Product Realization

Product Realization:
A Comprehensive Approach
Mileta M. Tomovic l
Shaoping Wang
Editors
Product Realization:
A Comprehensive
Approach
13
Editors
Mileta M. Tomovic Shaoping Wang
Batten College of Engineering School of Automation Science
and Technology Electrical Engineering
Old Dominion University Beihang University
Norfolk, VA Hai Dan District
USA Beijing
mtomovic@odu.edu China
shaopingwang@vip.sina.com
ISBN: 978-0-387-09481-6 e-ISBN: 978-0-387-09482-3

DOI 10.1007/978-0-0387-09482-3
Library of Congress Control Number: 2008928441
# Springer ScienceBusiness Media, LLC 2009

All rights reserved. This work may not be translated or copied in whole or in part without the written
permission of the publisher (Springer ScienceBusiness Media, LLC, 233 Spring Street, New York,
NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in
connection with any form of information storage and retrieval, electronic adaptation, computer
software, or by similar or dissimilar methodology now known or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks and similar terms, even if they
are not identified as such, is not to be taken as an expression of opinion as to whether or not they are
subject to proprietary rights.
Printed on acid-free paper
springer.com
Preface
Comprehensive product realization, or Product Lifecycle Management (PLM),

is a holistic approach to sustainable product development from market analy-
sis, concept definition, design and analysis, production, customer service, all the
way to the products recycle. It is defined by a set of technological tools and a
change in business practices that will provide effective feedback mechanism to
support continuous and timely improvements in product development. It is a
methodology for the use of tools and technology enabled by a digital collabora-
tive environment, with the goal to improve competitiveness through the effec-
tive sharing and use and reuse of information.
PLM is highly applied area which has significant impact on industry and
provides numerous benefits to the adopters of technology and business prac-
tices, including: (i) increase in revenue through shorter time to market, (ii)
decrease in product costs through knowledge reuse, and (iii) decrease in product
development costs through effective feedback mechanisms. Some of the com-
monly used strategic metrics to measure PLMs return on investment indicate
significant benefits of using PLM. According to IBM PLM On Demand Busi-
ness report, companies are reporting 20% increase in design productivity and
5080% reduction in the time required to modify complex design, ability to
explore 50% more design options fostering innovation, conducting numeric
control programming up to 10 times faster and machining up to 35% faster,
60% reduction in pallet manufacturing time, 40% decrease in the errors found
at the final assembly stage, etc. Today when global competition is forcing
industry to reduce the cost of product development and manufacture, and
increase product options through mass customization, PLM is proving to be
an enabling technology that can gives manufacturers a competitive advantage.
The applications of PLM are expanding rapidly. According to CIMdata
Inc., the overall PLM market is expected to exceed $20 billion in 2008. The Next
Generation Manufacturing Report funded by the NSF, published by Jordan
and Michel, and based on opinions of close to 500 industry experts, envisions
that manufacturing industry of 2020 will become highly dependant on Product
and Process Lifecycle Management (PPLM) to integrate, connect and com-
bine people, processes, systems, and technologies of the extended enterprise to
v
vi Preface
assure that the right information is available at the right location, with the right
resources, at the right time.
Although highly applied PLM has numerous theoretical challenges that need
to be addressed and solved in order to take full advantage of the comprehensive
approach to product development. The major challenges are related to areas at
the boundaries between various well defined technical fields, as well as to
human aspects of implementation of the complex interdisciplinary and globally
diverse organizational structures.
The book presents some of the latest scientific findings and ideas along with
technical developments in the area of Product Lifecycle Management and
covers broad range of interdisciplinary topics ranging from measuring the
impact of PLM, social issues of PLM, product design optimization, PLM and
virtualization of product information, and multidisciplinary optimization. The
authors wish to thank all contributors for their contribution to the general body
of knowledge in this emerging interdisciplinary field, and for granting the
editors permission to use their material.
Old Dominion University, Norfolk, VA, USA Mileta M. Tomovic

Beihang University BeijingChina Shaoping Wang
Contents
Product Life Cycle and Assessment Parameters . . . . . . . . . . . . . . . . . . . . 1

Joze Duhovnik
Collaborative Optimization and Application of Active Suspension. . . . . . . 21

Hao Wang, Shaoping Wang, and Mileta M. Tomovic
Back to the Future: Product Lifecycle Management and the

Virtualization of Product Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Michael Grieves
Measuring the Impact of Product Lifecycle Management: Process Plan,

Waste Reduction and Innovations Conceptual Frameworks, and Logic
Model for Developing Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Cynthia Tomovic, Abram Walton, Lisa Ncube, Michael Grieves,
Ben Birtles, and Brandon Bednar
Reliability-Based Collaborative Design Platform for Hydraulic

Actuation System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Shaoping Wang, Hao Wang, Fang Wang, and Jian Shi
The 4+1 Dynamic Management System of Lifecycle . . . . . . . . . . . . . . . . 91

Junichi Yagi and Eiji Arai
A PLM Tools Taxonomy to Support Product Realization Process:

A Solar Racing Car Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
D. A. Guerra-Zubiaga, E.D. Ramon-Raygoza, E.F. Rios-Soltero,
M. Tomovic, and A. Molina
vii
viii Contents
Social Issues of Product Lifecycle Management: Developing Cross Cultural

Virtual Teams; Supporting Todays Green Manufacturing Imperative;
Educating and Preparing Tomorrows Workforce; and Impacting
Inter-Organizational Relationships in Supply Chain Management . . . . . . 135
Hannah Anderson, Alyssa Anglin, Ligia-Varinia Barreto,
Sharron A. Frillman, Tanner J. Georgiades, Scott R. Homan,
Jonathan F. Kochert, Magdalena B. Lech, John E. Sukup,
Cynthia Tomovic, Kari L. Wilde, and Michael Wisma
Product Design Optimization: An Interdisciplinary Approach . . . . . . . . . . 169

Sergio Romero Hernandez and Omar Romero-Hernandez
A Comprehensive Business Approach to Product Realization

Using Service Oriented Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Vijay Srinivasan
Integration of Collaborative Engineering Design Using Teamcenter

Community in Mechanical Engineering Curricula . . . . . . . . . . . . . . . . . . . 205
Xiaobo Peng, Ming C. Leu and Qiang Niu
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Product Life Cycle and Assessment Parameters
Joze Duhovnik
Abstract The paper presents a review of some important articles about the
product life cycle. The phases of the entire cycle are specifically defined, from
product development to its use and elimination. On the basis of this phase
definition, a concept of unified assessments is elaborated, which essentially
defines the characteristics of both the product and the external influences on
its life cycle. The main phases are presented, as well as the use of specific fields
that essentially supplement the assessment of a products suitability during
individual product life cycle phases. The design parameters are used and sup-
plemented with constraints during the product conquering phase. Specifically,
product parameters during the phase of product use are presented as the key
elements of assessment. Finally, the product elimination parameters are also
used for assessment. Costs are included as well; these are specific to each
product phase and need to be discussed separately. All the three product phases
form the basis of a comprehensive assessment. The division to three separate
product life cycle phases and definition of parameters for validating the pro-
ducts characteristics constitute a significant contribution to a fuller under-
standing of the product life cycle process.
Keywords PLC PDM Lifecycle Assessment Product Validation Product

Perfection
1 Introduction
The term Product Life Cycle (PLC) denotes the products life in general. This
refers to products which constitute technical systems. Technical systems are for
example mechanical assemblies, electronic devices, information systems or
biotechnical systems. In principle, technical systems are intended for the fulfill-
ment of a certain technical process function [6]. Each technical process as a
J. Duhovnik (*)
University of Ljubljana LECAD Laboratory for Global Product Realization
e-mail: joze.duhovnik@lecad.uni-lj.si
M.M. Tomovic, S. Wang (eds.), Product Realization: A Comprehensive Approach, 1

DOI 10.1007/978-0-387-09482-3_1, Springer ScienceBusiness Media, LLC 2009
2 J. Duhovnik
cybernetic model has transformation of material, information and energy. A

technical process can exist only if certain resources are available. In a similar
manner as technical processes are analyzed, the IDEF0 foundations are given
for process description [7]. A certain technical system placed in nature usually
serves a particular purpose. After the analysis of natural and technical pro-
cesses, one can establish whether a technical process is a good substitute to a
natural process [3]. The more a technical process can substitute a natural one,
the better the technical system substituting this natural process.
Figure 1 shows that the levels of processes and systems are the same when
compared in two environments: nature and technology. The crucial question
here is one of transition from a natural process to a technical one. A natural
process must be recognized as accurately as possible in order to recognize a
natural system (or its functioning) in sufficient detail. It is crucial to execute
the process of transition from a natural to a technical process as thoroughly
as possible. But this can be achieved only if a comparison is made of the two
systems: the natural one and the technical one. During such a comparison,
one seeks to achieve the highest possible level of perfection of the technical
system.
A requirement is usually set that the result of assessment of the level of
perfection of the technical system should be its level of perfection. Therefore,
this level will be sought in the technical part of the assessment. This method of
searching and assessment proved good with the axiomatic design method [13],
with which an optimal solution for a certain design parameter can be found.
During product development, however, not just a few parameters are taken into
accounta greater number is needed, including the constraints, which are those
UNIVERSE, EARTH,
ENVIRONMENTAL
Product
NATURE Recognition, of the Man
Description Technical processes
Nature processes
ONTOLOGY
Natural and technical science
Description on TELEOLOGY basis
Product
NATURE
of the Man
Description,
Substitution Technical systmes
Nature systems
Comparision
Fig. 1 Connection between nature and technology vs. interaction of processes and systems
Product Life Cycle and Assessment Parameters 3
that enable an optimum solution in a given environment. It is the optimality of

the solution that usually constitutes the most critical element of new product
development. Optimality as a condition takes into account certain given bound-
ary conditions, which are normally related to the environment in which the
product will be used, throughout its scope. Specifically, boundary conditions
are related to the technical and technological situation, as well as the directions
of development in a certain environment. For this reason, balancing between a
real natural system on one hand and a real technical system on the other is the
basic, constantly recurring question for each new product.
A product is not excluded from nature, it is connected with it in every case.
Many literature resources [4], [5], [8], [16] etc. have recently been recognizing the
basic product cycle in nature. Some authors distinguish between individual life
cycles with respect to product type [16] etc., while others seek a generalized
model based on material foundations, processes, manufacture, sales, recycling
and disposal [4], [5] and others.
After process analyses (design, manufacturing etc.), the understanding of
product life cycle was positioned within a closed cycle of material, information
and energy flow. Recently, the very last portion of the product life cycle, at
which point the product has reached the end of its use and essentially constitutes
only the material, has been increasingly studied. The development of new
materials is mainly aimed only at generating them, without considering their
disposal after use [17]. The level of knowledge about nature is low and the
process of product elimination is analyzed only to a small extent, therefore one
can expect large problems when dealing with recycling and disposal of products.
During development and designing of technical systems there is still insufficient
emphasis on the optimization of parameters of the third phase of the product
life cycle. It is no coincidence that the analysis of the last products phase is
difficult; the reason for this also lies in the fact that the products analysis during
the first phase of its life cycle is inadequate. Therefore, in the case of many
products, the problem of recycling is still addressed only at the end, and not
during their conceptual design, unfortunately. Researchers Xirochakis and
Dimitris [17] and [18] have specifically drawn our attention to this problem.
When knowledge of the entire PLC and of the essential variables for analyz-
ing the fulfillment of individual product functions becomes more comprehen-
sive we can expect to be able to systematically monitor the product life cycle.
The objective of this paper is to first provide a review of the structure of
activities taking place during a product life cycle and then to define individual
variables for high-quality product assessment through its individual phases. To
make assessment as unbiased as possible, a general method is derived below for
a quantitative evaluation of the results. A special chapter presents a generalized
diagram for changing individual variables throughout the product life cycle. A
generalized LPP-T (Level of Product Perfection-Time) diagram will be given for
technical assessment; its logarithmic scale for the time variable significantly
simplifies the understanding of the product life cycle.
4 J. Duhovnik
The characteristic constraints of the proposed presentation of the product

life cycle will be discussed in the conclusion.
2 Description of Individual PLC Phases

Any product, which is intended for the execution of a specific technical process,
is defined in the conceptual design phase and during analysis, including testing.
Processing of new products from idea to realization is called the Development &
Design Process (D&DP). With respect to different levels of the design process
(from designing anew (i.e. from scratch) to adaptive design) [2], we sometimes
use the term Research & Development process (R&D) for designing a new, but
this includes only the developmental part of the process, in which special and
new product functions are researched and developed still in the form of a rough
prototype, and not a product undergoing manufacture. Such a process is
usually used for mass production (automobiles, pharmaceuticals, electronics).
During this period, a product is formed according to the basic requirements,
which are usually specified as a design task or a list of product specifications. In
some areas, some analysis like a RMS (Reconfigurable Manufacturing System)
present some method, which could be used for some particular evaluation in the
first step and finally used for development of the product. [1]. All methods of the
D&DP include an incorporated demand that the specification of requirements
be as clear as possible [6], [10], [12], [13], [14], [15] and others. On the basis of the
requirements, one can define assessment of the solutions or develop a mathe-
matical relationship for each product; this mathematical relationship is defined
on the basis of the characteristic parameters as the design parameter [13].
Thus some kind of axioms, which are product-specific, can be developed on the
basis of a design task. It is especially important to emphasize that this method is
impoverished, so to speak, because it does not take the constraints directly into
account; some researchers [11] do not even consider it as method for development
of new products. Because of the specificities of the D&DP, during which the
product is created and materialized, one can simply speak of product conquering.
According to various authors, the phase of product conquering usually
includes the following: an idea (which is related to the given situation based
on the natural environment), product development, engineering analysis,
designing, prototype, production, product manufacture and testing. The
descriptions of all these processes may vary somewhat with the author, but
the comparison of individual processes definitely reveals roughly matching
phases, as was explained in Fig. 2. In the presentations of some authors
discussing the product life cycle, the number of phases is usually reduced to
three: idea, conceptual design and product manufacture. The phase of product
conquering can be treated as a continual iterative process [3]. The essential
element at the iterative procedure of product conquering are activities taking
place in the designers room (a team with one expert from a particular field, or
Specification of design requirements:

Requirements for function, functionality, shape
model, shape, manufacture, costs, price, market
etc.
Desires, general limitations, regulations, laws
Derivation of the axioms (certion details).
Golden loop
of design in
FUNCTION
each step by
SHAPE
Use of results at N-th stage Designer`s
of manufacture in the the product
room
development-design process development
Presentation, expert
opinions Product
improvement
Fig. 2 Product conquering phase. Continuously iterative process with Golden loop [3]
several such experts). The flow of information from specifications to the pre-
sentation and use always revolves within the golden loops [3], and the objec-
tive of this process is to reach the products golden mean of sorts (i.e. a
harmonious product, perfectly adjusted to its purpose in a specific
environment).
Important part in the design golden loop is specification of design require-
ments. On the list of design requirements is not just technical data, but also
commercial, financial, production, user profile and ecological. All those
requirements are as a starting position for searching the better answer, which
we are aspect from the designer. Designer in that case is a man or team at all.
According our knowledge all those requirements are normally specified but
later in the PLC are forgot and generate once. We proposed in our PDM and
PLM systems that the basic requirements should be used through all PLC.
During the PLC the new knowledge of the product was generate or described
continuously. This knowledge we just added on the list of requirements. One or
part of requirements should be change during the development process. In that
case we changed it. In the search process was happened that some new axioms,
new function or new data was done. We should understood that fact on the
dynamical way. So we added the list of specification.
All data on the specification list is a goal for the designers. Usually the
particular result on some development stage or phase the goal was not fully
satisfied. The differences between the goal and real result we unified with the
6 J. Duhovnik
Product perfection during conquering phase
100%
Level of product perfection
80%
60%
Normal D&DP
40% Very fast first part

in D&DP
Significant
20% handicap in D&DP
0%
0.0 0.3 0.6 0.8 0.9 1.0
Time (unification) [nonlinear scale]
Fig. 3 Phase of product conquering during the PLC, taking into account the assessment of the
products level of perfection
percentage range. If the difference is too small we used more sensitive scale by
presentation.
During the phase of product conquering, several different approaches can be
undertaken, but irrespective of the approach, the teams quality always plays
the crucial role (Fig. 3). The diagram shows three ways of product conquering.
Series 1 shows normal flow of the product conquering process and Series 2 an
accelerated course. In Series 3, the process of product conquering is initially
slower, but later, with a planned inclusion of a team that is better in terms of
knowledge and equipment, it is possible to prove that one is achieving the same
results within the required time.
Any product that is manufactured is intended to be used. Therefore, it is
important what happens to a product during its use. In addition to sales
(distribution, start-up and initial use during the warranty period), the following
activities also take place: maintenance, reconstruction and perhaps even partial
recycling of consumable materials. At the same time that the product is placed
on the market, manufacture needs to be changed) in order to be harmonized
with sales in terms of quantity, quality and scheduling. The initial product
assessment criteria now change as well. The technical criteria are the most
important ones for creating good products, but business criteria are crucial
for good sales. Usually environmental aspect as by into nature not included into
the production and sale strategy, mainly because the manufacturer, sales man
or final customer do not pay attention, whether the product is environmentally
tolerable or not. For that reason we used additional criterion for ecological
assessment of products. Therefore, it is possible to use business parameters for
assessment during the process of products use. It is no coincidence that the concept
of redistribution of manufacture originated from business parameters [11] is a
possible way of reducing costs. During a products use, the basic parameter is
cost, both on the side of manufacture and on the side of the buyer. On the side
of manufacture, costs reduction is associated with greater profitability. On the
side of the buyer, lower costs are achieved via cheaper product use. In such
evaluation, costs appear as a generalized concept. In addition to costs, the
products competitive position in the market also has to be reviewed in this
period, since one of the main assessment parameters is the fulfillment of
customer expectations regarding the products competitiveness. When we
searched for a suitable parameter, a decision was made to use the products
level of perfection. The level of perfection does not include only the technical
characteristics or the level of technical perfection; it is complementary to the
criterion of competitiveness.
The phase of product use consists of two parts (Fig. 4), which need to be
balanced. The first is comprised of production, planning, purchasing, manu-
facturing, assembly, testing of the main product parameters, packaging and
warehousing. The second one comprises distribution, start-up, initial use during
the warranty period, current maintenance, investment maintenance and recy-
cling of consumable materials. The division of a product to two separate life
lines, which are interwoven with the use of concurrent engineering, but con-
ducted in sequence along the main timeline, is just one option for a clearer
analysis of the product life cycle in both the technical and business senses.
The product life cycle must be concluded within a natural environment, so
that product components remain part of nature. The completion of product life
Product perfection in use phase
100%
80%
60%
40%
20%
0%
1.0 3.0 6.0 7.8 9.0 10.0 30.0 60.0 77.8 90.0 100.0
Time (unification) [nonlinear scale]
'Normal D&DP and small lose product perfection in use phase

Very fast D&DP and latest bigger lose of product perfection in use phase
Significant handicap in D&DP and product perfection lose in short time
Fig. 4 Phase of product use. Several parallel flows are shown different influence from product
conquering phase
8 J. Duhovnik
cycle is enabled by various processes (End Of Life, EOL). The quickest way to
eliminate a product is to destroy it. However, this is accompanied by various
additional influences, since destruction in itself does not imply permanent
returning of materials to nature; it merely means destruction of a product.
Other processes used in the EOL cycle also enable partial or total return of
materials to the natural cycle.
We are therefore dealing with the principles of product reuse and recycling,
which enable its partial prolonged use or prolongation of the life cycle of
individual modules by separating its parts. With the reuse principle, each
individual module is understood as a partial or main function of an individual
product component. At the level of the products material structure, the prin-
ciples of recycling are applied, which make it possible for materials to be reused
as basic materials, changed materials or as filler. Some part of the product
normally remains and is technically or technologically useless, therefore it is
usually disposed of. It is important to understand that this process is manage-
able as well and should not be left to chance. At the EOL, the products level of
perfection falls to zero. An important criterion for deciding on process type are
costs, but these should not be estimated only from process costs; they should
comprise total costs, including those of persistence of the products remainders
in a certain environment, e.g. in nature.
The product elimination phase (Fig. 5) is crucial for understanding the
overall product life cycle. Processes taking place in this phase are destruction,
Product perfection in elimination phase
100%
80%
60%
40%
20%
0%
100.00 100.14 100.20
Time (unification) [nonlinea scale]
Natural material fast recycling process - dangerous material

and slow removeable process
process recycling is planed
destruction - bombing and final location on the refuse dump
Fig. 5 Phase of product elimination

reuse, recycling and disposal. Here as well there are three separate lines of
processing, but these usually run separately, without any special connections
between them. Processing depends on the decision regarding the product elim-
ination procedure, which is normally determined both in terms of time and
process already in the phase of product use. The decision on the EOL process
is ordinarily related to its costs. Any additional parameters that appear during
assessment of the suitability of an individual process can be used either as a
supplement to or substantiation of the decision. This also indicates a relatively
low level of knowledge about this part of the process and its importance in
places where the concentration of products is so high that nature can no longer
accept them, either in the form of product non-use (random disposal) or as its
destruction (destruction, no elimination).
In the Fig. 5 we have three different types in EOL phase:
Type 1: Elimination and longer disposal of product remains at appropriate
sites. Natural material made chance for fast recycling process in the
beginning. Dangerous material request special technology and made
slow removable process.
Type 2: Prediction of a appropriate product elimination and disposal
process. Strategy or scenario of EOL is planed and can be shorter as
it can.
Type 3: Product destruction (blasting) and depositing without sorting. Some
typical EOL when blasting was used. Example: situation by war where
bombing was used. Another example: some particular activities by ecolo-
gical and technical no regain consciousness people.
Finally, the variation of the products level of perfection throughout the
three phases of the product life cycle can be examined in Fig. 6. In the figure are
present three typical curves for product perfection during the product life cycle.
Product perfection during life cycle

100%
Level of product
80%
perfection
60%
40%
20%
0%
0.0 0.3 0.6 0.8 0.9 1.0 3.0 6.0 7.8 9.0 10.0 30.0 60.0 77.8 90.0 100.0 100.1 100.2 100.3
Product conquering + Product use + Product elemination
Time (unification) [nonlinerar scale]
Normal D&DP Very fast first part in D&DP Significant handicap in D&D process
Fig. 6 Product life cycle through all its phases

