist), 4, 112-3, 146, 279, 295 ment; Perception; Personality; Satisfaction; Symbo-
Paradigm (see also Model), 323, 332 lism; Uncertainty; Value), 24, 37, 51, 59, 65, 125,
Part (see also System), 29, 296-7, 319 145, 272, 282, 284, 295, 322, 330
Pattern (see also Connectivity; Creativity; Model; Product; Purchaser (see also Customer), 39
Shape), 3-4, 80, 105-9
Perception (see also <;:olour; Communication; Ergonomics; Quality (see also Product profile; Reliability), 45, 219,
Kinaesthetic image; Observation; Psychology; 225, 296, 298, 319
Recognition), 7, 11, 37-8, 52, 78-9, 131, 145, 148, Quantification (see Science)
264 Queueing theory (see also Operational research), 312
Perfonnance (see also Design product, function), 14, 28,
211, 220, 226, 296, 305-6 Ranking (see also Comparison; Evaluation), 24, 27
Persistence (see also Failure), 202, 236, 326 Realization (see also Assembly; Commissioning; Drawing;
Person-product (see Man-machine) Embodiment; Fabrication; Hardware; Production
Personality (see also Psychology), 12, 40, 67, 118, 137, engineer), 4, 9, 20, 23, 63, 108, 201-7, 289, 321, 332
183, 272, 279 Recognition (see also Perception), 12
Perturbation (see also Heuristic; Optimization; Synthesis), Reconciliation (see also Balance; Compatibility; Design
30, 133, 149, 323 activity; lnteraction; Optimization), 86, 88, 296-7,
Philosophy (see also Aesthetics; Choice; Value), 37, 322
121-9, 330 ~ecursi ve (see also Network; Pattern), 105, 107
Physiology (see also Ergonomics), 35, 51 Reductionist (see also Science; Technology), 151
Piece (see System) Redundancy (sec also Information theory), 141-5, 148,
Pilot plant (see also Model), 134, 136, 150, 155 282, 285
Planner, planning (see also Technologist), 62, 82, 295-6, Relationship (see also Morphology; Network; Pattern;
314 Shape), 91, 105
Poet (see also Artist), 3-4 Relativity theory, 113
SUBJECT INDEX 353
ReHability (see also Product profi!e; Statistics; System), Statistics (see also Mathematics; Reliabi!ity), 173, 225-7
14, 94.,201, 206, 219-33, 298, 319 Status (see also Certainty; Design agent; Sociology;
Requirement (see also Need), 59 Symbolism), 39, 43, 111, 136-7, 261-3, 271-5, 277,
Research (see also Information; Science), 16, 66, 136, 251 321
topics, 119, 239, 267, 277, 299-300, 309, 320-2, 329-30 Strategy (see also Heuristic; Methodical procedure; Ob-
Resources (see also Capability; Economics; Information; jective; Policy; Problem-solving), 8, 36, 77, 88,
Models; Technology), 5, 9, 36-7, 88, 97, 119, 149, 93-5, 134, 137. 143, 152, 154-5, 164, 168, 175, 188,
251, 276, 289, 291, 295, 303, 320 276, 291, 301, 303, 306-7, 311, 327, 329, 332-3
Response (see also Acceptance), 300, 304 choice
Responsibility (see also Decision; Management science), a1 ternati ves tree, 94-5
10-1. 1Q, 40, 132, 136, 140, 149, 292 objectives, 301
Restraint (see also Constraint). 16, 320 sub-problem, 134
Risk (see also Uncertainty), 67, 135 company, 36
Role (see also Sociology), 21, 37, 60, 292 defini tion, 333
Rule ( see also Problem-sol ving; Procedure), 138 design
activity, computer-aided, 168
Safety (see also Product profile), 56, 89, 132, 299, 319 maximum manoeuvre (ending), 134, 306-7
Satisfaction (see also Consumer; Customer; Need; Psycho- means-ends analysis, 155, 307
logy), 3, 7, 35-7, 41, 48, 301, 325, 333 minimum commitment (beginning), 81!, 134, 154-5,
Science, scientific method, sdentist, 6, 12, 19-20, 61, 296, 307, 311, 315
134, 143, 149, 295, 299, 323-4 objective, 301
Sculptor, sculpture (see also Artist), 4, 20, 146 product
Search (see also Exploration; Heuristic; ~ethodical pro- flexibility, 255
cedure), 6, 25, 29, 78, 202, 216, 301, 306, 332 simplici ty, 253
Selection (see also Choice; Evaluation; \faterials), 93, uncertainty, 77
175, 211-7 reduction of alternatives, 94, 137
Sel!ing methods (see also Market), 252 sub-optimization (fragmentation), Q4, 139, 161
Semantic (see also Communication; Information; Meaning; economics and design, 155, 164
Model), 7, 99 general problem-solving, 152, 155
Sensitivity analysis (see also Optimization), 86, 135, 138 'in-out', 'out-in' (see also Bridging; Tree), 307
Service (see also Complaints; lßaintenance); 163-4, 173, loop-breaking, 303
231-3, 250 management, 276
Set theory (see also Mathematics), 77, 93, 333 model usage, 143, 152, 154-5, 306
Shape (see also Aesthetics; Design product, general, practical example, 175, 188
function; Fabrication), 4-5, 35, 39-42, 114, 153, research, 291
235, 279, 296, 308, 325, 327 on design, 317, 327, 329
Shareholder (see also Company; Management science; Style (see also Aesthetics; Model), 42
Profit), 266 Subjective (see also Judgment), 17, 82, 135-6, 139, 164,
Shelter (see also Containment; Design product, specific, 326
houses; Need-fulfilment), 45 Sub-optimization (see also Optimization; Strategy; System),
Side-effects (see also Acceptance; ~esponse), 299 94, 139, 158-9, 161
Sifting (see also Evaluation; Search), 22, 78, 253 Success {see also Acceptance; Failure), 299
Simi!arity (see also Model), 149 9Jpport (see also Design product, general, function), 202,
Simplification (see also Work study), 2Q, 191, 242, 253 236, 326
Simulation (see also Model; Training), 29 Symbolism (see also Aesthetics; Psychology), 7, 37, 40,
Situation (see Design relationship, situation) 43, 295
Sociology (see also Behavioural sciences; Communication; Synectics {see also Creativity; Heuristic), 116-7, 154, 328
Ethics; Need; Image; Innovation; Interaction; ~oti Synthesis {see also Design activity; Heuristic; Methodical
vation; ,Organization; Procedure; Role; Situation; procedure), 12-3, 78, 85, 97-8, 102-3, 115, 167-9,
Status; Team), 37, 59-64, 276-7, 285, 306, 330 202, 302, 312, 323
Specification (see also Communication; Model; Open-ended), System (see also Artefact; Articulation; Assembly; 'Bleck
22-3, 28, 66, 72, 91, 241, 302, 306, 326 box'; Characteristic; Circuit; Complexity; Component;
Sponsor (see also Brief; Client; Need), 21, 27, 71, 300-2, Cybernetics; Design; Device; Element; Environment;
314 Equipment; Ergonomics; Function; General systems
Standard, standardization (see also British Standard Speci- theory; llierarchy; Interaction; 'Isolate'; ~an
fication; Design agent, computer-aided; Design machine; Management science; Methodical procedure;
product, general, open-ended; Work study), 5-6, 29, Model; Module; Morphology; Network; Objective;
39-40, 43-4, 53, 173, 204-7, 235, 241, 253, 255, One-off; Optimization; Part; Piece; Reliability;
298-9 Sub-optimization; Transfer function), 5, 7, 21, 52,
State-of-the-art (see also Technology), 12-3, 20, 40, 71, 55, 78-9, 111, 134, 145, 154-5, 201, 207, 236, 253,
85, 81!, 119, 136, 236 325, 330, 333
Statistical decision theory (see also Decision; Operational autonomic designet (see also Autonomie), 88
research), 135 Boulding degree, 148-Q, 152
System (Continued) Transfer function (see also System), 21, 327