10 J. Duhovnik
Those examples are used by experts system for planning and controlling the
D&D process which has important influence on the product use and product
elimination phase. We can predict some typical deviation in those two phases
according the activities in the D&D process.
3 Determination of Variables in Individual Phases

and Product Validation
Three typical phases of the PLC, i.e. three characteristic periods in the life of a
product were identified in the previous chapter. It is interesting that these three
periods can be compared with biological processes in nature (birth, life and
death). There are different criteria for validating the products success, i.e. its
suitability for its environment in each period of the PLC.
The product developed or search without any criteria which are defined clearly
especially for our new idea about the product. Designer must create all criteria by
himself and for that reason the creation of right criteria is a major task in the first
stage of development phase. Definition of criteria can be as an axiom or some
definition or limitations which are based on designers knowledge. Usually
definition or some parameter limitations are based on ontology. If designer has
more knowledge from soft science their criteria are more undefined (softly).
If designer has more knowledge on the natural-technological and engineering
science his criteria are more defined on the numerical or physics basis. Generally
all criteria depend of the designers knowledge about the nature.
During the development phase criteria should be deeply defined or changes.
By the design phase criteria are usually more precisely defined. When we discuss
about criteria we understood that finally we find out some parameters. In
generally the parameters have value inside the interval, which is defined by
designers. Parameters with one value should be used by adoption design. It is
strictly prohibited by the design a new or innovative design. The field of criteria
is very important and designers must specify them.
A few criteria per phase that are crucial for assessment were already men-
tioned in the previous chapter. Figure 7 shows the influence of the said criteria
in all phases of the PLC. The values of these criteria are classified into groups
below:
A. Product conquering phase
VPP,C Products technical level of perfection 5070%
VC,C Manufacturing costs during product conquering 2010%
VP,C Profitability 3020%
VO,C Other 020%
B. Product use phase
VPP,U Products level of perfection 2030%
VC,U Production costs 4050%
100
Others Others
Others
80
Profit Profit
Value of criterian in %
Profit
60
Cost
Cost
40
Cost
PP
20
PP
PP
0
PP Cost Profit Others Total PP Cost Profit Others Total PP Cost Profit Others Total
Product conquering Product used Product elimination
Influences of different criterians during PLC
PP / Product Perfection in some phase
Fig. 7 Influence of individual criteria during individual phases of PLC
VP,U Profitability 3010%

VO,U Other 1020%
C. Product elimination phase
VPP,E Products level of perfection 525%
VC,E Elimination costs 5070%
VP,E Products profitability during PLC 4510%
VO,E Other 015%
The presented criteria can be discussed within three basic categories: techni-
cal level of perfection, costs and profitability. The category Other usually also
includes supply, i.e. accessibility in real time. During the product conquering
phase, real time represents the customers expectations (market in a broader
sense), during the product use phase, which is also the customer or the market,
and the same applies to the phase of elimination. The impact of individual
criteria varies due to different work tasks and different expectations in indivi-
dual PLC phases.
The author believes that because of variable impact of individual criteria
over the entire PLC one can easily understand some entirely opposing expecta-
tions of the participants in individual phases. Technically oriented professionals
insist on the technical level of perfection, while economically oriented ones are
against high costs or even want no costs for new products. Financial profes-
sionals desire reliability of financial operations, i.e. a suitable, and not extreme,
degree of profitability. Balancing between all of the extremes is possible if the
goal, i.e. to have a new product on the market, is clear.
The magnitude of influences during individual phases needs to be determined
separately for each product. Limit values stated in the diagram are merely
12 J. Duhovnik
recommendations. Their general range cannot be determined with a sufficient

accuracy, but the list and extent of influences are intended for general and initial
assessment.
In addition to these criteria, which serve as the basis for assessment of the
suitability of a new product, the product must also be assessed according to other
criteria. Features (categories) important for manufacture, sales, distribution, use
and the environment need to be taken into account as well.
On the basis of a review of important categories, a decision was made on the
following ones, which in our opinion comprise the main influences throughout
the three phases of the PLC. The technical category denotes exclusivity con-
cerning engineering resources.
The category of production takes into account the entire manufacturing
complex in product materialization. The commercial and financial categories
refer to components of business operation. Users are listed in schematic way
because their decisions regarding the product have a significant impact on its
continued production. The category of nature is termed so in order to generalize
the influence of the environment.
By more detailed analysis we used individual criteria. Individual criteria
are more detail assessment, which give in the D&D process or better in whole
life cycle of product good and right view about the total quality of the
product.
Below are present and roughly divided assessments of individual criteria,
each comprising the relevant specifications:
Technical criterion
tec = TECH (function, shape, functionality, form, technology use, etc.)
Categories as a function, shape, functionality, form, technology use, etc.
are we defined closely to the products. For different categories we used
a parameter, which present technical criteria for product who is
developed.
Commercial criterion
com = COM (market, purchase, delivery, goods exchange, etc.)
Categories as a market, purchase, delivery, goods exchange, etc. present
environment, which has influence of man and their society. Categories
present commercial parameters which are reflected to product from envir-
onmental use.
Finance criterion
fin= FIN (profit, investment, security, bilateral & multilateral exchange,
etc.)
Categories as a profit, investment, security, bilateral and multilateral
exchange, etc. present international finance and banking systems, which
depend from country to country and international geographical and

political situation.
Production criterion
pro = PROD (cost, logistics, human resources, employment, social security,
etc.)
Categories as a cost, logistics, human resources, employment, social security
define production capabilities on the different location. Manufacturing
possibilities are included in tfunction.
User criterion
use = USE (better machine, better quality; easy operation, minimum main-
tenance, fast delivery, etc.)
By use we have important user assessment, which include many different
categories like a: better machine, easy operation, better quality, robust
use, minimum maintenance, etc. Definitions of those categories are com-
ing from virtual users, which are simulated in development and design
phase. In that case we usually used some expert system by some particular
market and users.
Eco (Nature) criterion
eco = ECO (environmental safety, no environmental impact, more reuse,
easier recycling, less disposal, security, etc.)
Categories, which present eco system are today more important. They are:
environmental safety, no environmental impact, more reuse, easier recy-
cling, less disposal, security. By the assessment at these categories plays an
important role the local (country) legal system.
All those criteria can be used for a general assessment by product validation.
The parameters, which are used by some criterion, were generated on the
requirements basis. List of requirements was used as a list of parameters,
which was defined in the early stage of D&D process. All parameters, as a
description of function, technical specification, description of law, financial
data etc. we can put in the validation process. The reason for that is an
assumption that the product is not depended of some particular fields. The
product is one part in the nature (Fig. 1) and his validation should be compare
to it.
It is evident that the above categories are not directly related to the criteria.
Many categories also have different influences on the criteria. This means that
the assertion about a direct, solid association between individual categories and
the criteria is incorrect. For each individual category, it is necessary to sepa-
rately define the parameters and their values. It is especially important to have a
separate definition of the parameters for each products location on the earth
(in the world (the universe in general)). The function values for individual
14 J. Duhovnik
categories also vary with time, as the system is dynamic and not static. A general
equation applying to individual products was written, independent of the dis-
tribution of the influence of individual categories on the criteria. Influence of
whole criteria can be summarized in function (1), which we can denominate as a
product validation function.
k h
P i
wj; x xj; t
j1
x; t k h i (1)
P
wj; x xj; max
j1
x = x (tec, com, fin, pro, use, eco) index of different criteria, like a
technical, commercial, financial,
production, use and ecological
j index for particular parameter
inside the different criteria
(x),(t) product validation function pre-
sent the product assessment for
typical criteria defined in time t,
where we made assessment
(x)j,(t) product validation function for
typical category and particular
parameter (indexing with j)
defined in time t, where we made
assessment
(x)j,max maximum value of product vali-
dation function for typical cate-
gory and particular parameter
(indexing with j) defined in time
interval from to till t, where the
assessment was made
w (j),x weight function for each particu-
lar parameter and each different
criteria
If we like follow dynamic changes by some criteria and validated it we used
particular equation which is developed from (1). Validation function for tech-
nical criteria is follow:
k h
P i
w j; tec tecj; t
j1
tec; t k h i (2)
P
w j; tec tecj; max
j1
j index for particular parameter inside the different criteria

(tec),(t) product validation function on technical basis, present the
product assessment for technical criteria defined in time t,
where we made assessment
(tec)j,(t) product validation function on technical basis for typical
category and particular parameter (indexing with j) defined
in time t, where we made assessment
(tec)j,max maximum value of product validation function on technical
basis for typical category and particular parameter (index-
ing with j) defined in time interval from to till t, where the
assessment was made
w (j),tec weight function for each particular parameter and each
different criteria
It is essential to be able made relation and influences of individual categories during
the assessment, so that in spite of different categories one can obtain their relative
influence on the final assessment (Fig. 8). It is vital to always obtain values between 0
and 1 for each assessment category. If a product is monitored over time t, from its
development and use to elimination, an interesting variation of the products assess-
ment is obtained. On the basis of changes in the trend of assessment for individual
categories, one can adopt decisions or institute measures over the entire PLC.
If one wishes to evaluate a product according to criteria from individual
categories and over a certain time period (t) during the PLC, a detailed graphical
presentation should be made for this time interval. As a rule, assessment becomes
valuable only when the product enters a more mature phase (over 60% of the time
required for its development). Assessment during earlier phases may even be
disturbing and stressful for the developmental-design and manufacturing teams.
On the other hand, a product can be assessed especially thoroughly at the time of
its direct use, which is part of the next phase. On the basis of the products first
entry into the environment, one obtains characteristic trends of product assess-
ment per criterion (Figs. 9a, 9b and 9c).
Product validation using different criteria

Value of indivudual criteria
1.00
0.80
0.60
FI technical
FI commercial
0.40 FI financial
FI production
FI user
0.20 FI eco
0.00
0.00 0.50 1.00 2.50 8.00 20.00 50.00
Product conquering and used phase [nonlinear scale]
Fig. 8 Variation of product assessments (various criteria) during the PLC

16 J. Duhovnik
(a) Product validation with technical FI technical

and commercial criteria
FI commercial
1.00
Value of individual
criteria 0.80
0.60
0.40
0.20
0.00
0.00 0.50 1.00 2.50
Fig. 9a Product validation with technical and commercial criteria
(b) Product validation with financial

FI financial
and production criteria
FI production
Value of individual criteria
1.00
0.80
0.60
0.40
0.20
0.00
0.00 0.50 1.00 2.50
Fig. 9b Product validation with financial and production criteria
(c) Product validation with user FI user

and eco criteria FI eco
1.00
Value of individual
0.80
criteria
0.60
0.40
0.20
0.00
0.00 0.50 1.00 2.50
Fig. 9c Product validation with user and ecological criteria

4 Trends of Assessment According to Individual Criteria
When assessing individual parameters in a certain category (tec,com,fin,pro,

use,eco), one attempts to define the comparability of assessments in absolute
terms and as accurately as possible, particularly their variations. Therefore, data
capture at predefined time intervals is very important. Irrespective of temporal
variations, though, one must also monitor the products location in the universe.
Sometimes, comparable locations can appear for individual products, even
though the difference in details between individual locations is very large.
Insufficient sensitivity of product assessment at various locations (sometimes
called remote viewing) is manifested as excessive oscillations of the achieved
results. As a rule, such locations must be identified, and the assessment para-
meters must be analyzed separately and monitored in order to be able to isolate
and analyze them separately in the event of changes.
Assessment can be performed on the basis of hard or soft data. Hard data, for
example, include physical data (capacity, power etc.), financial data (price, costs,
interest etc.). Soft data are for example technical data (industrial design, techno-
logical capabilities etc.), user competence (better work capabilities, good after-
sales services, etc.). For each parameter, an attempt is made during the
assessment process to determine its value as soon as possible on the basis of
hard data, not by capturing soft values. It is understandable that the problem has
several layers, therefore some assertions by individual authors found in literature
regarding hard data for assessment parameters which are in principle soft, are
very questionable. As a rule, such data can be formed only on the basis of global
capture of values and their special transformation into data. Since this is sensitive
information, global commercial complexes must protect them as business secrets.
Our team has simulated a few cases of products, with five different locations of
product conquering, use and elimination, and the results were very interesting.
When reviewing suitable mathematical formulations for evaluating the sen-
sitivity of assessment, an attempt was made to select the kind of function that
would be more sensitive at small differences and less at greater ones. This
method was chosen on the basis of experience in assessing the technical char-
acteristics of products according to the VDI 2222 reference. At absolute values
in the low range of assessments, linear determination of differences still enables
a rough determination of the successful solution. In the event of smaller
differences, i.e. with very similar technical solutions, however, the absolute
values are so small that they usually do not enable high-quality evaluation.
Therefore, a differentiated assessment method was chosen and defined with the
term First Step to Better Product (FSBP).
FSBP 1 x; tSEN (3)
x; t product validation function by typical criteria named x, defined in

time t, where we made assessment, see equation No.1
18 J. Duhovnik
Sensibilness of the FSBP function is mostly defined with potential coefficient

named SEN, which is defined with equation
q

SEN 1 x; t1 ; t2 min (4)
x; t1 ; t2 min
minimal value by the product validation function by
typical criteria named x in the investigation time inter-
val, from t1 to t2
So obtained FSBP values enable a better differentiation and easier decisions
even for values around 1.0. It should be emphasized, though, that other impor-
tant information besides the assessment includes minimal values, as well as the
time interval itself.
In Fig. 10 we presented FSBP curve for technical and commercial para-
meters. The interval in PLC is from 0.50 to 4.0. That interval is usually very
sensitive for product perfection value. Include the final phase at product con-
quering phase and first phase or starting period in product used phase. Assess-
ment in that phase made also good treatment value for product value on the
market and environment at all.
In generally we find out that such analysis give better and clear value for self
control process. The parameters are generated for product itself. If the chosen
FSBP assesment at technical and

commercial criteria
1.20
1.00
Value of indivudual criteria
0.80
0.60
0.40
0.20
0.00
0.50 0.75 1.00 1.50 2.50 4.0
FI technical FI commercial
FSBP technical FSBP commercial
Fig. 10 FSBP curve at technical and commercial assessment compare to the product perfec-
tion function in the watching interval from 0.50 till 4.00.
parameters are right or appropriate the D&D team can recognize during the
process. If they find out that the parameters are not appropriate at all they can
be added too. In that case the additional parameters must be used from starting
point of assessment process. The new parameters could not be used for assess-
ment from some time or for some time in PLC interval.
5 Conclusion
The paper presents three characteristic product phases. This means that pro-
duct planning needs to be performed over the entire PLC. The current knowl-
edge of product planning or economic resources planning is focused only on
manufacture, therefore an expansion of such an information system to PLC
conquering can be problematic. Certain processes occurring in the phases of
product use and elimination are therefore analyzed in lesser detail or very
specifically for individual product types.
For easier product conquering over its entire life cycle, six basic criteria have
been defined (tec,com,fin,pro,use,eco). By each of these criteria, we used some
categories for detail analyses. Categories should be defined with some para-
meters, which are presented like a value or described. PLC started with D&D
process, where we defined list of specification. The specification list is not static
but should be changed during the development of the product, generally during
the PLC. Maximum of the request is a goal of the products requirements. In
reality we try the best that the goal can be achieved. Differences between the
goal and our real results we like used as an assessment value. In the paper was
proposed unification measure with the percentage scale.
We also introduce the product validation function which can be used as a
timely assessment of the product during the PLC. For some particular reason
we can also individual criteria use as a (tec,com,fin,pro,use,eco).
Generally speaking we define the requirement for the selection of assessment
parameters. Product assessment was then performed at certain time intervals,
with the same starting points. When the variation of the assessment function for
all six parameters was reviewed (tec,com,fin,pro,use,eco), it was found that it is
possible to enforce the extrapolation principle for the variations expected in the
following time intervals. An estimate of the scatter of expected results was also
used. Timely intervention enables correct and more successful directing of
trends already at time t, when the last assessment of product is performed.
Among five products, interventions were made in two on the basis of assess-
ment by directing activities towards additional progress. Progress was first
enforced in the phase of product conquering (product development) and for
the second time in the phase of its use (after-sale services). For various reasons,
detailed data on product types cannot be provided here.
The method was applied in the standard PDM (three cases) and PLM (two
cases) systems. We added standard software with the data base of specification
list, which was followed all PLC. Through the PLC we made product validation
20 J. Duhovnik
function and the designers, researchers, salesman, controlling department and

marketing man follow their activities. Some particular action was done accord-
ing the trend analysis.
The method enables the analysis of those important categories that are
relevant for the product, not only in technical terms, but also generally, includ-
ing economic and user influences, but not neglecting the assessment of the
products environmental impact, i.e. its effects on nature.
References
[1] Bi Z.M., Gruver W.A., Lang S.Y.T.: Analysis and Synthesis of Reconfigurable Robotic
Systems, Concurrent Engineering, Vol. 12, No. 2, 145153 (2004).
[2] Duhovnik J. (1990): Contribution for natural classification of products related to CAD/
CAM, CIRP International Seminar on CA-Design, Ljubljana 1990
[3] Duhovnik J. (2002): New Challenges in Development and Design Process, CAMT,
Wroclaw
[4] Eisenhard, J.L. and Schepper, M.S., Approximate Life-cycle assessment in product
design, Proceedings of DETC00, Maryland, USA
[5] Galina A., and Iyigun, M.F. (2002): Outsourcing Globally During the Product Life
Cycles: A Theory and Some Evidence, Working paper No. 0208, University of
Colorado, Boulder, USA
[6] Hubka, V. (1976): Theorie der Konstruktionsprocesse, Springer-Verlag, Berlin 1976
[7] FIBS PUB 183 (1993): IDEF0 Process Modeling Standard, Federal Information Proces-
sing Standards Publications (FIBS PUBS), Nationals Institute of Standards and
Technology (NIST), USA
[8] Ishii, K. (2002): Modularity: A Key Concept in Product Life-cycle Engineering, Design
Division, Stanford University, CA 94305, USA
[9] Prasad, B. (1996): Concurrent Engineering Fundamentals, Vol. I, Integrated products
and process organization, Technomic, Lancaster, USA
[10] Prasad, B. (1997): Concurrent Engineering Fundamentals, Vol. II, Integrated products
and process organization, Technomic, Lancaster, USA
[11] Pack H. and Saggi, K. Vertical Technology Transfer via International Outsourcing,
Juornal of Development Economics, 65, 389415
[12] Decision Factors of Product Life Cycle Strategies, paper od Department of Precision
Engineering, School of Engineering, University of Tokyo, 2001
[13] Suh, N.P. (1990): The Principles of Design, Oxford University Press, Inc., New York
[14] Tavcar J. and Duhovnik J. (1999): Typical Models of Product Data Integration in Small
and Medium Enterprieses, The international Journal of Advanced Manufacturing
Technology, Springer Verlag
[15] VDI 2221, VDI Verlag, Duesseldorf, 1985, 1993
[16] Windley J.P. (2002): The discipline of Product Management, Paper of the Utah
Governement, State of Utah, USA
[17] Xirouchakis, P. (1998): Computer Support for planing the life Cycles of Eco-Effective
Product Systems, 5-th CIRP International Seminar on Life Cycle Design98, Stockholm.
[18] Xirochackis, P., Kiritis, D., Persson, J.G. (1998): A Petri net technique for process
planning cost estimation, CIRP, GA, Athens.
Collaborative Optimization and Application
of Active Suspension
Hao Wang, Shaoping Wang, and Mileta M. Tomovic
Abstract This paper addresses the current state of collaborative optimization

(CO) and its application to active suspension system. CO is a bi-level optimiza-
tion architecture, which preserves the autonomy to individual disciplines.
Through this decomposition, CO can solve the loose-composite problem.
When it comes to strong-composite cases, CO will have difficulty in conver-
gence because of the form of system-level constraint in its structure. In order to
change this nonlinear phenomenon, we use Response Surface Model (RSM) to
modify the interdisciplinary discrepancy. Central Composite Design method is
used to select the experiment point to construct the RSM.
Keywords Collaborative Optimization Compatibility constraint

Approximation approach Agent model Response Surface Model (RSM)
Design of Experiment (DOE)
Nomenclature
xsl Design variables in system level
xsl Optimal values for system level
xi The i th subsystem design variables
xi Optimal values for subsystem level
" Dynamic factor in interdisciplinary compatibility
lu ld lk lp Length of upper and lower wishbone, kingpin and knuckle
du dd dk dp Diameter of upper and lower wishbone, kingpin and knuckle
u ; d Yield stress of upper and lower wishbone
k Yield stress of kingpin and knuckle (considered as a whole)
l Angle of tire plane
l Horizontal slip of the tire
max Maximum value among u ; d ; k
H. Wang (*)
School of Automation Science and Electrical Engineering, Beihang University, Beijing,
China
e-mail: PeterWang.Leaf@gmail.com

22 H. Wang et al.
Constant in analysis
Density of the rod, gain = 7:8 103 kg=m3
Yield stress for steel, gain = 70 MPa
E Modulus of Elasticity, gain = 200 GPa
Poissons ratio, gain = 0.3
K1 Stiffness factor of sprung, gain = 1:3 104 N=m
C1 Damping factor of sprung, gain = 1000
M1 Sprung mass (quarter car model), gain = 330 kg
F1 Preload force of sprung, gain = 3236.2 N
K2 Stiffness factor of unsprung, gain = 1:7 105 N=m
C2 Damping factor of unsprung, gain = 0
M2 Unsprung mass (car model), gain = 25 kg
F2 Preload force of sprung quarter, gain = 3481.4 N
1 Introduction
Multidisciplinary design optimization (MDO) is a field of engineering that uses

optimization methods to solve design problems incorporating a number of
disciplines. Especially to the complex system, its performance is related to
many disciplines. Whereas the traditional design of complex system always
concerns some disciplines while it ignores others, it may lead to design conflict
among different disciplines (Fig. 1).
Aerodynamics
Weights
Traditional
design always
focus on one
discipline Structures
Manufacturing Propulsion
Noise
Fig. 1 Traditional design always focus on one discipline

Collaborative Optimization and Application of Active Suspension 23
Details of the CO method have been developed by several researchers,

primarily in the last 20 years, although closely-related work is described in
publications as early as 1977 (Peterson et al. 1977). CO is a promising decom-
position algorithm introduced by Braun (Braun et al. 1996) and further devel-
oped by Sobiesky (Sobiesky 1998) to solve the general (non-convex) MDO
problem. There are many applications in the aviation industry. Budianto
(Budianto, 2004) presents a study of collaborative optimization as a systematic,
multivariable method for the conceptual design of satellite constellations. Perez
(Perez 2006) presented an integrated control-configured aircraft design-sizing
framework. It overcomes the challenges that the flight dynamics and control
integration present when included with the traditional disciplines in an aircraft
sizing process. During these applications, there are two ways in subsystem level
analysis: CAE analysis and approximation method. Because of the difficulty
and time cost of the computation, some of the application used approximation
method instead of direct CAE analysis. Sangook Jun (Jun et al. 2004) used CO
to design the aircraft wing. Genetic algorithm is widely used for the system level
optimization while the gradient-based method as a subspace optimization
algorithm. The response surfaces are exploited to realize CO in subspace in
some research work. Hence, the human interface is crucially important to
enable engineers to control the design process (Enblom 2006). It is necessary
to establish a platform that enables data transmission between different com-
mercial uni-discipline software and realize the MDO. Xu Lin (Lin 2006) pro-
posed a platform that integrated aerodynamic, propulsion and structure
disciplines to design a missile based on MDO framework iSIGHT. S.-X.
Chen developed an email-based communication technology of global coordi-
native optimization of distributed structural system (Chen 2005). The applica-
tion of CO in the automobile field is rare. WU Bao-gui proposed a CO
integration platform that did successfully test the whole vehicle virtual model
in crash, aerodynamic and NVH (Noise, Vibration and Harshness) disciplines
(Wu et al. 2007). MDO through CO method and multidisciplinary integrated
simulation is a new attempt in active suspension design.
CO is a good method in MDO because its structure is like real division of
work in industry. However, CO is still a relatively immature decomposition
method because its compatibility constraint in system level adds much non-
linear characters to the optimization. This problem is still not solved completely
as little experience in true industrial environments is available. But some efforts
have been made to change this disadvantage of CO algorithm. Approximation
approach is one of the successful methods among the improvements. Approx-
imation approach is part of agent based model method and Design of Experi-
ment (DOE) is another part of it. In this paper we use Response Surface Model
(RSM) which is one of the approximation approaches widely used in MDO to
solve a two-discipline optimization problem on active suspension system.
The remainder of this article is organized as follows: Section 2 introduces the
CO methodology and its character. Section 3 provides agent model to solve the
discrepancies of the disciplines. Section 4 describes the application on active
24 H. Wang et al.
suspension system which concurrent design is used and two CAE model sharing
the same CAD model has been proposed to improve the design efficiency.
Section 5 is the conclusion of whole paper.
2 Collaborative Optimization
Before discussion of CO, brief introduction and comparison of four key MDO
decomposition frameworks are summarized from 2006 European-U.S. MDO
Colloquium (de Weck et al. 2007) (Table 1).
2.1 CO Methodology
With recent advances in the field of MDO, it is possible to transfer the tradi-
tional vertical design process into horizontal process, enabling concurrent
analysis and design. Among many MDO methods, CO shown below (Fig. 2)
is recognized as suitable method to design and optimize multidisciplinary
coupling problems.
CO method consists of two-level optimization architecture. The subsystem
must satisfy all of their disciplinary constraints. To achieve its design task with
given target variables from the system, the subsystem can choose their local
variables freely. However, when there is not enough degree of freedom in
choosing local variables, the subspaces are also allowed to change their target
copies with minimum departure from their target variables. The task of the
system is to adjust the target variables so that all subspaces can achieve their
own task, while minimizing the system-level objective. When using this
method, it is very similar to the real work. Every department and expert of
different disciplines can do their own design and optimization separately
(Perez 2006).
Table 1 Overview of MDO decomposition frameworks

Method
Characteristic BLISS CO ATC CSSO
System-level Analysis No No No Yes
Required?
Subspace Sensitivity No No No Yes
Analysis Required?
Number of Levels Two Two Multiple Two
Partitioned by: Discipline Discipline Object/ Discipline
Analysis Analysis Component Analysis
Subspace optimization Yes Yes Yes No
influenced by targets?
Autonomous Subspace Yes Yes Yes Yes
Optimization?
Fig. 2 Collaborative optimization method
At the system level (SL), the collaborative optimization objective is stated as:
X
m
min fxsl Subject toJi xsli ; xi xslij xij 2 "; i 1; ::::::; n (1)
j1
where fxsl represents the system level objective function which is also the
design objective function. Ji represents the compatibility constraint for the i
th subsystem (of the total n subsystems) optimization problem and xi are the i th
subsystem design variables whose dimension is m. Variables with a superscript
asterisk indicate optimal values for the subsystem level optimization. Note that
the system level constraint assures simultaneous coordination of the coupled
disciplinary values.
The lower-level objective function is formulated such that it minimizes the
interdisciplinary discrepancy while meeting local disciplinary constraints. At
the disciplinary level, the i th subsystem optimization is stated as
X
m
min Ji xsli ; xi xslij xij 2 subject to gi xi 0 (2)
j1
where gi is the specific disciplinary constraint; variables with a superscript

asterisk indicate optimal values for the system level optimization.
Considering the condition of three variables and two disciplines as an
example, the process of CO is illustrated in Fig. 3 (Kobayashi et al. 2005). In
the figure below, three coordinate axis represent the values of design variables
X, Y and Z. The two surfaces and represent the constraint of two disciplines.
Our goal is to find out the minimum value of object function !. From Fig. 3(a),
we can clearly find the optimal point is point 0 that located on the intersection of
two constraint surfaces. Point 1 represents a set of system-level design variables
that are given as the initial value. Then it will go to subsystem level. In order to
26 H. Wang et al.
Z Z
X
X
4 4
2 3

Y
0
decreasing Y 0 decreasing
(a) (b)
Fig. 3 Process of CO
satisfy the constraint, we need to find the closest point to point 1 on the two
surfaces according to geometrical meaning of the equation (2). The result of this
step is point 2 and 3. Then in system level, the geometry meaning of the
constraint is two spheres. The optimal point is among the intersection of these
two spheres. We can get point 4 here that is closer to the optimal solution. Then
go on the iteration to get closer to and arrive at point 0 (Fig. 3(b)).
2.2 Advantages of CO
Major characteristics of CO are its bi-level and distributed structure. CO has
computational and organizational advantages, and these advantages can be
maximized when used for a large-scale design optimization consisting of a large
number of design variables and disciplines (Jun et al. 2004).
Collaborative optimization is formulated to remove direct communication
among disciplines, so as to guarantee disciplinary autonomy. Computationally,
decrease of direct communication among disciplines enables a reduction of
computational cost, especially in large-scale problems that require large
amount of data exchange at each iteration. Furthermore, each disciplines
analysis tools can be directly integrated with a specific optimization algorithm
without much modification. Organizationally, the collaborative optimization
architecture provides a natural fit to the current disciplinary expertise structure
found in most design organizations and used by most project teams.
2.3 Weaknesses of CO
2.3.1 Problems from the Experiments

I. Kroo and V. Mannings experiments have suggested that the price that must
be paid for the advantages of decomposition is a somewhat increased
computational time (Kroo and Manni 2000). CO often leads to inefficient

convergence, especially when gradient-based method is used for system level
optimization. In this case of gradient-based method use, CO also strongly
depends on the initial condition for convergence (Jun et al. 2004). A few
difficulties caused by certain features of its architecture have been also reported.
The architecture, with discipline-level optimizations nested in a system level
optimization, leads to considerably increased computational time. In addition,
numerical difficulties such as the problem of slow convergence or unexpected
nonlinearity of the compatibility constraint in the system level optimization are
known weaknesses of CO (Jang 2005).
2.3.2 Reasons of the Problems of CO

The use of quadratic forms for the system level compatibility constraints
means that near the solution changes in system targets have little effect on
the constraint values. Specifically, the gradient approaches zero, leading to
difficulties for many optimizers, especially those that rely on linear approx-
imations to these functions. The implications of the singular Jacobi are dis-
cussed in (Alexandrov and Lewis 2000) using an SQP method at the system
level. For a simple test case, convergence of the system problem was not
achieved. This failure is attributed, not to a failure of the optimizer, but to
the specific CO formulation (Jang 2005). That is because the system-level
problem fails to satisfy the standard first-order necessary conditions
(Karush-Kuhn-Tucker or KKT conditions (Fiacco and McCormick 1990)).
3 Agent Based Model Approximation
From the above analysis, we know that the bi-level and distributed structure
allows CO to easily decompose MDO problem but also causes some unavoid-
able convergence difficulty in computation. Some alternate choices for the form
of subspace objectives, system constrains, and optimizers are discussed in
(DeMiguel, and Murray 2000).
Several ideas for modification have been explored which focus on alternating
choices for the form of interdisciplinary discrepancy. These modifications can
be categorized as following:
1. Loose system constraint method. The constraint in system level is equal
constraint which is ideal but very hard to satisfy. Robert Braun (Braun
et al., 1996) knows this problem from the beginning and proposed that we
should change the equal constraint to inequalityJi xsli ; xi s. The writer
has used this method to test a simple MDO problem and found that the
DONLP, LSGRG2 and SLP method all failed in the optimization because
they are emanative and could not find the optimal solution. For the value of
S, it cannot be so small because there will be a difference between xsli and xi
28 H. Wang et al.
so it can meet the subsystem level but cannot meet the system level. The S
cannot be big either because it is hard to meet this constraint in system level
optimization.
Li Xiang
improved this method by using a dynamic factor
" l X1 X2 2 ; l 2 0:5; 1in inequality constraint instead of station-
ary factor (Xiang 2004).
2. Penalty function method. Angel-Victor DeMiguel and Walter Murray pro-
posed a new CO formulation called Modified Collaborative Optimization
(MCO) which uses penalty function instead of the quadratic penalty func-
tion used by Braun (DeMiguel, and Murray 2000).
3. The use of response surfaces to model the system constraints. I.P. Sobieskin
and I.M. Kroo had expounded two kinds of method used in CO and refine-
ment was performed using ideas from trust region methods (Sobieski and
Kroo 2000).
While in this paper we use response surfaces (RS) to modify the system
constrains. RS method is an approximation approach which is widely used in
MDO field.
3.1 Response Surface Method (RSM)
Response surface methodology (RSM) is a collection of statistical and mathe-

matical techniques useful for developing, improving, and optimizing processes.
In general, suppose that the system involving a response y that depends on the
controllable input variables x. The relationship is y = f(x). Because the form of
the true response function f is unknown, we must approximate it. Usually, a
low-order polynomial in some relatively small region of the independent vari-
able space is appropriate. The 4th order or Quartic model is represented by a
polynomial of the following form:
X
N X
N X
N X
N X
N
Fx a0 bi xi cii x2i cij xi xj di x2i ei x4i (3)
i1 i1 1 i1 i1
where
N is the number of the model inputs
Xi is the set of model inputs Xi
a, b, c, d, e are the polynomial coefficients, the total number of coefficients is
(N+1)(N+2)/2+2 N.
3.2 Design of Experiment (DOE)

Experimental design, in which a prescribed set of experiments or trials (system
analyses) is performed, can be used to study the effects of design parameters on
the design states so that intelligent design decisions can be made. Primary
considerations in experimental design are as follows:
1. Number of experiments that can be performed (given cost and time
constraints).
2. Values for the parameters in each experiment.
3. Proper interpretation of the results.
During every subsystem-level optimization we need to construct the
response surface function according to some experiment point. In our RSM
process, we use Central Composite Design (CCD). CCD is a statistically based
technique in which a 2-level full-factorial experiment is augmented with a
center point and two additional points for each factor (called star points).
Thus, five levels are defined for each factor, and to study n factors using CCD
requires 2n+2n+1 design point evaluations. Figure 4 shows the CCD points
for three factors.
The center and star points are added to acquire knowledge from regions of
the design space inside and outside the 2-level full-factorial points, allowing for
an estimation of higher order effects (curvature). The star point(s) are deter-
mined by defining a parameter which relates these points to the full-factorial
points by
Supper b u b a
(4)
Sslow b b l a
where
b = baseline design
l = lower factorial point
u = upper factorial point, l < b < u.
Fig. 4 Central Composite

Design points for three
factors
30 H. Wang et al.
Although CCD requires a significant number of design point evaluations, it

is a popular technique for compiling data for Response Surface Modeling due
to the expanse of design space covered, and the higher order information
obtained (iSIGHT user reference guide 2000).
3.3 RSM and Whole Computation Flow
The whole computation process is described as follows:

1. According to CCD experiment method we mentioned before we select a
group of system level design variables vector Xsik , k = 1, 2, 3. . ..
(N+1)(N+2)/2+2 N, where N is the number of the input variables.
2. Performs the subsystem-level optimization in every discipline and get the
optimal result of the objective function Jik , i = 1, 2, 3. . ..n, k = 1, 2, 3. . ..
(N+1)(N+2)/2+2 N, where n is the number of the disciplines and N is the
number of the input variables.
3. Construct the RSM function between subsystem optimal objective functions
and system optimal design variables according to Jik and Xsik :
X
N X
N X
N X
N X
N
Jik a0 bi xi cii x2i cij xi xj di x2i ei x4i
i1 i1 1 i1 i1
4. According to formula 1 we can get the system level optimization problem is:
Min fxsl
Subject to
(5)
X
N X
N X
N X
N X
N
Jik a0 bi xi cii x2i cij xi xj di x2i ei x4i 0
i1 i1 1 i1 i1
5. Performs the system level optimization and get the new design variables
vectorXsi
Aforementioned process executes literately until the whole system
converged.
4 Application on Active Suspension

In performing this analysis, it is assumed that the mass of the members is
negligible compared to that of the applied loading. Friction and compliance
at the joints are also assumed negligible.
4.1 Subsystem-level Design and Analysis
4.1.1 Dynamic Subsystem

The Fig. 5 above shows the double wishbone suspension structure. The length of
upper and lower wishbone is lu and ld . lk and lp represent the length of the kingpin
and knuckle. The height of the wheel is h. We need to measure the angle of wheel
plane l and horizontal slip of the wheel l. When the wheel beating from top to
bottom, we just need to design the length of every rod and make a correct layout.
Then we can make the changes l of l and in the scope of the limit.
Now we will deduce the relationship between l,l and wheel structure.
When the wheel beat from top to bottom, the upper and lower wishbone will
rotate around rotation center. The angle of rotation is a very complex and
nonlinear function so we cannot get them easily. We will use lu ; ld ; lk and
lu ; ld ; lk to express them. We define that the projective length on the ground
of OA, AD, DE, BC are l0d ,l0k , l0p and h0 . According to the geometrical relation-
ship from the Fig. above, we can easily deduce following formula:
l0d ld cos lu ; ld ; lk (6)

1
l0k lk cos lu ; ld ; lk
2
l0p lp cos l
1
h0 h sin l
2
l ld lp l0d l0k l0p h0 ld lp

1 1
ld cos lu ; ld ; lk lk cos lu ; ld ; lk lp cos l h sin l
2 2
B lu
C
lp
E D
Car body
lk
A

ld O
Fig. 5 Dynamic analysis of

a two double wishbone
active suspension l
32 H. Wang et al.
From the formula above, we can consider that l is related to lu ,ld ,lk and lp .
The height of tire h is also in the function but we do not consider it as a design
variable. The slip displacement of wheel l should be not more than 5 mm, that
is l 5 mm. This is the allowable displacement of elastic deformation.
4.1.2 Structure Subsystem

To the structure discipline, load is the active force provided by hydraulic
cylinder (Fig. 6(a)) (Happian-Smith 2002). Assume Fw is the wheel load and
Fs is the maximum active force we calculated in Section 2 through multidisci-
plinary collaborative simulation.
When the free body diagram of the wheel and knuckle is considered (Fig.
6(b)), the directions of Fw and FB are known and together establish the point of
concurrencyP1 , for the three forces that act on the body. If the magnitude of Fw
is known, the magnitudes of FA and FB can be determined from the triangle of
forces through:Flkw Fluk pF
2
A
2
(Fig. 6(c)). For the free body diagram of AO
lu lk
(Fig. 6(d)), the point of concurrency is at P2 and with FA , Fs and lu ; lk known,
Fo can be found from the second triangle of forces (Fig. 6(e)).
To the lower wishbone OA, the load and geometry parameter is shown in
Fig. 6(d). According to the model, we can get the formula from the mechanism
handbook (Joseph et al. 2004). The cross-section G is the dangerous one
because it is located at the maximum bending moment (absolute).
Fs a2
FA 3ld a=coslu ; lk
2l3d
Fs a2 2
Mmax 3ld a2 4ald =coslu ; lk (7)
2l3d
Mmax
d d
Wz
Fig. 6 Forces analysis of a double wishbone active suspension

Iz
where Wz ymax represents the flexural coefficient of the cross-section;
Iz denotes the moment of inertia to the neutral axis that is related to the
geometry of the cross-section; ymax represents the maximum displacement
from the point of stress tensor to the neutral axis. To the plastic deforma-
tion for ductile material such as metal, we use Von Mises yield criterion.
The von Mises yield criterion, as a function of the principal stresses, is
defined as:
s
1 2 2 2 3 2 3 1 2
(8)
2
In the formula, 1 ,2 ,3 , are fist, second, third principal stress and yield
stress. The cross-section G can be considered that only the first principal stress
works. The constraint is that maximum stress tensor is smaller than the yield
one1 d ld ;lu ;ldk ;a;dd 40.
To the upper wishbone CB, we can make the analysis similar to OA:
Fa2
FB FA tan 3ld asinlu ; lk =cos2 lu ; lk
2l3d
Fa2 lu
Mmax 3ld asinlu ; lk = cos2 lu ; lk (9)
2l3d
Mmax
u u
Wz
The moment of inertia is similar to the expression above. We can consider

that the stress tensor of CB is related to design variables ld ; lu ; lk ; a; du . The
constraint is that maximum stress tensor is smaller than the yield
one1 u ld ;lu ;luk ;a;du 40.
Considering kingpin and knuckle as a whole, we can also get the constraint is
k lk ; lp ; dk ; dp k .
Through the analysis above, we can use CO method to solve the structure
and dynamic coupling problems. The analysis result is summarized in Fig. 7.
4.2 Implement of the Design and Optimization
One of the most challenging aspects of multidisciplinary design and optimiza-

tion is the sharing of disciplinary data between the various analysis codes.
MDO frameworks iSIGHT is chosen to build up the MDO model. The model
constructed in Pro/ENGINEER can be used in Patran and ADAMS. Through
34 H. Wang et al.
Fig. 7 Expression of coupling problems of structure and dynamic discipline in CO
this kind of sharing model, the speed of design and optimization can be
improved significantly. In optimal cycle, we cannot run the CAD/CAE soft-
ware every time by Graphics User Interface (GUI). Instead of it, we execute
command files that automatically record every operation step, save the com-
mand flow at the first time, and modify them according to our need. The data
transmission among every step is shown in Fig. 8.
4.3 Result
Before the optimal process, we need to normalize the unit of every parameter
before we calculate the discrepancies of the disciplines. In the suspension
Fig. 8 Automatic data transmission in iteration
system, we use linear normalization to change the value to the domain from 0 to 1.
The calculation rule is as follows: Xn XXX min
max Xmin
Xn 2 0; 1
The initial and final value are XandXn . The normalization is preceded in the
step S5 and D3.With the collaborative design and hierarchy optimization, we
can get the optimal values listed in Table 2:
With the CO method, we can get the mass of the suspension deceased by
22.52%. The dynamic and structure performance of the system are greatly
improved.
36 H. Wang et al.
Table 2 Result of the optimization