closed, 78, 88, 131, 134-5 TransportaUon (see also Communication), 15, 21, 45-7
c:omponent-system interaction, 175, 189-97, 202 Tree (see also Network), 16, 25-6, 30, 79, 94-5, 133
definition, 21, 78, 201, 333 Trial (see also Test), 229-30
flow, 297-8, 323, 332 Twelve-note method in music, 113
input-output, 78, 327 Two-culture situations (see also Communication; Team),
interc:onnection, 78 59-60, 64
mechanical, 201-2, 297
open, 78, 82, 131, 135 Uncertainty (see also Decision; Information), 77, 111,
relation, 88 133-8, 149, 289, 291, 295-6, 324
supersystem, 318 Unsteady-state (see also Life-cycle), 327
transfer, transformation, 326-7 Uprating potential (see also Flexibility; Strategy, choice,
Systematic procedure (see also Methodical procedure), 7 sub-problem), 255
Systems engineering. 7, 323, 325 User (see also Customer; Man-machine; Need; Operator),
7, 26, 31, 39, 221, 230, 321, 330
Target (see also Objective), 8, 276 Utility (see also Economics), 82, 131, 138
Task (see also Design activity; Project), 16
Team (see also Committee; Design agent; Management Value (see also Action; Choice; Costs; Economics; Ethics;
science; Sociology), 21, 26, 29, 59, 63-4, 70-2, 85, Evaluation), 3, 6, 41, 48, 64, 81-2, 121, 124, 128,
133-5, 157, 215, 260, 267, 273, 275-6, 281, 292, 134-5, 328-9
308-9, 328 analysis, engineering, 9, 14, 215-6, 253
Technical planning (see also Critical path), 11, 15-6, 292 Variety reduction (see also Standard; Work study), 173,
Technologist, technology (see also Architect; Builder; 253, 298
Chemical engineer; Ci vil engineer; Electrical en- Verbalizer, 7, 81
gineer; Electronics engineer; Materials technologist; Verification (see also Design activity; Evaluation; Feasi-
Mechanical engineer; Metallurgist; Planner; Prod- bility; Test), 20, 28, 112
uction engineer; State-of-the-a'rt), 4, 7, 11, ·13, 15-6, Vibration, 90, 149, 203, 223
31, 42, 44, 60-1, 66, 117, 136, 146, 239, 249, 259, Visualization, 7, 81
269, 284, 292, 324-5
Test (see also Critique; Evaluation; Pilot plant; Proto- Warranty (see also Complaints; Guarantee; Service), 231-2,
type; Specification; Trial), 14, 29, 54, 61, 66, 78, 255
86, 132, 136-7, 224, 226-9, 249, 298, 330 Work study (see also Activity sampling; Checking; Com-
Tolerance (see also Assembly; Reliability; Specification), munication; 'Aethodical procedure; Model; Redun-
87, 225-6 dancy; Simplification; Standard; Variety reduction),
Trade-off (see also Reconciliation), 14, 139 6, 7, 9, 26, 28, 77, 191-4, 242, 276-7, 287, 291
Training, 67, 183-5, 256, 266-7, 287 Workspace (see Erl'>nomics; Man-machine)