Initial Min Max
Nomenclature value value value Result Unit
lu 35 26 38 36.26 cm
ld 50 40 58 41.65 cm
lk 32 28 40 31.28 cm
lp 26 20 30 24.50 cm
du 40 30 50 38.05 mm
dd 40 30 50 41.60 mm
dk 40 35 45 34.80 mm
dp 40 35 45 40.40 mm
a 40 35 50 33.60 cm
l 12.5 1.8 mm
max 97 50.8 MPa
G 14.3033 11.0826 kg
5 Conclusion
This paper summarizes the advantages and weakness when using CO method to
solve MDO problems. Bi-level structure and nonlinear system level constraint
makes CO hard to converge and even fail to find the optimal solution. Three
kinds of modifications of CO have been summarized and compared.
4-order response surface model was used in this paper to the application of
active suspension. The presented result showed that the CO coupling process
worked. The analysis and optimization models were simple enough to develop
the desired methodology without involving high computational load. If the
design variables become more and then we need more experiments and time
because it requires O(n2) function evaluations.
In the future, the complexity of models in each discipline will be increased to
analyze realistic active suspension models. Furthermore, a stronger coupling
will be developed by including the control effects of the hydraulic components
in hydraulic and control disciplines. Finally, several decomposition will be
implemented on the active suspension design process and products that are
more comprehensive can be undertaken.
References
Alexandrov, N.M. and Lewis, R.M (2000) Analytical and Computational Aspects of Colla-
borative Optimization, NASA TM 2000-210104.
Braun, R., Gage, P., Kroo, I. and Sobieski I. (1996) Implement and performance issues in
collaborative optimization, NASA, and ISSMO. In 6th Symposium on Multidiscilinary
Analysis and Optimization, Bellevue, WA.
Budianto, I. and Olds, J.R. (2004) Design and deployment of a satellite constellation using
collaborative optimization. Journal of Spacecraft and Rockets, vol. 41(6), pp. 956963.
Chen, S.X. (2005) A communication technology for internet-based collaboration optimization

for distributed structural systems. Structural and Multidisciplinary Optimization, vol. 29(5),
pp. 391397.
de Weck, O. et al. (2007) State-of-the-art and future trends in multidisciplinary. 48th AIAA/
ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference,
Honolulu, Hawaii.
DeMiguel, A. and Murray, W. (2000) An analysis of collaborative optimization methods. In
8th AIAA/USAF/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimi-
zation, Bellevue, WA.
Enblom, R. (2006) Two-level numerical optimization of ride comfort in railway vehicles.
Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid
Transit, vol. 220(F1), pp. 111.
Fiacco, A.V. and McCormick G.P. (1990) Non-linear programming: sequential unconstrained
minimization techniques. Classics in Applied Mathematics. SIAM, Philadelphia.
Happian-Smith, J. (2002) An Introduction to Modern Vehicle Design, Warrendale, Pennsylvania:
SAE International.
Jang, B.S. Yang, Y.S. and Jung, H.S. (2005) Managing approximation models in collaborative
optimization, Structural and Multidisciplinary Optimization vol. 30(1), pp. 1126.
Jun, S., Jeon, Y., Rho J. and Lee, D. (2004) Application of collaborative optimization using
response surface methodology to an aircraft wing design. In 10th AIAA/ISSMO Multi-
disciplinary Analysis and Optimization Conference, Albany, New York.
Kobayashi, T. and Kroo I., (2005) The new effective MDO method based on collaborative
optimization. In 35th AIAA Fluid Dynamics Conference and Exhibit, Toronto, Ontario,
Kakamigahara.
Kroo, I. and Manni V. (2000) Collaborative optimization status and directions. In 8th AIAA/
USAF/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimization,
Long Beach.
Lin (2006) System design of missile based on CO with variable complexity modeling. Collection
of Technical Papers AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics
and Materials Conference, pp. 56185624.
Perez, R.E. (2006) Multidisciplinary optimization framework for control-configuration inte-
gration in aircraft conceptual design, Journal of Aircraft .vol, 43 (6), pp. 19371948.
Peterson J. et al. (1977) Journal of Spacecraft & Rockets, vol 11.
Shigley, J.E., Mischke, C.R. and Budyknas, R.G. (2004), Mechanical Engineering Design, 7th
edition, pp. 261, New York: McGraw-Hill Professional.
Sobieski, I.P. and Kroo, I.M. (2000) Collaborative optimization using response surface
estimation. AIAA Journal, vol. 38(10), pp. 19311938.
Sobiesky I.P. (1998) Multidisciplinary design using collaborative optimization. PhD thesis,
Stanford.
Wu, B.-G., Huang, H.-Z. and Tao, Ye. (2007) Functional digital prototype-based MDO &
simulation. Journal of System Simulation, vol. 4, p. 861.
Xiang, L. (2004) MDO and its application in aircraft design. Doctoral Thesis, Northwestern
Polytechnical University, China.
Back to the Future: Product Lifecycle
Management and the Virtualization
of Product Information
Michael Grieves
Abstract Product Lifecycle Management (PLM) is new as both a term and an

acronym, with their usage occurring only in this new century. Since PLM is new,
it is necessary to have an understanding of how it is defined. PLM is based on the
substitution of product information for wasted physical resources and using that
information for the entire lifecycle of the product. One of the attractions of PLM
is that it is a return to a concept as old as the concept of a product. That concept is
a product-centric view of the product from creation to disposal. However, mass
production and division of labor had made that concept unworkable. The ability
to virtualize products allows a return to a product-centric view. The Information
Mirror Model (IMM) is a conceptual framework that describes how this product-
centric view will be operationalized and used through the four stages of the
products life: create, build, support, and dispose.
Keywords PLM Product Information IMM Product-Centric View Back

to the future
1 Introduction
Product Lifecycle Management (PLM) appears to be a new and different
approach to product information. As a defined term and acronym, Product
Lifecycle Management and PLM are new. However, on closer examination,
PLM is really a back-to-the future concept.
PLM trades information for physical resources, specifically wasted time,
energy, and material. This is not a new concept. Humans have attempted to
do this for as long as they have had the capacity to reason and remember.
However, this ability to reason and remember was on an individual effort and
limited by the capabilities of the human brain. It only worked on a small scale.
M. Grieves (*)
Product Lifecycle Management Center of Excellence, Purdue University Discovery
Park, Burton Morgan Building, 610 Purdue Mall, West Lafayette, IN 47907
e-mail: mgrieves@purdue.edu

40 M. Grieves
However limited this ability, humans did concern themselves with the lifecycles
of their artifacts or, as I will refer to them, their products. Humans created products.
They built those products. They maintained those products in good working order.
When those products outlived their useful life, humans disposed of them.
Because for the most part, humans relied on mental representations that they
maintained within their own memory, this model worked under limited conditions.
The products were simple. The numbers of products were small. The products
remained in close physical proximity to the product producers.
Product proximity was necessary. Information is an intrinsic characteristic of
a product. Length, width, weight, and composition, to name some of information
characteristics that products contain, are embedded within a product. Humans
could and did extract that information in order to be more efficient in using
physical resources. However, except for the memory of humans and their notes
and sketches, information about the product did not exist independent of the
physical product itself.
The ability to strip and record this embedded product information took a long
time to develop, but did get increasingly better. Standards of measurement and
instruments to perform such measurements, blueprints, and detailed product
descriptions represented product information that was embedded into the pro-
duct. However, it was only with computers and communication technology
development in the last half of the 20th century that product information could
not only be fully described but also could be dynamically manipulated regardless
of proximity to the actual product.
This opened up the possibility of linking a physical product with the information
about that product. This stripping, modeling, and linking of product information
on computers virtualizes the physical product. The Information Mirror Model
(IMM) is conceptually related to mental representations of humans. However, it
is richer, more extensive, more precise, and can be shared.
The global nature of the Internet makes this virtualized model available to
anyone independent of their geographical location and proximity to the physical
product itself. This linking of the physical product with its virtual equivalent
throughout the life of the product with the purpose of minimizing the waste of
physical resources is the basis for Product Lifecycle Management or PLM.
2 Defining PLM
Product Lifecycle Management and its acronym are literally 21st century creations.
If a search by ABI Inform for industry or academic articles about PLM as Product
Lifecycle Management before the year 2000 is done, the results are that there are no
articles found. While as discussed elsewhere (Grieves 2006), PLM came about as
other technologies and concepts, such as Computer-Aided Design (CAD), Com-
puter-Integrated Manufacturing (CIM), and Product Data Management (PDM)
were integrated into a product-centric view of product information over the life of
the product.
Product Lifecycle Management and the Virtualization of Product Information 41
While there was some usage of the term by software vendors and information
system analysts at the beginning of the century, one of the first periodicals to
attempt to define Product Lifecycle Management (PLM) was CIO Magazine. The
article, Theres a New App in Town (Stackpole 2003), was a seminal article
about PLM. It was the first time an executive-oriented magazine featured Product
Lifecycle Management as a topic that the executive suite should be interested in.
The definition in its present form was developed for my first book on
Product Lifecycle Management (Grieves 2006), although elements appeared
in a previous paper (Grieves 2005). There are some other definitions of Product
Lifecycle Management that differ, and some that differ very radically, from the
definition that I will present here. However, I would argue that the practitioners
in the field would generally agree that this definition captures what most people
familiar with Product Lifecycle Management would consider when thinking
about PLM.
The definition for PLM is as follows:
Product Lifecycle Management (PLM) is an integrated, information-driven approach
comprised of people, processes/practices, and technology to all aspects of a products life,
from its design through manufacture, deployment and maintenanceculminating in the
products removal from service and final disposal. By trading product information for
wasted time, energy, and material across the entire organization and into the supply
chain, PLM drives the next generation of lean thinking.
While there are many interesting implications of this definition, the focus of
this chapter will be on the why an integrated information approach has rele-
vance to the four phases of a products life, namely the create, build, support,
and dispose phases, and what a model for such an approach would look like.
3 Information as a Substitute for Wasted Time, Energy,

and Material
Any discussion of virtualization needs to start with the rationale for doing so. It
has only been fairly recently that economists have recognized the economic value
and difference of information as compared to other resources (Warsh 2006). The
concept of trading atoms for bits has been proposed before (Negroponte 1995).
However, the rationale for doing so has not always been made clear. It often
appears that going digital is goal in itself. However, there should be a better
rationale than that.
For PLM, it is in the definitional phrase stated above, By trading product
information for wasted time, energy, and material across the entire organization
and into the supply chain, PLM drives the next generation of lean thinking.
Lean thinking is the idea that we minimize the use of physical resources in the
four cycles of the products life: create, build, support, and dispose.
We do this by using information about the product or what we will call the
virtualized product to reduce the use of wasted time, energy, and material. We
42 M. Grieves
cannot eliminate all usage of time, energy, and material. PLM is about physical
products that are atom-based. Material has to be moved, shaped, assembled,
repaired, and disposed of. However, we can attempt to minimize the amount of
resources to do those tasks. The resources above that minimum are by defini-
tion waste.
There are many, many examples that illustrate the waste of resources that
occurs during the products lifecycle: redesigning parts that already exist;
manufacturing parts that do not fit together at assembly; ordering repair
parts that are not the right part which is discovered only when the product
has been disassembled; digging parts out of a landfill because they are toxic and
should have been disposed of differently.
Time, energy, material, and information are not directly comparable. Three
of them, time, energy, and material, are quantifiable, but have different units of
measurement. However, we do have a mechanism to equate them. That
mechanism is the costing function. We can apply the cost of each unit of
measure times the quantity of measurement to arrive at a common unit of
comparison, cost in dollars.
Information is more difficult to deal with. Unlike time, energy, and material,
information is not a rival good. The cost of information is not in its application
since the use of information does not diminish it. However, there is a cost in
collecting, organizing, and transmitting information. For purposes of this
analysis, the costs of the resources and computer and communications equip-
ment to do this collecting, organizing, and transmitting over the life of a specific
task can serve as a proxy for the costs of information.
Figure 1 is illustrative of this substitution of information for wasted time,
energy, and material. In this figure, the two bars represent the costs of any
designated task involved with a products lifecycle. The task could be the
designing of a part that has certain functionality. It could be the manufacture
and assembly of that part. It could be the repair of that part when it ceases to
function correctly.
The bottom parts of the two bars are identical. They represent the most
efficient use of time, energy, and material to perform the task in question. For
Fig. 1 Information as Time,

Energy, Material Substitute
designing the part, the bottom part of the bar represents the minimum amount
of resources it would take to create a part that met the functional requirements
if one had the ability to evaluate all possible approaches. For the manufacturing
and support tasks, the same analysis would apply evaluating all possible
approaches and choosing the one (or ones if there were multiple, equal solu-
tions) that minimized the use time, energy, and material.
The top part of the left bar represents the costs of resources that are used in
excess of the minimum. It represents the resources used in designing the part
and testing it only to find that it does not quite meet the requirements, necessi-
tating redesign and retesting. It represents rework in manufacturing the part so
that two components fit properly together. It represents having to dissemble
and reassemble the part in for repair because wiring was not connected in the
proper sequence.
The top part of the right bar represents the cost of information that replaces
the wasted resources of the task. It is engineering information that predicts that
a design will meet specific requirements. It is a manufacturing simulation that
produces a process so that rework is not required. It is a step-by-step repair
procedure that eliminates the need for disassembly and reassembly.
For purposes of illustration, the cost of information reflected in the right bar
is less than the cost of wasted time, energy, and material. For simple tasks, a
trial-and-error approach may be less costly than collecting, organizing, and
transferring information. However, for complex tasks and repeatable tasks, it is
fairly self evident that the relationship of information being less costly than
wasting time, energy, and material holds true. Learning or experience curves,
lean manufacturing (Womack, Jones, et al. 1990), and the Toyota Production
System (Liker 2004) are all based on trading information for wasted time,
energy, and material.
It is also important to note that this is an idealized representation. For
complex tasks, there is no way to know the task method(s) that uses the minimum
amount of resources. Information will only be a replacement for some of the
wasted time, energy, and material. There still may be some wasted resources.
The lifecycle of a product is made up of a number of tasks. When looking at
the trade-off of information for wasted time, energy, and material, the sum of the
tradeoffs must be done over the life of the product. If the cost of the resource for a
task is given by C(x), with x being Information (I), wasted Time (Tw), wasted
Energy (Ew), wasted Material (Mw), then for this substitution of Information to
make economical sense, the sum of the information needs to be less than the sum
of the wasted time, energy, and material (C(I) < C(Tw, Ew, Mw) over all these
tasks.
It is also important to note that different views of the product lifecycle will
yield different results. If product producers only consider the tasks necessary to
create and build the product, their calculation of costs may differ considerably
than if costs of supporting and disposing the product are taken into considera-
tion. The trend due to legal and regulatory pressures is for product producers to
take a broader view of their responsibility.
44 M. Grieves
4 Back-to-the-Future
Product Lifecycle Management is a new phrase and PLM is a new acronym.

However a product-centric view of the four phases of a products lifecycle, i.e.
build, create, support, and dispose, is not a new concept. One of the intrinsic
appeals of PLM and a source of confidence in its effectiveness is that the
concept has worked in the past. PLM is in fact an old concept that has gained
new life due to advances in information technology.
The idea that a product manufacturer concerns itself with the four phases of
a product is as old as product manufacturers themselves. The realization that
information could be substituted for wasted time, energy, and material would
have been known to even the most primitive manufacturers of products. The
product producers who did engage in this substitution would have prospered
more than those that did not. The resources that these increasingly efficient
producers did not waste could go into to more productive efforts.
In addition to this understanding, product producers of the past would have
been concerned with the entire lifecycle of their products. Product producers
were part of relatively small communities and had long relationships with their
products and the users of their products.
By way of example, it is reasonable to assume the manufacturers of chariots
were practitioners of PLM. These ancient craftsmen would have designed their
chariots with an end purpose in mind: transportation, battles, or races to name
some.
The chariot craftsmen would have manufactured the chariots. They hewed the
lumber to build the frames, weaved the basketry for the passenger compartment,
formed wood into V shapes for wheel spokes and lashed wooden segments with
rawhide to form the tires, and beat metal into shape to form the amour.
These same craftsmen would have supported the product when it was in use.
They would have fabricated new parts when old parts broke. They might have
improved certain aspects of the chariot, such as the placement of amour, after
seeing how the chariot performed in battle. They would have changed out parts
to make the chariot suitable for new terrain.
It certainly may have been that product producers would have remained very
involved in the support of their products. Standards of measurements and the
measurement tools required for interchangeable parts were not developed until
the mid 1800s. Before then, each product was unique. The individual best
capable of supporting the product was the producer of the product.
In the disposal phase, these craftsmen would have attempted to salvage all
the parts that could be reused on other chariots. They would have scavenged
all the metal parts for either reuse or melting down and repurposing for other
armament, such as spear tips. While there is no historical record of used chariot
lots, it certainly is not unreasonable to postulate that chariot manufacturers
facilitated the transfer of chariots from old (or dead) warriors to young warriors
coming of age.
The coming of the Industrial Age and the ability to mass produce diminished
the focus and the ability of product producers to concern themselves with the
entire lifecycle of a product. As pointed out by Adam Smith (Smith and Jenkins
1948) in his pin factory example, division of labor was a method of reducing the
resources required to produce a given quantity of products.
While this division of labor increased productivity due in part to a substitution
of information for time, energy, and material, there was a loss in information
about the product as a whole. Product information became siloed in functional
areas such as engineering or manufacturing. However, as products became more
complex to create, build, and support, the requirement to have a product-centric
view of product information became more compelling. However, until computer
and communications technologies developed sufficiently, it simply was not
possible.
5 Information Mirror Model
The ability to return to a product-centric perspective requires the ability to

access and manipulate product information as a construct separate from the
product itself. As discussed in the introduction, product information is
embedded within the physical product itself. While humans have had the
capability to extract this information, previous technologies for capturing,
manipulating, and transferring this information could only deal with abstracted
and limited subsets of this information.
However, in order to maintain the division of labor and distinct functional
areas that have allowed for increases in productivity, we need an informational
model of the product that is both rich and shareable. The mental model of an
individual who would be involved with a product from birth to death no longer
suffices. Modern, complex products require many individuals to be involved
with the information about a product throughout its life. However, if these
individuals are to do so effectively, they need to have all the information of the
product available without the requirement of being in proximity to the physical
product throughout its life.
In a product-centric approach, the requirements for physical-to-information
mapping exist at all phases of the lifecycle, although the direction of the flow of
information may differ. In the creation phase, mental designs need to be fully
described and tested for feasibility. In the manufacturing phase, the steps that
minimize resource usage needed to turn these designs into physical products need
to be developed and the actual physical configurations (as-built) need to be
captured. In the support phase, the as-built information needs to available so the
product can be maintained to its designed level of functionality (as-maintained),
ideally before it fails. In the disposal phase, the design intent to provide for safe,
recyclable disposal is known and can be carried out.
46 M. Grieves
It is only until recently that computers and communications technologies could

deal with the vast amount of data needed for robust representations. Additionally,
prior to computers, manipulation of this information in a dynamic fashion could
only be done internally in human minds and then only in a limited fashion. The
simple rotation in three-dimensional space of a mental model would be considered
a feat that only a limited number of people would be capable of performing.
We are reaching a point in time where the technical capability of computers
and communications makes possible the rich representations of products. In
addition to rendering the geometric shapes with great fidelity in three dimen-
sions, we are developing the capability of simulating physical characteristics of
these products with increasingly better and better fidelity and precision. Simu-
lating complex characteristics such as the crash testing of automobiles or
showing the flow of air on an entire airplane is now possible.
What will be required to fully realize PLM is a model that represents
products and informational constructs. What I have proposed is to virtualize
physical products and maintain that linkage throughout the life of the product.
This virtualized product can then be used to substitute for wasted time, energy,
and material. The intent with the virtualized product is to use bits wherever
possible in place of atoms. This model was developed to simply describe the
connection between products and their associated information. It was first
called the Mirror Spaces Model (Grieves 2005). The name was subsequently
changed to the Information Mirror Model (Grieves 2006). However, the sub-
stance of the model itself remained relatively unchanged, although refinements
and clarifications were made.
The model consists of three elements: Real Space, Virtual Space(s), and a
linking mechanism, referred to as Data and Information connecting Real Space
and Virtual Space(s). As described in Fig. 2, this is a model that uses elements
that are embedded into our very way of thinking. However, when used as a
model for managing information, it changes some very fundamental ways
product data is captured, organized and used. Specifically it de-silos product
information, such that data is not organized by its function, but by the emergent
or instantiated physical object with which it is associated. Doing so brings us
back to the future in terms of our interaction with product information.
Fig. 2 Information Mirroring Model

5.1 Real Space
As humans living in real space, we are ontologically compelled to think in terms

of physical spaces. For real objects, it is obvious that a space-based model is
natural. As much as we attempt to conceptualize and abstract the world around
us, we are inextricably bound to our view of the world that comes to us via our
senses. It is embedded in our very language (Lakoff and Johnson 1980). We
experience our world in terms of spaces having physical dimensions that govern
our understandings. No amount of abstraction can separate us from our
connection to physical space and our perception of that space (Latour 1999).
We constantly think in spatial terms even when the concepts are not spatial
concepts, such as up/down, above/below, here/there, and over/under (Lakoff and
Johnson 1999). Common examples abound. I am feeling up today. The fore-
man is over the assembly line workers. My thinking here is... We think in
spatial terms even when thinking about abstract, non-spatial concepts, some-
times unconscious that we are doing this and at other times very deliberately
(Grieves 2000).
5.2 Virtual Spaces
As humans, we have been able to create virtual spaces in our minds as far back
as there is history. While imperfect and transient, we have the ability to create
virtual spaces and control what happens in those spaces. This is the process we
know as imagining (Casey 1976). We can build representations of the real world
and change these representations either as things change in the real world, or as
we postulate things could change.
Drawing in the dirt was probably our first mechanism to share our virtual
space. However, langue was and is our main mechanism to share these private
virtual spaces. Rearranging atoms (dirt, clay, papyrus, ink and paper) gave both
our drawings and writings permanence, albeit in static form and with some
forms less permanent than others. The advent of computers allowed for
dynamic spaces that could be shared on a local basis by all connected to that
particular computer system.
The advent of computers also created a virtual space that possessed inde-
pendent processing capabilities and therefore could be dynamic. Communica-
tions technology allowed this virtual space to be shared by individuals in
geographically diverse locations. The Internet took communications to the
level of ubiquity where anyone anywhere in the world could access and interact
with the shared virtual space.
The constraints or limitations with earlier computer systems were such that their
representations of the real world were constrained by the state of computing
technology to only a very limited amount of information about real world objects.
This necessitated substantial abstraction to only the coarsest characteristics. For
48 M. Grieves
example, early computerized information about a part consisted of its part number
and limited amount of characteristics such as dimensions, color, weight, etc.
In addition, manipulations were limited by processing power. Even if real
world objects could be described sufficiently, the processing power to do anything
useful was unavailable. For example, in the 1970s, the mathematical calculations
required to rotate even a simple geometric representation required dedicating a
multi-million dollar mainframe computer.
Thanks to Moores law and its corollaries in computer storage and commu-
nication bandwidth, these limitations have reached the threshold where computing
systems can accommodate a desired functionality with respect both to mirroring
the description of complex objects and to manipulating them. Mathematical
descriptions of parts geometries are such that they can directly drive manufactur-
ing machinery to create functional parts without human intervention. In addition,
these complex objects can be combined to form even more complex objects with the
correct spatial orientations within the context of their use. This was and is a shared
space where an object could be created and manipulated. Multiple people could
agree on its interpretation because they all see and understand the same thing.
However, it was not until the global diffusion of the Internet that easily
shared, universally accessible virtual spaces were made possible. Prior to the
Internet and World Wide Web becoming readily accessible, the ability to link
multiple individuals into this shared space, especially if separated geographically,
was technically challenging and very expensive. Connections to this shared space
took planning and dedicated communications resources. The Internet enabled
nearly anyone with access to a computer and communication line to access these
shared spaces.
Per the IMM, this shared space implies a singularity of information. Unlike
real space, it is almost costless to reproduce and manipulate copies of data
objects. While a powerful advantage of virtual spaces, multiple copies can also
be a source of waste and inefficiency. In engineering design, multiple copies can
lead to wasted work being done on old versions of the object. In support, it can
lead to multiple copies of the product that reflect different components that
can only be resolved by expensive physical inspection.
Where there is a corresponding physical object, the issue of singularity of
information is obvious: the virtual space mirrors the physical space. The issue
does arise where there is no physical equivalent as in the create phase. In fact, an
advantage of PLM is to forego the costs of developing physical prototypes and
do as much work in designing and testing digitally and virtually. In this case, the
people dealing in the creation phase need to treat their virtual object as if it did
have a single, physical equivalent. This representation has been called the
controlling virtual object (Grieves 2006, p. 78). Failure to observe this con-
vention can easily result in incompatible work being done on different versions
of what should be the same virtual object. The result is wasted resources in
diagnosing the situation and substantial recovery and rework.
Where real space has only a single, unique existence, virtual spaces can be
duplicated. As indicated in Fig. 2, these subspaces (VS1 . . . VSn) can be used
recreate the shared virtual space in order perform individual work on products.
Different configurations and simulations of those configurations can be per-
formed in these private subspaces without affecting the shared space.
While a subspace may be identical to the shared space, any activity taking
place is independent of the shared space. The results of any work done in these
subspaces can be moved to shared space and become a controlling virtual object
only under the conditions specified by the owner of the shared virtual space.
The simulations afforded by these subspaces allow for much more experiment-
ing and testing than can be done in real space. As computers become more
powerful, more testing will be done via virtual subspaces. While physical testing,
such as crash testing, will continue to take place, it will be done not as testing but
as validation of the virtual tests that have taken place in virtual space.
5.3 Linking Mechanisms
Once virtual spaces are able to be built, accessed, and manipulated, the key to
IMM and consequently to PLM is the ability to link the virtual space to its real
space or physical counterpart. This is what will allow PLM to have real utility.
It is the ability to access the virtual representation of an item and know that it
substantially mirrors the state of the real world object.
To make full use of virtual spaces, linking mechanisms are required in both
directions. Data must be collected and transmitted from real space to virtual
space where it is organized into information. That information must be acces-
sible and available in real space. In both cases, this currently is a human-
intensive activity. With the exception of process control systems, the interface
between the real and virtual spaces has primarily been humans coding and
entering data and requesting information via tactile devices, such as keyboards,
touch screens, mice, graphical tablets, etc.
This means that the create phase has advanced and will continue to advance
rapidly. The creation of a physical product requires that the designer first create a
virtual product. The technologies that exist, while not of Minority Report science
fiction quality, do a sufficient job of capturing designs originating in real space and
will continue to improve. Displaying those designs back to the designers and their
colleagues in full scale (Power Walls) and in 3-D statically and even dynamically
(caves and interactive caves) also exist and are advancing. Creating physical objects
directly from this information via various technologies such as stereolithography
and laser/powder metal forming is feasible and advancing.
In the manufacturing phase, capturing the data on the creation of physical
products (as-builts) and building mirrored virtual products is occurring
slowly. The three components required to do this are: a) physical measurement
and capture equipment such as Coordinate Measuring Machines (CMMs) and
sensor equipped machinery; b) data collection and consolidation software; c)
Manufacturing Execution Systems (MES) to organize and store the virtual
50 M. Grieves
product. As manufacturers become aware of the ability to use this information

for such purposes as virtual quality control, the capabilities and technologies
will advance more rapidly.
The support phase is still in its infancy. Product manufacturers are becoming
increasingly aware of the need to instrument their products and use that data to
remain linked to the product after the product has left the factory door. As
manufacturers recognize that this creates new revenue opportunities and as
instrumentation and associated communication technology advances, this area
will grow rapidly.
The disposal phase has been ignored for the most part. However, the ability
to dispose and recycle appropriately requires that the virtual product with its
information about how it was designed to be recycled and disposed be available.
It is only a matter of time until regulatory and legal pressures make this a
priority.
For IMM to reach its full potential, the linkage between the object in virtual
space and its counterpart will have to be robust, accurate, and timely. This area
of linkage is where substantial investments and advancements in technology are
currently taking place. On the data collection side, the advancements are
occurring in areas such as RFID, mechatronics, sensors, and video recognition.
On the information presentation side, advancement areas include search
engines, virtual reality technologies (e.g. VR goggles, caves, holographic and
haptic equipment, etc.). For both areas, wireless communications technologies
are a key enabler.
6 Summary
Product Lifecycle Management as a term and PLM as an acronym are 21st

century constructs. PLM is a product-centric perspective of the information
about a product throughout the four phases of its life: create, build, support,
and dispose. The premise underlying PLM is that less expensive information
can replace more expensive wasted time, energy, and material.
The intrinsic appeal of PLM is that, although new in its current construction,
it is a back-to-the-future approach to product information. When products
were simpler and localized, information about the product could reside in
mental models of the individuals who created, built, supported, and disposed
of the products. These individuals could use the information they had about the
products to replace wasted time, energy, and material. As products became
more complex, more voluminous, and less localized, these mental models were
incapable of maintaining the information required. Division of labor was the
method needed to efficiently support such a change in product production.
However, this meant information became siloed and a product-centric perspec-
tive was lost.
It required the capabilities of the modern computers with their current

capacities in computing capability, data storage, and communications band-
width to return to a product-centric model. The requirements to fully virtua-
lize products and maintain information throughout the life of the product
required 21st century technology.
The model presented here, the Information Mirror Model (IMM), is an ideal
representation of what a virtualized product will entail. The ability to maintain
a virtual product that reflects the state of the real product and vice-versa will
allow individuals to trade this virtual information for physical resources,
namely time, energy, and material. While this is an idealized model, the trend
of applications in the PLM area is supporting the direction of this model.
While PLM is a back-to-the-future concept, the technologies of today are
qualitatively incomparable with what the chariot producers of eons ago had at
their disposable, namely their mental models and minimal drawings. Correspond-
ingly the potential trade-off of PLM information for wasted time, energy, and
material is also incomparable with greater savings as the technologies enabling the
Information Mirroring Model continue on their improvement trajectories.
References
Casey, E. S. (1976). Imagining: A phenomenological study. Bloomington, Indiana University
Press.
Grieves, M. (2005). Product Lifecycle Management: the new paradigm for enterprises. Int.
J. Product Development 2(Nos. 1/2): 7184.
Grieves, M. (2006). Product lifecycle management: Driving the next generation of lean thinking.
New York, McGraw-Hill.
Grieves, M. W. (2000). Business is war: An investigation into metaphor use in Internet and
non-Internet IPOs. Weatherhead School of Management. Cleveland, Case Western
Reserve University: 219.
Lakoff, G. and M. Johnson (1980). Metaphors we live by. Chicago, University of Chicago
Press.
Lakoff, G. and M. Johnson (1999). Philosophy in the flesh: The embodied mind and its
challenge to Western thought. New York, Basic Books.
Latour, B. (1999). Pandoras hope: Essays on the reality of science. Cambridge, Mass.,
Harvard University Press.
Liker, J. K. (2004). The Toyota way: 14 management principles from the worlds greatest
manufacturer. New York, McGraw-Hill.
Negroponte, N. (1995). Being digital. New York, Knopf.
Smith, A. and A. H. Jenkins (1948). Adam Smith today: An inquiry into the nature and causes of
the wealth of nations. New York, R.R. Smith.
Stackpole, B. (2003). Theres a New App in Town. CIO Magazine: 92101.
Warsh, D. (2006). Knowledge and the wealth of nations: A story of economic discovery.
New York, W.W. Norton.
Womack, J. P., D. T. Jones, et al. (1990). The machine that changed the world: Based on the
Massachusetts Institute of Technology 5-million dollar 5-year study on the future of the
automobile. New York, Rawson Associates.
Measuring the Impact of Product Lifecycle
Management: Process Plan, Waste Reduction
and Innovations Conceptual Frameworks,
and Logic Model for Developing Metrics
Cynthia Tomovic, Abram Walton, Lisa Ncube, Michael Grieves, Ben Birtles,
and Brandon Bednar
Abstract Companies implement Product Lifecycle Management (PLM) as a

means to reduce costs associated with wasted time, energy, and material with
the expectation that captured resources can be reallocated to product and
process improvements and innovations, potentially resulting in new revenue
streams. As organizations prepare to implement PLM, it is critical that they
understand their expectations of PLM and that they track the impact of PLM
relative to achieving their organizational goals. A PLM Process Plan, PLM
Waste Reduction and Innovation Conceptual Frameworks, and PLM Logic
Model for Developing Metrics are suggested as a guide to organizations tasked
with determining the impact of their PLM initiative.
Keywords Product lifecycle management Performance measurement Metrics
1 Introduction
Product Lifecycle Management (PLM) is a holistic business strategy built on

lean-thinking principles. Whereas lean-thinking focuses on reducing organiza-
tional wastes associated with time, energy, and materials, PLM focuses on
capturing wasted resources and reallocating them in support of product and
process improvements and innovations, potentially resulting in new revenue
streams. While numerous studies have been conducted on measuring an organi-
zations level of waste reduction, literature on how to accurately measure the
revenue generation capacity associated with PLM is relatively new (Grieves 2006;
Stark 2004; UGS 2007; Symmonds 2005).
The vision of PLM is to transform the way companies manage the product
lifecycle phases including initial ideation, concept design, product design, manu-
facturing design, production, and delivery to the customer, in-service support,
C. Tomovic (*)
Darden College of Education, Old Dominion University, Norfolk, VA, 23529, USA
e-mail: ctomovic@odu.edu

54 C. Tomovic et al.
and retirement from use (Miller 2005). Today more than ever before, if compa-
nies are to sustain a competitive advantage, they must exploit their innovative
capabilities or develop such capabilities to address the disruptive effects of
emerging technologies, customization demands, new market entrants, shorter
product life cycles, geopolitical instability, and market globalization (Muller
et al. 2005).
As PLM transforms the way companies do business, it is important that
companies understand how well they are doing, that is, whether or not PLM is
making a difference in their operations and to their bottom line. In recent years
there has been a major change in how business performance is measured. While
traditional performance measures include net income, operating income,
revenue, and while these measures will always be relevant and important in
determining profitability and successes of an organization, the role played by
project theory in developing, operating and evaluating projects is being recog-
nized (Barchan 1999).
Projects in general, but PLM projects in particular, tend to be complex
projects consisting of many different interlocking components. The failure to
pay attention to all phases of product lifecycle, particularly the end of the
lifecycle when sales dip and administrative costs soar, can result in a negative
impact on business performance (Marien 2006).
After having invested millions of dollars in technological infrastructure,
training and support, senior level executives are anxious to see reliable data
supporting their investments in PLM. Moreover, without knowing the bottom
line impact of PLM on cost-savings and revenue-generation, executives are
unable to accurately estimate the level of risk associated with future PLM
investments. The primary focus of this chapter is to suggest a PLM Process
Plan, PLM Waste Reduction and Innovation Conceptual Frameworks, and a
PLM Logic Model for Developing Metrics to guide organizations as they define
the process and the metrics used to measure how effectively they are PLMing.
2 PLM Process Plan
Businesses need to develop and follow a PLM Process Plan (Fig. 1). For every
company, there must be clear and compelling reasons to engage in PLM. It is
critical that companies consult their own strategic plan and question how PLM
can help them achieve their particular goals. In essence, businesses should first
conduct a benefits appraisal. By focusing on the goals and potential benefits,
companies will less likely focus on technology alone and instead come to see
PLM as a total system involving people, processes/practices, and technology.
Even in the early stages of considering PLM, it is important that companies
begin to think about what metrics to embrace; so that they can later measure
actual benefits achieved from the PLM related solutions deployed (CIMdata
2002). Once having consulted the strategic plan and determined on which goals
Measuring the Impact of Product Lifecycle Management 55
Fig. 1 PLM Process Plan
to focus, and once defined appropriate metrics, including baselines and targets
for each metric, the next step is to define the methodology to collect the data.
Finally, procedures should be defined as to how data should be fed back into the
system to ensure that the PLM solution is delivering expected benefits, thus,
contributing to the cycle of continuous improvements as it relates to the PLM
initiative (Deming 1993).
3 PLM Basics Capturing Wastes and Reallocating

for Innovations
Literature suggests (Grieves 2006; CIMdata 2002) that approximately eighty
percent of the overall cost of a product is based in engineering. Consequently,
most executives focus their PLM efforts on the plan/design, and manufacturing
phases of PLM. While these efforts do not represent the entire cycle of PLM, those
who are successful in their implementation of these PLM components are being
heavily rewarded with increased market share, profit margins, brand recognition
and the like. Nonetheless, in order to continue their support and investment in
PLM, executives are asked to justify expenses associated with PLM and its
potential. Alternatively, as is often the case, executives make the case that without
PLM, they stand to go out of business. In either case, once the decision to invest in
PLM is made, executives struggle with how to effectively and successfully orches-
trate such a large corporate-wide initiative, and how to demonstrate its impact.
One of the most news-worthy examples of the success and pitfalls of PLM
involves the number one and number two airplane manufacturers in the world.
Boeing and Airbus each embarked on their own multi-billion dollar venture, to
revolutionize air travel. Boeing, with its Dreamliner and major improvement in
operating efficiency and ultimate passenger experience, and Airbus with its
A380 and unprecedented passenger capacity loads, were in a first to market
race, and each had hopes of capturing the others market share. Due to version
control issues, Airbus ended up billions of dollars and years behind their target
(Duvall and Bartholomew 2007). Boeing, on the other hand, escaped this pitfall
by having maintained tighter version control.
As many other engineering and manufacturing based companies are learning to
traverse the world of PLM implementation, they are finding that there are vast
opportunities for PLM initiatives. With American companies facing issues such as
global warming, rising fuel costs, and increased global competition, pioneering
companies are finding innovative ways to address and capitalize on these new
challenges. Engineers from around the globe who pool their information to design
better products are engaged in follow-the-sun strategies, e.g. finding innovative
ways to reduce vehicle emissions, reduce vehicle weights, increase gas mileage, all
while increasing brand recognition and passenger safety (Kriz 2007).
Since implementing PLM, GM has saved thousands of hours and reallocated
their savings to improving product safety. Today, GM will simulate 175 crash
tests for every 1 full-vehicle crash test (Brown 2007). Not only are companies
simulating more prototypes and tests, but the simulations are vastly more
complicated than those prior to PLM technologies. Simulated computer models
could solve 1,000 equations simultaneously, prior to PLM, while today they can
solve over 30 million equations (Brown 2007). This level of complexity allows
engineers to analyze crash results in frames of 100th of a second.
These types of innovations coupled with a corporations willingness to invest
in PLM and undertake the more challenging opportunities are proving to be
lucrative and environmentally friendly. In an effort to reduce their impact on
global warming by decreasing new vehicle emissions, Toyota and BMW
designed cars with fewer emissions which received the attention of many new
customers and improved their market share. John DeCicco suggests that efforts
such as these are a clear example of innovative designs paying off for the
bottom line and the environment (Thomas 2007, p. 1).
3.1 PLM Waste Reduction Conceptual Framework
PLM users require product knowledge at each particular product phase and
access to product information throughout the entire product lifecycle. Sections
3.1.1 through 3.1.3 describes the PLM elements (Grieves 2006) on which the
proposed PLM Waste Reduction Conceptual Framework is based (Fig. 2).
Waste Reduction PLM Elements

Components People Process Practices Technology
Time
Energy
Materials
Fig. 2 PLM waste reduction conceptual framework

3.1.1 People
A commitment to supporting people by educating them is essential to effective
information flow. As with any change implemented at the organizational level,
willingness to guide people through the transitional period can mean the
difference between success and failure (Bridges 2003). With significant
resources already invested in a technological infrastructure it would be foolish
not to help employees change their current paradigm with regard to both how to
access and how to communicate within a system of information.
Within a traditional organization, established lines of communication sel-
dom follow a product or service through the product lifecycle. Departments
tend to act independently of one another, handing off a completed module at
the end of a departments value-adding process. Establishing clear lines of
communication between departments and facilitating real time communica-
tions clearly affects peoples behavior and will necessitate a change if increases
in productivity are to be realized.
Supporting a culture where information is shared instead of being guarded
during a products lifecycle is paramount to PLM success (Bridges 2003). An
organization can capitalize on identifying potential errors when they exist in
virtual space instead of scraping products due to errors at the manufacturing
phase. Recognizing errors earlier in the process requires employees to develop a
greater sense of trust. It also requires that each department be involved in the
entire lifecycle of a product.
Ways in which to think about the potential impact of PLM on waste reduction
with regard to time, energy, and materials across people might include: time
time to complete product cycle; expect a reduction in product cycle time as
upstream changes can be communicated immediately to persons downstream
such that changes can be accommodated without delay; energy energy used to
support meetings; expect a reduction in energy required to support people in face
to face meetings as they begin to work in virtual space; materials number of times
raw material is delivered correctly to each person in the manufacturing process,
expect an increase in the number of correct times raw materials are correctly
distributed to each person in the manufacturing process as the pull-system can be
simulated and adjusted based on actual sales figures versus historical data.
3.1.2 Processes and Practices

A focus on processes and information flow must be established in tandem with
the creation of an information system if PLM is going to be successfully
implemented. Best practices often are created through the natural norming
practices of organizations. Prior to the advent of PLM, communication
between departments may or may not have been codified. Upon review and
standardization, such communication practices may be defined as acceptable
processes.
A simple yet useful distinction between processes and practices are that
processes are input and routine driven, while practices are output or results
driven. With processes, the output or result is the results of applying the
specified routines or processes to the defined inputs. With practices, the desired
result or output is attempted to be obtained by selecting inputs and routines
from the universe of all possible inputs and routines.
The information requirements for processes and practices differ consider-
ably. With processes, the information needed is only that input that is called for
by the routines or processes. With practices, all information that is relevant, and
maybe even information that appears irrelevant, needs to be made available so
that the practitioner can explore as many potential situations as possible to
yield the desired results.
With the advent of a PLM system, members can work together in real time
on digital models. A change in design can be immediately communicated to
engineering, who in turn can adjust component requirements, which is commu-
nicated to purchasing and budgeting. Since these changes are happening con-
currently, the representative of each department can focus on their tasks as they
relate to the entire product. Due to the transparency of the PLM process,
employees will be less likely to inadvertently duplicate one anothers work
(CIMdata 2003a, b).
Accessibility to internet databases becomes more important as virtual design
and testing become a part of accepted business practices. With an adequate
computing system and access to the internet, team members no longer need to
be in close physical proximity (Friedman 2006). The ability to collaborate from
a distance reduces costs in gathering a product team in one location and allows
organizational leaders more flexibility when making decisions involving staff-
ing and team composition (CIMdata 2006).
Ways in which to think about the potential impact of PLM on waste reduc-
tion with regard to time, energy, and materials across processes/practices might
include: time number of times designs are reused, expect a reduction in time as
engineers build and store a digital catalogue of parts which may be reused in
future product designs; energy amount of energy required to sustain a manu-
facturing line, expect a reduction in energy used to run a manufacturing line as
alternative subassembly arrangements may be virtually created and consulted
with regard to energy use in advance of building the actual manufacturing
process; materials amount of inventory, expect a reduction in inventory as the
sharing of information in an information core should better support the timing
of purchasing decisions.
3.1.3 Technology
A solid technological infrastructure must be in place to support the demands of
a meta-database that can provide real time access to information by all mem-
bers of a project team. With the emergence of information technology, such as
the Internet and high-speed fiber-optic data transfer, organizations have been
able to share ideas, transfer designs, and stimulate collaboration among depart-
ments with increasing ease. Limitations such as geographic proximity to other
team members have become much smaller barriers now that data can be shared
in virtual space (Friedman 2006). By substituting data bits for physical objects,
huge reductions in time and resources can be realized.
A lifelike rendering of the product must be presented to members of various
departments if they wish to virtually manipulate product capabilities. To do
this, programs such as CAD/CAM have been created. CAD/CAM and similar
programs greatly enhance the effectiveness of engineers and allow organiza-
tions to inexpensively move their planning and design functions to geographical
regions that offer comparatively inexpensive technical designers and data
analysts (CIMdata 2003a, b; Engardio 2007).
One of the areas most able to benefit from this new technology is destructive
product testing (Grieves 2006). If a product must be destroyed to test its static
or dynamic load limits, significant costs will be incurred during the design and
testing phase. Destructive testing costs limit the amount of testing that can be
conducted. If a digital model can be tested, prior to physical testing, then many
of the incompatibilities of early designs can be resolved. Furthermore, if a
digital model can be housed in virtual space, where members of each product
lifecycle phase have access to it, an additional savings in time can be achieved as
the team members are able to work concurrently, instead of consecutively
(Gould 2003).
In the last two decades, better information systems have allowed the automotive
business to grow in the billions of dollars (IBM 2004). More efficient information
systems have resulted in a reduction of wasted time, energy, and material and
allows for the sharing of increased product-based information. According to one
study, engineers and designers find information about ninety percent faster in
digital environments than in non-digital environments (IBM 2004).
Ways in which to think about the potential impact of PLM on waste reduc-
tion with regard to time, energy, and materials across technology might include:
time time to locate information, expect a reduction in time to locate informa-
tion as it would take less time to access data in an information core than in
potentially multiple personal hard copy or electronic files; energy amount of
energy spent in distribution of parts to subassemblies, expect a reduction in
energy due to better factory layout designs than by physical trial and error;
and materials amount of scrap used in testing, expect a reduction in scrap due to
increased use of virtual testing versus physical testing.
3.2 PLM Innovations Conceptual Framework
Assuming wasted resources (time, energy, and materials) are captured; the
opportunity to reallocate these resources to innovation is available. In essence,
resources can be reallocated in support of product quality improvement and/or
Innovation PLM Elements