Gregory SA Ed The Design Method

Uploaded by

Copyright:

Available Formats

Gregory SA Ed The Design Method

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Gregory SA Ed The Design Method

Uploaded by

Copyright:

Available Formats

THE

Springer Science+Business Media, LLC

Suggested U.D.C. number: 601

Part I. The Design Method

1. Design and the Design 1VIethod

2. The Design Method in Practice

3 Definitions and Methodologies

Part II. The Human Perspective

4. The Human Perspective, the Design Situation and its Opportunities

5. Design and Human Satisfaction

6. Invention, Design and Market Research

8. Sociology and Design

Part 111. The Elements of Design

10. A More Detailed View of Design

12. Infonnation for Design

15. Choosing and Evaluating

16. Design and Decision

Part IV. Design Techniques

17. Models in Practical Design

18. The Implications of Economics in Engineering Design

19. The Use of a Digital Computer in Design Offices

20. A Practical Design: An Oil Bumer for Large Watertube Boilers

21. Practical Design Basedon Method (Fundamental Design Method)

22. Design Realization

23. Selection of Materials

24. ReHability and Maintenance

25. Design of Materials

26. New Ideas in the Drawing Office

Part V. Management and Design

'1:7. Design Policy Fonnutation

28. Innovative Design as a Policy Function

29. Effect of Organizational Procedures on Design - An Outline of the Problems

31. Design, Management, Realization and Change

Part VI. Design Research

32. Design Methods Reviewed

33. Technologies and Varieties of Design

34. Preliminary Research into Electronics Design

35. Design Science

References and Bibliography 335

Name Index 343

Subject Index 347

H. V. BECK, M.A., B.Sc., C.Eng., M.I.E.E.

A.H. BRIGGS, B.Eng. (Hons)

GEOFFREY H. BROADBENT, B.A., A.R.I.B.A.

C. HEARN BUCK, B.Sc., A.lnst.P., C.Eng., A.M.Inst.Gas E.

C.T. CORNEY, B.Sc., C.Eng., A.M.I.E.E., A.R.C.S.

A.L. DAVIES, B.Sc., Ph.D., A.I.M.

ROBERT DAVIS, C.Eng., A.M.I.Mech.E., A.M.I.Prod.E.

W.E. EDER, lng. (Vienna)

F.L. ELLIS, C.Eng., A.M.I.Mech.E., A.M.Inst.R.

J. FARRADANE, B.Sc., A.R.C.S., A.R.I.C., F.I.Inf.Sc.

S.A. GREGORY, B.Sc., C.Eng., M.I.Chem.E.

R.J. MCCRORY, Mem.A.S.M.E.

P. MCMULLEN, A.C.G.I., B.Sc., D.I.C.

E. MATCHETT, C.Eng., A.M.I.Mech.E., A.M.I.Prod.E.

W.H. MAYALL, A.F.R.Ae.S.

A.M. NEEDHAM, B.Sc. (Eng.)

A.D. NEWMAN, B.Sc., A.Inst.P., C.Eng., A.M.I.E.E.

KENNETH CHARLES PARTON, B.Sc.(Eng.), C.Eng., A.M.I.E.E.