Components Product Process
Functionality
Quality
Fig. 3 PLM Innovations conceptual framework
increases in product functionality, and/or improvements in processes. Thus, the

benefits of PLM are not only in waste reduction, but opportunities to poten-
tially generate new revenue streams present themselves as a function of product
and process improvements (Fig. 3).
3.3 PLM Data Characteristics
The current body of scholarly literature focuses on PLM as an approach to

control information flow and access to this information in support of the
product lifecycle, e.g., plan/design, build, support, and dispose (Fig. 4).
Before operationalizing the proposed conceptual frameworks above, it is
important to review the underlying characteristics that are an integral part of
PLM, namely, the singularity, correspondence, cohesion, traceability, reflec-
tivity, and cued availability of the data in each phase (Grieves 2006). If these
characteristics are not intact, data in the information core will be compro-
mised, and the PLM initiative will likely fail. PLM can only work effectively if
the data in the information core reflects reality, and that that reality, at all
times, is recorded or reflected in the data of the information core. In this
Fig. 4 Product lifecycle

manner, there should be a one to one correspondence between the data that
describes and is stored about a particular product and the physical reality of
that product.
Singularity Much like in version or document control, singularity refers to
the identification and agreement on which version of the product informa-
tion everyone is working. The issue of controlled versions is critical. In the
case when a product is tangible, data about that product can be readily
assessed, verified, and documented. However, in the case of a virtual pro-
duct, there has to be system by which changes from one version of a product
file to the next version can be tracked and catalogued. In this manner, the
cataloguing feature of an information core helps to prevent the potential for
wasted time, e.g., the time individuals erroneously spend on working off an
older version of a product data file.
Correspondence Refers to the linkage between a physical object and the
data that describes that object. If one begins with a tangible product,
correspondence is largely an exercise in data extraction, e.g., developing
the methodology and technology that allows the physical features of a
product to be coded, catalogued, and digitally shared. Even in the case of a
virtual product, the data that is extracted about the virtual product should
permit the complete physical build-out of that virtual product. The concept
of start or smart parts is dependent upon a one to one correspondence
between the virtual product and its component parts in the physical world.
Because engineers can fully understand the constitution of a product and its
parts, be it virtual or physical, they can choose to start with a pre-existing
part that will serve their purposes, versus wasting the time and energy
required to design and create a redundant new part.
Cohesion Refers to the ability to reconcile different representations of a
product and its parts, depending on the perspective, e.g., mechanical, elec-
trical, three-dimensional, etc. The concern here is not with just the represen-
tation of the product and its parts, but the potential impact a change in one
perspective may have on another, e.g., the impact of a change in the electrical
perspective on the mechanical perspective. Given the new generation of
software development, changes in one perspective may automatically create
changes in the remaining perspectives, thereby potentially reducing time and
energy wasted. Time required to reconcile multiple versions of a bill of
materials (BOM) may soon be a thing of the past.
Traceability Refers to the ability to follow the developmental path of a
product over time due to it having been documented. Digitized data over time
allows one to run and re-run tests over multiple virtual products, from which
multiple analyses can be conducted. Start and smart parts, for example, would
have already been tested and the results of their tests would be traceable. In
this manner, time, energy and materials can be saved. Additionally, given the
increases in lawsuits, product liability cases, and claims of violations of
governmental regulations, traceability is important in a companys defense.
Reflectivity Is the ideal that changes in physical products or parts are

captured, without any or with little lag, in virtual counter-part products.
Because of reflectivity, information can be substituted for what would have
been otherwise wasted time, energy, and material. Capturing as-built infor-
mation and updating this information through out the life of a product is one of
the benefits of PLM.
Cued Availability Is simply being able to have the right information at the
right time. An added benefit of this feature is that usable information may be
made available as a function of the search engine. This feature is expected to
improve in the future as relational search technologies mature.
4 PLM Logic Model

A clear understanding of a projects conceptual framework can contribute to
the successful performance of many important developmental and evaluative
tasks along its life span (Savaya and Waysman 2005). A structured approach
that analyzes a product at all stages of the lifecycle for maximum effectiveness
and efficiency is therefore essential. The project logic model assists in describing
and articulating the project theory by guiding and structuring the process
(Savaya and Waysman 2005). A logic model is a diagrammatic representation
of the logical or causal relationships among project elements and the problem to
be solved, thus defining measurements of success (W.K. Kellogg Foundation
1998). A logic model illustrates the logic or theory of the program or project
while focusing attention on the most important connections between actions
and results. Furthermore, the project logic model helps to build a common
understanding among staff and with stakeholders which is essential for the
successful implementation of PLM. Another important characteristic of logic
models is that it helps project personnel manage for results and informs
project managers of gaps in the logic of a program and focuses attention on
the need to resolve them (Nakabayashi 2000, Bell 1998).
4.1 Developing PLM Metrics Based on a Logic Model
The model (Fig. 5) combines two views on innovation. It provides the perspec-
tive for a suite of potential metrics that help assess and develop a companys
capacity for PLM (Muller et al. 2005).
The various components of the model include:
Inputs/resources addresses the allocation of resources to effect the balance
optimization (tactical investment in the existing business) and innovation
(strategic investment in new businesses) (Simmons 2000). The resource
inputs include capital, labor, and time.
Fig. 5 PLM logic model for developing metrics (examples provided)
Process assesses the extent to which a companys competencies, culture, and

conditions support the conversion of innovation resources into opportunities
for business renewal (Harbour 1997).
Outputs are the product or service delivery/implementation targets for
PLM. These include:
Completion Performancevisibility of how well teams are doing in terms of
conformance to stage gates and specific milestones and delivery points
Resource Optimization reporting team efficiency and productivity, captur-
ing and reporting resource usage around key processes, productivity indica-
tors, resource planning and profiling for follow the sun global operations
Change Control and Change Capacity visibility of the impact of changes
in terms of direct and indirect costs, reporting on frequency and scope of
change transactions, volumes of changes, speed to completion, e.g., of the
Engineering Change Order (ECO) cycle
Configuration Management Metrics the single record capability of the
PLM system to report changes to the various baselines
Project or Product Quality capture and reporting of re-work and scrapping
levels, identification of compliance levels and right first time outcomes.
Product Development and Portfolio Management Linking in sales yield and
estimates of re-use value to determine which products bring the best returns
Commercial Cost of Risk Project slippage, additional resourcing, and

commercial penalties, among others (PTC 2004)
Customers refer to the users of the products/services (Harbour 1997).
The target audience the PLM project is designed to reach.
Outcomes are the changes and/or benefits resulting from PLM activities
and outputs (Simmons 2000). The short-term outcomes refer to the changes
in learning in terms of knowledge, skills, and attitudes that have come about
as a result of the implementation of the PLM project. Intermediate outcomes
refer to the changes in behavior, practices, or decisions, which are a result of
PLM. Long-term outcomes are the changes in conditions which include such
results as waste reduction, innovation and new products, continuous
improvement and sustainable green manufacturing.
Impact is the return on investment in strategic innovation of the PLM project.
Every project has factors that are outside of the control (positive or negative)
of the company that may influence the outcome and impact of its program/
project and these factors are referred to as the External Influences. External
influences include the economic climate, the global market, and government
regulations among others (McLaughlin and Jordan 1999).
The difference between accurately measuring whether or not success has
been achieved is crucial in the decision making process. Metrics for the input
assess whether or not adequate and appropriate resources have been allocated
to the PLM project. Process metrics on the other hand, focus on whether the
PLM activities are being conducted appropriately. These metrics focus on
quality issues (Harbour 1997). The efficiency and effectiveness, which are the
most critical measures, apply to the outputs and outcomes. Output measures
determine the extent to which the manufactured products and services are
meeting customer requirements and needs. Outcome metrics measure the
effectiveness and impact of the PLM project on meeting organizational
goals. These metrics focus on the extent to which the PLM project has met
its originally stated objectives waste reduction, innovation, and sustainable
manufacturing.
5 Summary
The goal of this chapter is to provide a PLM Process Plan, PLM Waste
Reduction and Innovations Conceptual Frameworks, and a PLM Logic
Model for Developing Metrics for organizations interested in assessing how
well they are PLMing. As previously discussed, many organizations have yet
to implement PLM fully as the question of return on investments remains
elusive. As a normal course of business, executives must justify their current
and future PLM investments as a function of organizational performance. Since
2001, a number of businesses have launched a PLM initiative based on the
understanding that if successfully implemented, PLM leads to enterprise effi-

ciencies and opportunities. However, industry reports that benefits of PLM are
difficult to assess as the same benefit can be expressed as a function of time, cost,
or quality; and often times, the benefits overlap (Stark 2004). While important,
defining metrics that accurately measure the impact of PLM is no easy task.
Acknowledgment The authors would like to acknowledge the Center for Advanced Manu-
facturing and the Product Lifecycle Center for Excellence, Purdue University for their support
of this ongoing research.
References
Army Performance Management Coordination Office (2006) Performance Measurement
Tool Kit 703-692-7412
Barchan M (1999) Measuring success in a changing environment. Strategy & Leadersh
27(3):1215
Bell S (1998) Managing and learning with logical frameworks: The case of an MIS project in
China. Hum Syst Manage 17: 1527
Bridges W (2003) Managing transitions: Making the most of change. Perseus Books Group,
Jackson, Tennessee
Brown SF (2007) Making cars safer With simulated crash testing in the design stage. Interna-
tional Herald Tribune. http://global.factiva.com/aa/default.aspx?pp=Print&hc=Publication.
Accessed 7 September 2007
CIMdata (2002) PLM benefits appraisal guide. http://www.cimdata.com. Accessed 30 June 2007
CIMdata (2003a) PDM to PLM: Growth of an industry. http://www.cimdata.com. Accessed
4 March 2003
CIMdata (2003b) PLM and ERP integration: Business efficiency and value. http://www.
cimdata.com. Accessed 8 March 2003
CIMdata (2006) Product lifecycle management. http://www.cimdata.com. Accessed 1 June 2006
Deming E (1993) The new economics for industry, government, education. Massachusetts
Institute of Technology Press, Cambridge, Massachusetts
Duval M, Bartholomew D (2007) PLM: Boeings Dream, Airbus Nightmare. Baseline: The Project
Management Center. http://www.baselinemag.com/print_article2/0,1217,a=200193,00.asp.
Engardio P (2007) Chindia. The McGraw-Hill Co. Inc, New York
Friedman TL (2006) The World Is Flat. Farrar, Straus and Giroux, New York
Gould LS (2003) The ABCs of PLM. Automotive design & production
Grieves M (2006) Product Lifecycle Management: Driving the Next Generation of Lean
Thinking. The McGraw-Hill Co., Inc., New York
Harbour JL (1997) The basics of performance measurement. Productivity Press, Portland, Oregon
IBM (2004) Integral Power train: Driving down the time and cost of power train engineering
with PLM. IBM Eurocoordination, Paris, France
Kriz M (2007) Hot Opportunities. National Journal Insiders Poll. http://global.factiva.com/
aa/default.aspx?pp=Print&hc=Publication. Accessed 9 September 2007
Marien EJ (2006) Meeting the product lifecycle challenge. Supply Chain Manage Rev
10:7580
McLaughlin JA, Jordan GB (1999) Logic models: A tool for telling your program performance
story. Eval and Program Plan 22:6572
Miller E (2005) Innovation: A key to competitiveness. COE NewsNet. http://www.coe.org/
newsnet/Apr05/industry.cfm. Accessed 10 September 2007
Muller A, Valikangas L, et al. (2005) Metrics for innovation: guidelines for developing a
customized suite of innovation. Strategy & Leadersh 33(10):3745
Nakabayashi S (2000). The Japanese version of product cycle management: Adoption, adaptation
and application of ZOPP A comparison analysis of methods and methodologies. Working
Paper Series No. 319. Institute of Social Studies. The Hague, The Netherlands
PTC (2004) PTC Executive briefing PTC Aerospace and Defense. The definition and imple-
mentation of PLM technologies. http://www.ptc.com/go/a-d. Accessed 10 September 2007
Savaya R, Waysman M (2005) The logic model: A tool for incorporating theory in development
and evaluation of programs. Adm in Soc Work 29:8591
Simmons R (2000) Performance measurement & control systems for implementing strategy.
Prentice Hall, Upper Saddle River, New Jersey
Stark J (2004) Product Lifecycle Management: 21st century Paradigm for Product Realisation.
Springer, New York
Symmonds M (2005) PLM Metrics. Boeing Technology, Seattle, Washington
Thomas K (2007) Report: Toyota and BMW reduce emissions and improve market share. The
Associated Press. http://global.factiva.com/aa/default.aspx?pp=Print&hc=Publication.
UGS (2007) Best Practice Brief. http://www.ugs.com. Accessed 15 February 2007
Kellogg W K (1998) Foundation. Evaluation Handbook. http://www.wkkf.org. Accessed
10 September 2007
Reliability-Based Collaborative Design Platform
for Hydraulic Actuation System
Shaoping Wang, Hao Wang, Fang Wang, and Jian Shi
Abstract This paper presents a reliability-based collaborative design platform

(RCDP) for hydraulic actuation system wherein the reliability design and relia-
bility analysis are implemented in whole design process. The central idea of
RCDP is to decentralize the complex system into several subsystems, in which
the propagation of designed parameters as well as shared parameters are con-
sidered throughout the multilevel hierarchy and the integrated optimization can
be achieved with distributed optimization. Since system parameters are not
necessarily deterministic, the probabilistic design is exploited to guarantee the
system performance both in structural strength and reliability in lower level of
product design. Upload the local designed parameters that are related to upper
level, we can solve the bi-level optimal design problem through establishing the
appropriate objective function and constraint. Taking full advantage of inherent
interface of commercial design software or developed medial file, RCDP can
exchange information, access public data and verify the design variables dyna-
mically within different disciplines so as to achieve the performance-based opti-
mization with reliability constraint. RCDP has clear advantages compared with
other single discipline design method, i.e. high interoperability, rapid developing
process, well performance verification and enough product uncertainty design.
On the basis of several common commercial software, e.g. material, mechanical,
dynamics, control, hydraulic and reliability, RCDP provides suitable design flow
that can join the multidisciplinary design method and reliability analysis synthe-
tically to satisfy the interdisciplinary requirement. Application of hydraulic
actuation system design indicates that the reliability-based collaborative design
can give suitable room to accommodate the conflict and coupling among differ-
ent individual discipline design in time that can save the design time, decrease the
loss due to improper design and achieve the design optimization.
S. Wang (*)
School of Automation Science and Electrical Engineering, Beihang University,
37#Xueyuan Road, Beijing 100083, China
e-mail: shaopingwang@vip.sina.com

68 S. Wang et al.
Keywords Reliability-based collaborative design Multidisciplinary design

optimization (MDO) Uncertainty Performance-based Product Data
Management (PDM)
1 Introduction
With the increasing of the demand for the high reliability and safety, more and
more engineering designers think much of reliability-based design and multi-
disciplinary design optimization (MDO) when they develop a complex system.
However, direct integration of reliability-based design and MDO may present
tremendous implementation and numerical difficulties (Agarwal et al. 2003).
How to design a complex product with integrated optimization in design
process considering the reliability is an important problem demanding urgent
solution. Especially to the complex control system, its performance is related to
the structure, dynamics, frequency width, power matching and reliability whose
integrated performance depends on several disciplines such as mechanics, kine-
matics, dynamics, cybernetics and reliability, so reliability-based MDO facil-
itates considering the multidisciplinary design to achieve the integrated
performance optimization. Whereas the traditional design of control system
always concerns some disciplines while ignores others, it may lead to design
conflict among different disciplines. For example, traditional engineering
design of aircraft actuation system is carried out on static parameters (cylinder
geometric size, servo valve and hydraulic power selection) and dynamic perfor-
mance (frequency width, amplifier coefficient and control parameters), but it
seldom considers its weight, cost and reliability. If the initial designed para-
meters can not meet the requirement of other disciplines, the compensating
redesign process likely results in waste of resources and extension of the design
cycle, and frequently pushes to the limits of design constraint boundaries, leav-
ing little or no room to accommodate uncertainties in system design. So the
field of MDO has seen rapid advancement from uni-discipline design (Xiaoping
et al. 2008) in mission definition stage, conceptual design phase and preliminary
design phase that can aid system engineers to identify interactions among
different disciplines and improve integrated performance of comprehensive
product. MDO allows designers to incorporate all relevant disciplines simulta-
neously since it can exploit the interactions between the disciplines (Michael
et al. 2004). Practically, it is really difficult to realize multidisciplinary optimi-
zation for comprehensive product design because participating disciplines are
intrinsically linked to one another. Furthermore, such an integrated implemen-
tation is also subjected to complexities introduced as a result of a large number
of design variables and constraints.
Several approaches for MDO have been developed during the 1990s, all of
which had as an objective the coordination of the interacting disciplines during
the design optimization process (Bennet et al. 1997). At the same time, the
increasing concentration on economic factors and the attributes known as the
Reliability-Based Collaborative Design Platform 69
ilities make it possible to achieve optimization in performance, manufactur-

ability, reliability and maintainability. So the innovative MDO framework and
software are mushrooming. In 1990 s, Dr.Mark B. Tischler developed a MDO
software-CONDUIT (Control Designers United Interface) in NASA, and
applied it into UH-60 helicopter and X-29 aircraft (Andersson et al. 1998).
Dr. Eric Hallberg, affiliated with American navy graduate school, developed
RFTPS (Rapid Flight Test Prototyping System) to realize the multidisciplinary
simulation (Hallberg et al. 1999). The latest MDO framework emphasizes on
automation, optimization and integration of design process, which is widely
used in engineering application such as aircraft, launch vehicle and generator
(Takahashi et al. 2005). In 2002, Jean Yues Trepanier and Abdulsalam Alzubbi
developed a MDO framework VADOR that compose of graphic interface, data
server, implementation server and data management system to realize the
interdisciplinary design (Abdulsalam et al. 2000). Subsequently, Weiming
Shen presented MDO environment-WebBlow, which adopted Web interface,
agent computational resources and data management based on HML (Shen
et al. 2002). It is reported that MDO can improve product quality, decrease cost
and shorten the design period to 1/5 the original.
Since system parameters are not necessarily deterministic, the system para-
meters (e.g. material properties, boundary conditions, loads etc.) should be
stated probabilistically (Robert and Mark 2000), this class of problems is called
reliability-based multidisciplinary optimization. Reliability design method used
to calculate the failure probability or reliability of current point (only a design
point), and reliability-based optimized design method is not only to identify the
nature of the design results largest, smallest or more goals, but also to meet the
constraint conditions when it comes to the minimum reliability (or maximum
failure rate).Therefore, in order to create a problem of the reliability-based
optimization design, the general characterization of optimal design will require
some changes, such as adding some random variables, and change the qualita-
tive constraint to be random reliability constraint. R. Suess reliability-based
multidisciplinary design optimization randomizes the system parameters, such
as material performance, boundary condition and load, and optimize them with
integrate MDO method under uncertain factors (Sues and Cesare 2000). Xiaop-
ing Du presents concurrent sub-space uncertain analysis method to design a
system with Monte Carlo method (Du and Chen 2005). Dr. Harish Agarwal
utilized response surface to realize the multidisciplinary design optimization
(Harish and John 2004). D. Padmanabhan did some research work on relia-
bility-based optimization with approximate technique (Padmanabhan 2003).
Although the existing reliability-based MDO methods aforementioned are
effective in system multidisciplinary design, they pay more attention to the
component designed parameters whereas the system performance is seldom
taken into account.
Hydraulic actuation system is an important element of aircraft whose
performance and reliability influence its flight character and track cap-
ability directly. How to design an optimal hydraulic actuation system
70 S. Wang et al.
becomes bottleneck of flight control system when redundancy and fault-

tolerance are adopted in system design. This paper presents a reliability-
based collaborative design platform (RCDP) based on product database
for hydraulic actuation system whose trait is to carry out reliability
design, allocation and analysis traversing all the design process. In addi-
tion, the hierarchical decomposed multilevel optimization strategies are
utilized to achieve the more robust design including multiple disciplines
simultaneously and decrease the large computational effort. An effective
component database, engineering knowledge library and product data
management (PDM) are provided for designers to be interactive and
access them conveniently.
Based on actual commercial uni-discipline software, such as mechanical
design software Pro/E, reliability analysis software Relex, dynamic simula-
tion software ADAMS, hydraulic system analysis software AMESim, finite
element analysis software ANSYS and control software Matlab, RCDP
makes full use of various disciplines design and analysis tools to get more
reliable design results under graphical user interface, independent drive
engine, methods of rapidly integrated legacy and suitable computing envir-
onment. Herein, the knowledge and information of various disciplines
should be managed properly so as to achieve the entire optimization in
product life cycle design.
The outline of the paper is as follows. Section 2 introduces the idea and
design flow of RCDP, in which the probabilistic design, interactive interface,
multilevel optimization, data management and realization of RCDP are dis-
cussed. Section 3 provides the application of hydraulic actuation system that
indicates the effectiveness of RCDP. We present discussions and conclusions in
Section 4.
2 Reliability-based Collaborative Design Platform
An overview of the reliability-based collaborative design platform is shown

in Fig. 1 wherein reliability oriented and performance-based are considered
synchronously based on related commercial softwarecalled RCDP.
RCDP was designed to facilitate rapid development of complex hydraulic
servo system considering reliability and achieve integrated performance
optimization.
Generally, users should be up on the design specification and requirement
when they start to design a product that depends on several condition and
constraint. RCDP provides some standard scheme in engineering knowledge
library; designers can select them to obtain their preliminary design. If the
preliminary design parameters pass the back-check in other disciplines includ-
ing reliability, we can get the final system design and generate prototype to
verify its performance. Once the all performance meet the design specification,
Start
Input the
Design
Requirement Engineering
Knowledge
Library
Is There any N
Reliability Distibution
Standard Scheme?
Y Mechanical Product Design Electrical Product Design
Preliminary Component Protel

3-D Model by Pro/E
Design Library
Optimization
ANSYS Modeling Reliability

Dynamics EDA
&Analysis Prediction Optimization
by Simulation
by Relex
Reliability by ADAMS
Check the Reliability NESSUS
Parameters Prediction
Fulfill the Local
Y N
Subsystem
Fulfill the Local Performance?
Y Subsystem N
System Design
Performance? Y
Y
Fulfill the N Hydraulic System Reliability
Performance ? System Modeling
Simulation by Prediction Optimization
by Matlab
AMEsim by Relex
Y
Y Fulfill the Performance? N
Y
The
Final
System
Virtual Reliability
End
Prototype Assessing
Fig. 1 Overall description of RCDP
RCDP provides the reliability assessment report to accomplish the complex

product design. On the other hand, users can design the product indepen-
dently with abundant component library, convenient design platform and
necessary tools if RCDP can not provide the suitable design scheme. Because
the reliability-based design method is quite different in component design and
system design, RCDP decomposes the design process of complex hydraulic
products into two levels, viz. component level and system level, in which the
probabilistic design is adopted by treating stochastic quantities as random
variables in component level and the performance-based analysis is exploited
to realize the design optimization considering reliability constraint in system
level. With the appropriate interface, it becomes possible to verify the design
variables in two different disciplines through exchanging information dyna-
mically, and model the propagation of uncertainty throughout the hierarchy
of elements that utilize the simulation-based technique to joint the relation-
ship between two levels. In the event of product design, multidisciplinary
reliability oriented is provided that consists of the following disciplines:
72 S. Wang et al.
mechanics (ProE), electronics (Protel), hydraulics (AMESim), cybernetics

(Matlab), dynamics (ADAMS), finite element analysis (ANSYS) and
reliability (NESSUS and Relex).
2.1 Collaborate Design Based on Probabilistic Distribution
Since the performance of complex hydraulic servo system depends upon several
disciplines which interact and conflict each other sometimes, the collaborative
design platform should support the multi-discipline design software, assemble
the subsystem with appropriate interface and manage the different kind of data
effectively. In hydraulic control system, servo valve and cylinder is typical
mechanical product whose performance depends on mechanics, dynamics and
reliability. A solution of integrated design is put forward based on Pro/E,
ANSYS, ADAMS and NESSUS, in which the 3D model is firstly designed
with CAD software package Pro/E according to static performance of hydrau-
lic servo system. Then transfer the designed geometrical parameters in *.igs
document to ANSYS to carry out the structural finite element analysis that
provides the automatic command flow recording function with generating
*.lgw document. Meanwhile, ADAMS can load the 3D model through the
*.igs file to analyze its dynamics. Considering the product parameters are not
necessarily deterministic, the reliability design software, named NESSUS, is
adopted to perform probabilistic analysis for designed structure. As a result of
command flow from ANSYS, we can get the determinate design parameters,
then select the random variable and probability density from list and establish
response function of individual parameters. Then select the random variables
and their probability density from built-in list and define the failure probability
expression with stress-intensity interference model. Herein, the response func-
tions are utilized to build the bridge of individual parameters and adaptive
importance sampling (AIS) is exploited to calculate the failure probability of
product whose sampling space is defined using a limit-state boundary. For
instance, the tensile load acting on an uniform bar (L) and the cross sectional
area of the bar (A) are assumed to be independent random variables wherein the
corresponding probability distribution function submits normal distribution as
follows:

1 1 L L
fL p exp
2pL 2 L
(1)
1 1 L A
fA p exp
2pA 2 A
where L ; L denote mean value and standard deviation of the variable L

respectively; L ; L are mean value and standard deviation of the variable A
respectively.
The force in bar is:

1 1 P P
P fL; A L = A; fP p exp
2pP 2 P
s
(2)
2 @fL; A 2 2 @fL; A 2
P L = A ; P L A
@L @A
If material strength of the bar is s and its probability density function isgs,
the failure probability of the bar can be described as:
2 3
Z1 Z1
F P s P50 1 f P 4 g s ds5 dL (3)
1 s0
Select response model with user-defined as
P fL; A (4)
Let the coupling function as

s P
z p (5)
2s 2P
Then the failure probability of product is:
Pf 1 z (6)
It is easy to obtain z through referring Normal distribution table. With

NESSUS and ANSYS, we solve the problem as follows.
Firstly, build the bar finite element model in ANSYS, and edit the command
flow *.lgw file.
Secondly, describe the problem statement in NESSUS as:
Pf Pfs L50g
(7)
P fL; A
Select the P fL; A expressions response model as *.inp file editing from
the ANSYS fore-defined *.lgw file, then mapping the variable parameters to the
*.inp. NESSUS controls ANSYS to calculate failure probability under updat-
ing the input parameters (viz. L and A) with the sampled values by AIS and
extracting relevant results (viz. stress) from ANSYS output *.rst file.
Finally, NESSUS can provide the statistic report. Figure 2 shows the opera-
tional principle and data flow of NESSUS.
In NESSUS, the start point of reliability design is the product failure mode
that is subjected to design parameters and failure criteria. User can select the
74 S. Wang et al.
User
Execution Random Variable

Command and Inputs and
input/output Files Mappings
Input file
NESSUS
(ansys-job.inp)
Restart
Database
Output File
NESSUS (file.rst,file.rth)
Fig. 2 NESSUS operational principle
probabilistic distribution from provided list for design parameters, and estab-
lish the status equation or response function to calculate the failure probability
of product. The executive process of collaborative design based on probabilistic
distribution for mechanical product with Pro/E, ANSYS and NESSUS is
shown in Fig. 3.
Figure 4 shows the main interface of RCDP, in which the design methods are
listed on the left such as component design, system design, reliability analysis
and system optimization. Users are convenient to select or design the hydraulic
product with 3D modeling based on component database and engineering
library in the middle field and add the designed component in defined format
to the database when the designed product meets all the design requirements.
Once the product design is accomplished with mechanical package or electronic
package, users can design the hydraulic system with collaborative design
method considering electronics, dynamics, cybernetics and reliability synchro-
nously. In system design, we focus on the performance-based design under
multiple disciplines, so system optimization strategy is adopted to balance the
design requirements in different fields, in which the reliability analysis is exe-
cuted in whole design process.
Since much attention has been paid to the development of procedures to
couple powerful multidisciplinary optimization techniques with probabilistic
design method, it is easy to achieve the integrated optimization solution based
on different disciplines after collaborative design. For the system design, we
adopt current state-of-art optimization that emphasizes the performance by
cooperating control performance software-Matlab with reliability analysis soft-
ware-Relex. The Relex Reliability Studio integrates a suite of reliability analysis
modules encompassing reliability prediction, Reliability Block Diagram (RBD),
Failure Mode Effect Analysis (FMEA), Fault Tree Analysis (FTA), maintain-
ability prediction, Markov Modeling and LCC. It can calculate reliable
Maximum
RCDP Redesign
Stress, Size,
Framework
Reliability, etc.
The Fault
Determine the Determine the
Tree
3-D System Failure Random Design
Trail.txt, Structure
Modeling Criteria Parameters
*.drw g-function.
by Pro/E
With Dependent or Independent
ANSYS Finite Random Variables, Constants ,and
Reliability Statement :
Element Numerical Functions (Defined by
Analytical Expression
Modeling Numerical Model)
Fortran Syntax
*.log Edit
Dependent Distribution type
Files Mean, Standard Deviation, etc.
Random
Variables Change Parameter Values
ANSYS Assign Numerical Assign *.inp Files Edited from *.log Files
Analysis Response Model Assign *.rst Files to Extract Response
Updated Renewed Variable Locate Random Variables in the *.inp files

*.rst Files *.log files Mapping for Further Update.
Sample
Performance Random Response Specify Load Case ,Response Location in the
g-Function Statistic and Variables Selection Framework, and Concerned Time.
Visualation
Parameters
Probabilistic
Perturbation
Analysis
Analysis
Meet the N
Demand?
Generate ReliabilityReport
Fig. 3 The collaborative design process of mechanical product
probability or failure probability and obtain the weak item of the analyzed system
with appropriate algorithm. In addition, Relex can describe the system perfor-
mance degradation and repairable process dynamically, so the reliability-based
collaborative design platform can realize the reliability identification in whole
design process.
2.2 Interface Between Multidisciplinary Software

In the process of collaborative design of complex product, we obtain the design
parameters that are rooted from specific discipline, and then we may ask the
question: Will this design perform well under other discipline? To answer this
question, the convenient way is to transfer the design parameters to other
discipline software to verify its performance, which can guarantee well perfor-
mance in different discipline for designed product. Fortunately, most of the
76 S. Wang et al.
Fig. 4 Main interface of RCDP
commercial design and analysis software have interface with other software
package so as to improve their collaboration capability. The interface makes it
possible to exchange information, access public data and realize the integrated
optimization among different disciplines. Even if there is no interface between
two packages, it is also easy to develop executable code based on input and
output application file mode. Users can export files designed by one discipline
and import it into another discipline environment to check if the designed
parameters satisfy the performance or not. Developers can create their own
relationship on the basis of their design demand that makes the change of the
model easier. So it is important to obtain the well designed interaction interface
between two disciplines that can exchange data dynamically and save the
developing time effectively in complex product design. Table 1 lists the standard
interface and self-developed interface between current commercial software
packages.
Taking the Relex and Protel as an example, although both packages dont
provide mutual programmable interface, we establish the medial files via Excel
as their bridge. User can design the electronic or electrical circuit with Protel,
and export the components list in Excel format. Then the *.xls file is standar-
dized by the MDO platform and guided to Relex to predict its reliability (shown
in Fig. 5). It is obvious that the MDO platform is employed to reduce the
computational expenses significantly with the appropriate interface that makes
information exchanging easier.
Table 1 The interface of commercial software

No. Software1 Software2 Interface Implementation
1 AMESim Matlab There are the AMERun-MATLAB interface, AMESim to
Simulink Interface, Simulink to AMESim Interface in
the AMESim suite.
2 AMESim iSIGHT AMESim can create files directly readable by iSIGHT
using the MDOL format.
3 VC Matlab Visual C++ employs Matlab as a computation engine by
Component Object Model (COM) on Windows
4 ANSYS NESSUS NESSUS employs *.log files and variable mapping to
control ANSYS to operate.
5 ANSYS iSIGHT iSIGHT imports ANSYS code as simcode.
6 VC iSIGHT iSIGHT imports *.exe as the simcode.
7 VC SQL Visual C++ accesses SQL server by Open
Server Database Connectivity driver (ODBC)
8 Excel Relex Relex outputs and inputs part list in *.xls file.
9 Excel Protel ProtelDXP outputs part list in the .xls file.
10 Pro/E ADAMS ADAMS can load *.igs files generated by Pro/E.
11 Pro/E ANSYS ANSYS can load *.igs files generated by Pro/E.
12 ADAMS Matlab Output * .mdl files to Matlab/Simulink
13 ADAMS iSIGHT iSIGHT imports Adams analysis code as simcode.
14 Matlab iSIGHT iSIGHT embeds Matlab into the Insert toolbar.
Fig. 5 Medial interface between Protel and Relex

78 S. Wang et al.
2.3 Optimization Based on Hierarchical Decomposition
If hydraulic servo system is complex enough, design optimization can be

accomplished by decomposition. The system is divided into several subsystems,
and the subsystems can be divided into components and so on. Hierarchical
decomposition facilitates employing decentralized optimization approaches
that aid systems engineers to identify interactions among elements at lower
levels and to transfer this information to higher levels, and has in fact become
standard design practice, as evidenced by the organizational structure of engi-
neering companies. In RCDP, the PDM is exploited to provide user necessary
information about component and engineering experiences.
Multidisciplinary design optimization is an effective method and tool for
comprehensive engineering system design whose objective is to find the overall
optimal solution by means of the collaborative effect caused by multidiscipline
interaction and shorten the design period concurrent computation. While mul-
tidisciplinary design optimization is also faced with much challenges such as
computation complexity and difficulty of information exchange, and some
method and technology such as system decomposition, approximate analysis
and totally sensitive analysis. Recently, collaborative optimization is consid-
ered as a promising MDO method, which aims at complex system design,
distributed, multi-level optimization. Its motivation is dividing multidisciplin-
ary complex system design problem into subjects (or subsystems) design opti-
mization problems, and reducing the size of optimization problem, through
system-level constraints to coordinate sharing design variables and coupling
design variable between disciplines shown in Fig. 6.
In Fig. 6, the system consists of 3 levels, viz. level 0 (system), level 1
(subsystem) and level 2 (component). Suppose the local design parameters of
component iare rij i 1; 2; n; j 1; 2; ni in level 2, in which jmeans the
number of component. Herein the parameters ~ r1 ; ~r2 ; ~rq come from level 2 are
exported to level 1 to affect the subsystem design that is named shared para-
meters. If xij i 1; 2; m; j 1; 2; n0m denote the local design parameters
Level i=0 System y1 ,... yl
x~1 ,... ~
xp
Level i=1 Subsystem j=1 Subsystem j=2 Subsystem j=m
x1,1 , ... x1n1 x2,1 ,... x2 n2 xm ,1 ,... xmnm
~
r1 ,... r~q
Level i=2 Component j=1 Component j=2 Component j=n
r1,1 ,... r1n1 r2,1 ,... r2 n2 rn ,1 ,... rnnn
Fig. 6 Hierarchical decomposition of complex system

of subsystemi, the design procedure must balance both local variables and
shared variables to achieve well performance. Similarly, the system design
should consider the influence of the shared parameters of x~1 ; x~2 ; x~P come
from level 1 and local design parameters y1 ; y2 ; yl of system. Hence, both
local design and interaction between lower level and higher level should be
integrated in product design to obtain the optimization design results under
multilevel hierarchy.
Directing to the complex product in real application, this paper presents an
optimization method in the multilevel hierarchy. The original optimization is
carried out from component viz. the lowest level to get the designed variables, in
which some of the variable values pass up to higher level as shared design
variables. In the higher level, both local design variables and shared variables
coming from lower level are considered to achieve the optimization wherein
minimize the gap between what higher-level elements want and what lower-
level elements can. Through numbers of iterative optimization between the
higher-level and lower-level, we will eventually find an optimal design of the
consistency based on the state variables coupling and design variables sharing.
Suppose that i and jdenote the level and element respectively, the optimal
objective function of system can be described as:
x1 ; x~P ; y1 ; yl
min f~ (8)
Subject to

gij ~
x1 ; ; x~P ; y1 ; ; yl 0
(9)
hij ~
x1 ; ; x~P ; y1 ; ; yl 0
with
(
xij fij xi;1; ; xin0j ; r~1 ; r~q
(10)
yk fy1 ; ; yl ; x~1 ; ; x~P
where the vectors rij ; xij and yk denote the local design variables in each level;
r~1 ; r~q and x~1 ; x~P indicate the bi-level shared variables; gij and hij denote the
design inequality and equality constraints respectively.
In RCDP, iSIGHT is adopted to realize the multidisciplinary design optimi-
zation that can deal with the interaction among disparate disciplines and
balance the target difference to achieve the integrated optimization results.
Figure 7 shows how to balance the interaction between cybernetics (Matlab)
and hydraulics (AMESim) with iSIGHT.
Once we accomplish the design of hydraulic servo system with AMESim
according to the static and dynamic requirement, we can get the parameters of
servo valve, cylinder and displacement sensor such as flow coefficient KQ , flow-
pressure coefficient Kc and area of cylinder A. Introduce them to Matlab to get
80 S. Wang et al.
Hydraulic
PID Controller System
Component Model
by Matlab Simulation
by AMESim
In *.bat
Format as
Simcode
*.txt A New
*.txt as
as Input Analysis
Output Files
Files Task
Define Objective Fuction,

Select KP,KI,KD,etc as Optimization Constraints
Design Varibles Plan
Select Optimization
Algorithms
The Optimum
Design iSIGHT
Optimization
iSIGHT Run-time Interactive Route Custom an iSIGHT Programe Route
Fig. 7 Optimization between cybernetics and hydraulics with iSIGHT
the system transfer function. In order to satisfy system control performance,

appropriate PID controller should be designed, in which the proportional
coefficient KP , integral coefficient KI and the differential coefficient KD can
be adjusted dynamically. If it is impossible to achieve the system performance
requirement under current hydraulic components no matter how to adjust the
PID variables, we should redesign the hydraulic components dynamic para-
meters with AMESim. On this condition, it is necessary to balance the design
variables in two disciplines. Herein, we select the optimization software
iSIGHT to realize the design optimization. We select KP ; KI ; KD as design
variables and set the optimization of hydraulic actuator as a new analysis
task. Defining the response time as objective function and reliability as con-
straint, iSIGHT can recommend suitable algorithm to obtain the optimal
KP ; KI ; KD based on cybernetics. If the response time of system or reliability
can not satisfy the requirement, redesign in hydraulic discipline and cybernetics
discipline will be implemented until the integrated performance is fulfilled.
Especially, iSIGHT can run other software and launch the iteration process
automatically by assigning input parameter file, output file, simulation code
and control flow. Users are convenient to choose the CAD/CAE tools, math-
ematical and programming tools, internal developed codes and the intelligent
design engine, make these tools assembly and establish the integrate analysis
process through a simple graphical interface. iSIGHT enables the data to
transfer from one tool to another in a seamless manner operation. For instance,
iSIGHT can run Pro/E with the notepad type file and run ANSYS with the *.igs
type file. During Pro/E running, it will automatically record every operation
step and save the command flow to a trail.txt file. iSIGHT automatically passes
on the Pro/E geometrical model to ANSYS, and reads the performance para-
meters such as the greatest stress, the maximum deformation from ANSYS
output *.rst document. The intelligent design engine has four modules:
experimental design, optimization algorithms, quality engineering methods
and approximation method that can be used individually or synthetically. For
optimization, users can adopt the optimizing plan recommended by iSIGHT
which is the most suitable for the design problem, or combine the iSIGHT
optimizing algorithm or develop internal programs to get the most reliable and
robust strategy according to the nature of the problems and professional
experience.
2.4 Data Management
Database is the center of the platform, and it is the foundation of other function
modules. According to the operational process shown in Fig. 1, RCDP is
supported by two databases, viz. the engineering knowledge database and
system component database. These two databases play very important role in
the entire design process. They not only provide useful design information to
designers but also attribute and control method of components, and new items
can be added into the database on finished design. So the databases can be
continually expanded to facilitate future design. Data can be uploaded or
downloaded between local resource library and center database by the transfer
function module of database.
Figure 8 shows the database and PDM of RCDP, in which the knowledge
database can be divided into three categories:
1. Model knowledge database Toward some subsystems of special application,
it is easy to establish its design template whose design parameters construct
its model knowledge database.
2. Engineering design knowledge database It is obtained based on engineering
experience of research object such as typical displacement servo system, force
servo system and velocity servo system. Such databases mainly include
knowledge of design characteristic, design sequence, design experience dur-
ing engineering design that can provide recommended primary scheme for
designer at the beginning of design process.
3. Parameters design knowledge database This database includes knowledge
of interactive parameters and scopes of design parameters. Such database
will provide basis to multidisciplinary design parameters optimization.
In the beginning of new product design, the designers are accustomed to
consult the similar product design. If there is ready-made design in RCDP, users
may generally select a specific scheme by the aid of engineering database
82 S. Wang et al.
Result of the Analysis Guiding Design
CAD Tools Data Transfer Tool Collaborative Design

CAE Tools CAD Model Transfer Simulation Modeling
Control Flow Data flow
PDM System
File Version Staff Competence Engineering System

Management Management Design Database Database
Fig. 8 Database and PDM
according to the design requirement. Otherwise, the designers have to follow the
reliability-based design process to realize the multidisciplinary design optimiza-
tion. In this case, the system will be divided into subsystems by the appropriate
reliability allocation, and then the collaborative design based on component
database is carried out. After verifying the performance and reliability, the new
successful design application can be accessed into database with standard data
format.
The function of PDM is to manage the product database and product R&D.
Collaborative design platform can take advantage of powerful data integration
and management capabilities to achieve effective management to design activ-
ities which including workflow simulation, simulation dedicated database and
model libraries, design files. PDM system is the best CAD/CAPP/CAM inte-
gration platform, which helps a team or an enterprise to achieve information
integration, process integration and functional integration in heterogeneous
and distributed computing environment. In RCDP, SQL Server 2000 is adopted
to integrate and manage the different defined data that can interactive transfer
with data accessing technology. Table 2 shows the function parameters descrip-
tion in RCDP.
Table 2 The function parameters description

No. Character name Data type Description
1 ParaId int Parameter number, PK
2 EngId int Engineering number, FK
3 ParaName Varchar 20 Parameter/component/scheme name
4 ParaValue decimal Parameter value
5 PartNo int Component number (FK)
6 ParaImage image Graphical value
7 ParaType Varchar 20 Parameter type: system, structure and reliability
8 ParaDescribe Varchar 20 Parameter/component/scheme name
2.5 Realization of RCDP
Aimed at the characteristics of hydraulic servo system, a solution of integrated

design architecture is put forward aforementioned, which is multidiscipline
oriented and hierarchy. The collaborative design and simulation of all dis-
ciplines is performed concurrently whose results are transferred into the
unified PDM database for future applying. While some tools provide the
interfaces for collaborative simulation listed in Table 1, it is convenient to
check whether the shared variables which are determined by the uni-discipline
satisfy with the requirement in the other disciplines or not. So the integrated
optimization should be executed to guarantee the optimal performance and
reliability.
RCDP is designed based on VC++ 6.0 and SQL Server 2000, which can
carry out distributed collaborative design including mechanical design,
electronic design, reliability design, hydraulic design, control design and
unified data management. Herein, iSIGHT plays an important role in
collaborative integrated design with graphical interface and convenient
assembly. Figure 9 shows the integrated software, data standard and
operation principle wherein STEP standard is adopted in collaborative
design platform.
Fig. 9 The data standard of RCDP

84 S. Wang et al.
3 Application of Hydraulic Actuation System
Hydraulic actuation system (HAS) is a hydraulic displacement control system

that can manipulate the rudder according to the flight profile. To design this
kind of system, the designers will use many disciplines knowledge such as
mechanistic, kinematics, dynamics, cybernetics, hydraulics and reliability.
Figure 10 shows the structure of hydraulic actuator whose design requirement
is listed in Table 3.
With RCDP, the designer can design HAS according to the Pro/E-
>ANSYS->NESSUS sequence by aid of the component database and engi-
neering database as shown in Table 4. The dimension model of actuation is
shown in Fig. 11.
After finishing the components design we will enter the stage of system
design. After deduction we can get the transfer function of HAS:
KV
Gs 2 (11)
s !2 2
s
!h
h
s 1
h
where KV KA Ks Kq Kf =A is the gain coefficient; Ka means amplifiers coeffi-

cient; Kf denotes feedback coefficient; Ks expresses coefficient of servo valve; Kq
shows the flow coefficient of servo valve; !h and h are inherent frequency and
damping of dynamical structure respectively.
Hydraulic Actuation System
Instruction
Servo Valve
Input Signal +
Signal Adjust

Hydraulic Power
Sensor
Fig. 10 Structure diagram of hydraulic actuation system
Table 3 Design requirement of hydraulic actuation

Number Meanings Symbol Design requirement
1 load F 294000 N
2 pressure P 7 MPa
3 distance S 0.1m
4 velocity V 0.12m/s
5 reliability R(t) 0.999
Table 4 List of components

Code No. Name View Reliability
ZDQ0603001 piston 0.99987
ZDQ0603002 earring 0.99988
ZDQ0603003 sleeve 0.99991
ZDQ0603004 seal 0.99999
ZDQ0603005 seal 0.99999
ZDQ0603006 bearing 0.99999
ZDQ0603007 bearing 0.99999
Fig. 11 Dimension model of actuator
Firstly, we will define the variables to be design and optimize. Let X1 is

Ka ; X2 is Kf ; X3 ; X4 ; X5 are PID controller coefficient Kp ; Ki ; Kd respectively,
X6 expresses piston external diameter, X7 is piston internal diameter, X8
indicates load force and X9 denotes hydraulic cylinder material strength .
Herein, X6 ; X7 ; X8 are sharing design parameters which participate in the
two-subsystem design optimization process.
Secondly define the objective function of HAS as:

Rise time Tr 50:05s
(12)
Amplitude Margin 46dB
Constraint:
86 S. Wang et al.
8
> reliability40:9999
>
>
>
> 130 X1 150
>
>
>
>
>
> 1 X2 1:2
>
>
< 6:5 X3 7
(13)
>
> 8 X4 12
>
>
>
> 2 X5 2:5
>
>
>
>
>
> 0:1 X6 0:3
>
:
0:01 X7 0:15
With reliability-based collaborative design and hierarchy optimization, we

can get the optimal values listed in Table 5.
Table 5 Optimizaition results of HAS

Parameter Min value Max value Step Result of optimization Unit
X1 130 150 0.5 149.5
X2 1.00 1.20 0.01 1.12
X3 6.5000 7.0000 0.0005 6.8555
X4 8.0000 12.0000 0.0005 10.0000
X5 2.0000 2.5000 0.0005 2.4065
X6 0.10 0.30 0.01 0.16 m
X7 0.01 0.15 0.01 0.08 m
X8 294000 N
X9 280 MPa
After optimization, we can obtain the system step response and bode chart of
open-loop curves as shown in Fig. 12.
At the same time, we can obtain the fault tree of HAS and provide the
reliability prediction shown in Fig. 13.
With the computation algorithm, we can get the reliability of HAS is
0.999999933 in 10 hour that not only satisfies the design requirement but also
approaches the real condition.
4 Summary and Conclusion
This paper presents a reliability-based collaborative design platform based on

corresponding commercial software and appropriate interface, in which the
design optimization of hierarchically decomposed multilevel systems is carried
out under uncertainty. We extended the deterministic formulation of product
design to account for uncertainties with structural reliability design software
Fig. 12 Step response of system
NESSUS, in which designer can select the probabilistic distribution according

to the product characteristics. We adopt inherent interface or develop the
medial file as interface between two packages that builds the bridge connecting
the interdisciplinary interaction and makes it easy to achieve integrated opti-
mization among multiple disciplines. In addition, we modeled the propagation
of uncertainty with shared parameters and run the software in given variable
domain iteratively with intelligent design engine of iSIGHT. With local
88 S. Wang et al.
Fig. 13 Fault tree of HAS
variables and bi-level shared variables, the performance-based design optimiza-

tion can realize. Finally, the reliability analysis with Relex is carried out to
assess the system reliability, maintainability and availability to guarantee the
integrated performance optimization.
The following issues deserve special attention:
1. Exploiting the reliability-based collaborative design method to design a
complex product can fully consider the interaction among different disci-
plines and achieve performance-based integrated optimization.
2. Making the most of the inherent interface or provided medial files in RCDP
can facilitate designer to obtain the performance verification of other dis-
cipline and achieve the interdisciplinary design optimization.
3. Effective database and PDM can reduce the developing period and make the
designer easy to obtain the reliable, robust and best design scheme.
Application of HAS indicates that RCDP can improve the design effective-
ness, shorten the development period, economize designing resource and
achieve the reliability evaluation.
Acknowledgments Thank the support of Chinese Education Ministry Program 985, Pro-
gram of Introducing Talents of Discipline to Universities and Program for New Century
Excellent Talents in university (NCET), Program of Introducing Talents of Discipline to
University (Program 111) and Beijing Teaching Innovation Program.
References
Abdelhalim S, Ahmed C and Pierre S (2004) Comparison of the Various PDM Control
Modes, 2004 IEEE International Conference, Vol. 2, pp. 810; pp. 574579.
Abdulsalam A, Amadou N, Babak M, Francois G, Benoit O, and Jean Y T (2000) On the use
of JAVA and RMI in the development of a computer framework for MDO, AIAA-2000-
4903, Symposium on Multidisciplinary Analysis and Optimization, 8th, Long Beach, CA,
p. 4903.
Agarwal H and John E R (2002) Reliability Based Design Optimization for Multidisciplinary
Systems Using Response Surface[R]. AIAA Paper, p. 1755.
Agarwal H, Renaud J E and Mack J D (2003) A decomposition approach for reliability-based
multidisciplinary design optimization, 44th AIAA/ASME/ASCE/AHS/ASC structures,
structural dynamics and materials conference, Norfolk, VA, p. 1778.
Alexandrov N M and Kodiyalam S (1998) Initial results of an MDO method evaluation study,
AIAA Paper, p. 4884.
Andersson J, Krus P and Nilsson K(1998) Optimization as a support for selection and design
of aircraft actuation systems, Seventh AIAA/USAF/NASA/ISSMO Symposium on
Multidisciplinary Analysis and Optimization, St. Louis, Missouri, pp. 24.
Bennet J A, Botkin M E, Koromilas L, Neal R V and Zwiers R I (1997) A Multidisciplinary
Framework for Preliminary Vehice Analysis and Design. The ICASE/NASA Langley
Workshop on Multidisciplinary Design Optimization, pp. 1321.
Dapeng W and Greg F N (2005) Collaboration Pursuing Method for MDO Problems, AIAA
Paper, p. 2204.
Dhanesh P, Harish A, John E R and Stephen M B (1997) Monte Carlo Simulation in
Reliability-Based Optimization Using Approximation. Proceedings of the Fourth Inter-
national Symposium on Uncertainty Modeling and Analysis (ISUMA03), pp. 16.
Du X and Chen W (2001) A most probable point-basch and A.S.ed method for efficient
uncertainty analysis, Journal of Design and Manufacturing automation, pp. 4766.
Du X P and Chen W (2002) Efficient Uncertainty Analysis Methods for Multidisciplinary
Robust Design, AIAA Journal, Vol. 40, No. 3, pp. 545552.
Hallberg E, Komlosy J, Rivers T, Watson M, Meeks D, Lentz J and Kaminer I (1999)
Development and applications of a rapid flight test prototyping system for unmanned
air vehicles, Instrumentation in Aerospace Simulation Facilities, ICIASF 99.18th Inter-
national Congress, pp. 26/1.
Harish A and John E R (2004) Reliability based design optimization using response surfaces
in application to multidisciplinary systems, Engineering Optimization, Vol. 36, No. 3,
pp. 291311.
Jasmin T, Jean Y T and Christople T (2004) Interaction and Multidisciplinary Design
Optimization of a Simplified Gas Turbine Model Using Perl and iSIGHT, AIAA Paper,
p. 4425.
Jean Y T, Francois G and Ozel B (2002) The Architecture and Design of Distributed Frame-
work for Multidisciplinary Analysis, AIAA Paper, p. 597.
Mitteau (1999) Error evaluations for the Computation of Failure Probability in Static
Structure reliability Problems, Probabilistic Engineering Mechanics, Vol.14,
pp. 119135.
Michael K, Zissimos P Mourelators, Panos Y and Papalambros (2004) Design optimization
of hierarchically decomposed multilevel systems under uncertainty, Proceedings of
DETC04 ASME 2004 design engineering technical Conference and Computer and Infor-
mation in Engineering Conference, Salt Lake City, Utah, pp. 113.
Mohammand A, Youngwon S, Robert H Sues and Y-T Wu (2002) A framework for relia-
bility-based MDO of aerospace systems, AIAA/ASME/ASCE/AHS/ASC structures,
structural Dynamics and material Conference, Denver, Colorado, p. 1476.
90 S. Wang et al.
Nichols F B and John R O (2005) Evaluation of Multidisciplinary Optimization (MDO)

Techniques Applied to a Reusable Launch Vehicle, AIAA Paper, p. 707.
Padmanabhan D, Agarwal H, Renaud J E and Batill S M (2003) Monte Carlo simulation in
reliability-based optimization using approximation concepts, Fourth International Sym-
posium on Uncertainty Modeling and Analysis. ISUMA, pp. 298303.
Parashar S and Bloebaum C L (2005) Decision Support Tool for Multidisciplinary Design
Optimization (MDO) using Multi-Domain Decomposition. AIAA Paper, p. 2200.
Parviz M Z, Vassili V T and Alastair S W (2004) Use of Global Approximations in the
Collaborative Optimization Framework, AIAA Paper, p. 4654.
Robert H S and Mark A C (2000) An Innovative Framework for Reliability-based MDO,
AIAA Paper, p. 1509.
Shen W M, Wang Y and Li Y S(2002) A Web/Agent Based MDO Environment, 7th Inter-
national Conference on Computer Supported Cooperative Work in Design, Brazil,
pp. 2533.
Sues R and Cesare M (2000) An Innovative Framework for Reliability-based MDO, AIAA
Paper, p. 1509.
Takahashi H, Turner H and Birney H (2005) A Multi-Disciplinary Survey of Advanced
Subsonic Tactical Cruise Missile Configurations, AIAA Paper, p. 709.
Tatsuya S and Young W K (2003) Reactive scheduling for large-scale manufacturing systems
based on hybrid Petri Net Modeling, SICE Annual conference in Fukui, Japan,
pp. 244249
Tischler H (2001) Optimization and comparison of alternative flight control system design
methods using a common set of handling-qualities criteria, AIAA Paper, p. 4266.
Wang S (2000) Engineering Reliability, Beijing Aeronautics and Astronautics Press
Wan-jun Wand, da-you Liu, ni-ya Li and quan Liu (2004) Product lifecycle based on j2ee in
PDM, Proceedings of 2004 International Conference, Vol. 2, pp. 872877.
Wu Y T, Millwater H R and Cruse T A (1999) Advanced probabilistic structure analysis
method of implicit performance functions, AIAA Journal, Vol. 28, No. 9, pp. 16631669.
Xiaoping D, Guo J and Beeram H (2008) Sequential optimization and reliability assess-
ment for multidisciplinary system design, Structural and Multidisciplinary Optimization,
Vol. 35, No. 2, pp. 117130.
Xiaoping D and Chen W (2005) Collaborative reliability analysis under the framework of
multidisciplinary systems design, Optimization and Engineering, Springer Netherlands,
Vol. 6, No. 1, pp. 6384.
Yang H and Jaroslaw (2001) Multidisciplinary design optimization of a full vehicle with high
performance computing, AIAA Paper, p. 1273.
The 4+1 Dynamic Management System
of Lifecycle
Junichi Yagi and Eiji Arai
Abstract The life cycle management of a complex product like a constructed

product needs to reconsider its engineering framework. The recent development
of science, specifically physical and brain research provides us with an ingenious
clue. The virtual control of time is now within our scope. This manuscript
suggests its possible engineering implementation.
Keywords Lifecycle management Virtual time machine Sensor network

Unification of parts and packets Complexity of system
1 Introduction
Control of construction process must deal with a bulk of scheduling elements

that are produced by multiple heterogeneous players, contractors, sub-contrac-
tors, and parts makers. Their scheduling elements are necessarily intermingled
and linked with each other. The whole schedule is managed and adjusted as one
giant flow of processes. One small change in a tiny part of the whole schedule
may possibly propagate its influence into the whole. Every such complex system
which is non-linearly linked often shows some chaotic behavior, and at worst
leads to a catastrophe. Upon this peculiar condition, construction process
management is conducted on site holistically at every moment from the com-
mencement and completion of a construction project. There are varieties of
causes that force scheduling change. It is of critical issue when a signal of change
is dispatched within a complex system. If timing gets lost, disastrous conse-
quence may result with severe cost loss. It is also critical where a change occurs
among multiple players, as well as what happens.
Two to three million parts needed for a building, depending on how to count,
and yet the vast bulk of construction parts should be registered somehow in a
database, if the entire lifecycle management of a constructed product should
J. Yagi (*)
Shimizu Corporation, Institute of Technology, 3-4-17 Ecyujima, Koto-ku, 135-8530
Tokyo

92 J. Yagi, E. Arai
ever be achieved. Parts comparability with this parts-level information retrieval

will make possible replacement of parts in a building to its full extent, even
perhaps structural modules, not to mention replace of the infill. It will even-
tually erase the current dichotomy between renovation and newly-build in its
entirety.
The lifecycle of constructed products is quite lengthy, thirty years or longer.
The construction parts have various lengths of life-span, or different clocks,
days to years or to decades. So, the constructed products have temporal
hierarchies in addition to structural hierarchies. However, compared to con-
struction stage where the construction environment is dynamically changing
within a rather compact period, the time flows much slower at the operational
stage, gradually decaying and reaching to the point of threshold.
A proper engineering decision making is expected to be quite hard for the
lifecycle management of any constructed product of high complexity with quite
distinct characteristics over stages. This manuscript tries to suggest how to
attack this non-trivial issue.
2 Physical Background
The lifecycle of housing begins with construction by assembly of a bulk of

building components from bolts and nuts to a large span of columns and
beams. The resulted product is a complex physical hierarchy of hierarchies of
components, exerting directly or indirectly characteristic influences upon
other components both within and across the hierarchies, some stronger and
some weaker. Truly, it is tried in design to minimize couplings among the
hierarchies and between components within a hierarchy. Such architectural
effort is typically exemplified by modular construction, which joins structu-
rally and functionally modules with modules but not with the constituent
components within a module. The modules are encapsulated and relatively
independent so that they interact through a rather small window of inter-
action at their surface envelope. Much effort is extended along the same line to
decomposing of skeleton and infill, where structural skeleton like beams and
columns detach functionally from interior settings like equipments and fix-
tures so much as it becomes possible to replace the interior modules, while
keeping the structural skeleton intact. It seems very reasonable to keep the
skeleton in use longer, for the skeletons have longer clock for their lifecycle
compared to the infill. Keeping the skeleton longer fits the purpose of saving
energy and resources.
However, it is hardly believed that the modular independence remains
intact at the second stage of inhabitation or living stage, much longer period
for lifecycle, where the decaying process proceeds with full of interactions by
operations, human or otherwise. Such complex system often renders pro-
gressive deterioration in dissipation so that the modular independence will
The 4+1 Dynamic Management System of Lifecycle 93
be eroded away, and the essential interdependence among hierarchies and

components becomes dominant to determine the fate of the system for the
rest of the lifecycle. It may be said the second stage of living or operation is a
phase of melting pot of interactions. The flow overshadows the rigidity of
structures.
It is therefore not so appropriate to hold a presupposition of classical
mechanics of rigid body for consideration of dynamic transformation of the
complex system throughout lifecycle. It is true at construction stage most of
design follows the classical mechanics of rigid body. Some of structural
analyses already goes beyond it and employs nonlinear model for fracture
analysis of concrete, the coupling between the upper structure of building
and earth soil, not to mention the analysis of seismic or wind effects upon
building. It must go beyond further for modeling the dynamic flow of the
system transformations. Eddingtons argument (1958) [1] is quite suggestive
in that regard. Eddington, who argued that, in a purely algebraic approach
to physical phenomena, there are elements of existence defined not in terms
of some hazy metaphysical concept of existence, but rather that existence is
represented by an idempotent in the underlying algebra of transformations.
He suggested the physical phenomena are not so much as they exist but that
they persist. They persist not in the sense that they are static, but rather that
they continually transform into themselves (idempotent p*p= p, when it acts
on itself, it remains itself or it persists) [2]. Physical phenomena should be
viewed as relatively persistent phenomena under the underlying algebra of
ceaseless transformations. This view gives us much better representation of
the reality that nothing constructed is immune from transforming into other
states or decaying. Changing is more basic than existing (or persisting,
saying properly).
The temporal order of movement is thus the primary concern as long as the
management of lifecycle is at issue. However, a real management project of
lifecycle cannot be represented solely by the parametric time, for it consists of
series of engineering decisions for all sorts of this or that; this event or that
event, this sequence or that sequence. At every moment of decision, all the
engineering potentials or possibilities must be activated as alternative choices
and references [3], [4]. Two streams of activities in the form of potentials then
flow into the current activity at present, one from the past and the other from
the future. Two streams get convoluted and interacted at present, bounce
around their possible ways of engineering settlement, and then come to a
final settlement for their optimal or better mixture in the end. Thus the
whole of engineering potentials is transformed into a unique event at present
or decision now. Hiley (2000) [5], who provided an algebraic interpretation for
Bohms implicate order, gave a very suggestive and vivid metaphor for this
account;
The movement of a symphony involves a total ordering which involves the whole
movement, past and anticipated, at any movement. We hear new notes reverberating
94 J. Yagi, E. Arai
within the memory of the previous notes. This together with the anticipation of future
notes constitutes an unbroken movement. We comprehend movement in terms of a
series of inter-penetrating, intermingling elements of different degrees of enfoldment all
present together.
The last line in italic is particularly important to make proper engineering

decision, that is, to regress freely back to any point of time in the past,
anticipate the world paths to future, and return back to the current engineer-
ing work at present. It is the issue dealt in this manuscript how to implement
this mechanism in engineering decision making throughout the lifecycle of
engineered products.
3 4+1 Dynamic Mangement of Lifecycle
3.1 Introduction
An engineering decision making faces a series of opposite alternatives for

choice, seemingly equally valid, from which it is forced to choose one, often-
times under severe contention which way to go may lead to possibly far different
consequences or distinct patterns of consequences. It is the norm for decision
making for lifecycle management, rather than otherwise, to make a choice out
of multiple alternatives. We confront with burning potential of opposites
almost at every critical decision point to the left or to the right, up or down
metaphorically, A or A, speaking most generically, where both opposites
coexist in the acting potential (Whitehead, 1978 [6]), and both are rushing
toward realization but neither yet are chosen.
The primary requirement for lifecycle management is therefore to engineer
an adequate mechanism to make a proper choice among ordered information
represented by the spontaneous symmetry breaks of potentials in result of
interaction with the engineering environment or boundary conditions. It is no
less than to engineer the convolution of two streams of information flow, one
from the past and the other from the future, interpenetrating two temporal
phases, freely coming-and-going between them, and eventual settlement at
present with a focused image mirroring the past and future.
The etymology of the word, ubiquitous goes back to Latin, ubiqu- every-
where + itas ity, meaning Being or seeming to be everywhere at the same
time; omnipresent. The stem of the word, ubiqu refers to everywhere in space
at present or omnipresent in that regard. It is not omnitempo, lacking of the
meaning, every time, anytime. The National Institute of Advanced Indus-
trial Science and Technology (AIST), Japans largest public research organi-
zation began a project, called Virtual Time Machine [7] project, which tries
to extend the ubiquitous technology to the omnitempo technology that
covers real-time sensing, pattern recognition, data mining, behavior mining,
active tag, active database, the world simulation for possible scenarios, world
Virtual Time Machine
the PRESENT State Possible FUTURE States
Dynamic Simulation
the possible Doubles of

Pattern Recognition
Scenario Generators
Continual Dialog with

Real Time Sensing
b
Time-Space
World Storage
the Future
c
Environment d
Contexts
Data Compression
Pattern Recognition projection

projection
backward forward
Data Mining Behavior Mining
Continual Dialog with

the Double of the Past in PAST States in Piles
the World Storage
Fig. 1 Virtual time machine (Report on Virtual Time Machine by National Institute of
Advanced Industrial Science and Technology, 2004)
storage, and virtual temporal transportation (Fig. 1). It is an ambitious

engineering project indeed, but not so much ambitious as Hawikings real
time machine, whose probability of realization he figures 10 to the minus
power of 10 to the power of 60 [8].
The world storage may have a storage capacity as large as petabytes (10 to
the power of 15 bytes), which stores multimedia information captured from
active sensor networks (Google uses two to three petabytes for indexed infor-
mation worldwide in total). The network nodes are constituted of various
sensor devices and tags. The key technologies (Fig. 2) are (1) data and behavior
mining; monitoring and analysis of large amount data through sensing devices
and extraction of hidden behavioral intents or meaning from a bulk of data,
(2) temporal human interface; to contract or stretch time, realizing free walk-
through in time, moving forward and moving backward, (3) possible worlds
simulation; to simulate various world paths, future consequences or scenarios
enacted by current choices to be made with enabling technologies; genetic
algorithm, vision technology for flow line, sound spectrograph, robust sound
recognition, information storing in 4D virtual space, volume graphic cluster,
ultra-sonic sensing for human movement, Bayesian network, grid technology,
multi-agent simulation, ubiquitous agents, automatic generation of tags, mag-
neto resistive random access memory.
96 J. Yagi, E. Arai
Behavior Mining Temporal Human Interface Possible World Simulation

monitoring and analysis of large temporal compression, extraction, simulation of possible
amount data of people and society move forward, move backward consequences or scenarios enacted
by social or personal choices
extraction of hidden behavioral creation via continual dialogs with
intents from a bulk of data the doubles in the past and in future
[Sensing Archive + Data Mining] [Hyper Interaction with Time] [Simulation + Data Mining]
Ubiquitous network
World Model
Data Compression
Active sensor network
World VTM
Real Time Sensing
Super speed Grid Technology Real World
Fig. 2 Virtual time machine structure (Report on Virtual Time Machine by National Institute
of Advanced Industrial Science and Technology, 2004)
3.2 Requirements
One of the basic requirements for management technology for living stage of
inhabitation is to retrieve freely the needed information of architectural com-
ponents from the past archive which contains history of change in attributes of
the components and structures. Then automatically or semi-automatically
reconstituted is the whole 3D design, which faithfully represents the current
architecture of complex structures.
The causal chain among the functional requirements, physical system beha-
viors, and human behaviors is shown at Fig. 3. The horizontal box represents
the living stage, where the physical system environment interacts with inhabi-
tants continually throughout lifecycle. The vertical box represents the construc-
tion stage, where the inhabitants requirements determine the housing
functionality. It must be noted that the construction stage includes continual
retrofit, re-make, and repairing long after newly built. Maintenance and
upgrading of the physical system must be considered as critical activities for
the lifecycle management. The horizontal and vertical boxes are the view of the
same lifecycle process from different perspectives; the former represents the
challenges imposed by change in the relation between human and physical
Fig. 3 The causal chain among the functional requirements, physical system behaviors and
human behaviors
environment, and the latter represents the counter-acts against the engineering
challenges encountered during the lifecycle.
3.3 4+1 Adaptable System of Lifecycle Management
The 4+1 adaptable system of lifecycle management is a housing version of

AISTs virtual time machine or omnitempo engineering. It is an engineering
response to science, Eddintons algebraic interpretation of physics, and Hileys
ingenious view of physical orders, and temporal order in particular, mentioned
at Section 2. In order to complete our response, our list of research tasks is
presented below;
1. Parts oriented construction with hierarchical system of components covered
by distributed network of multi sensors, radio frequency identification tags,
RT middleware and super micro central processing unit, micro-actuators,
and long-lasting battery in addition.
2. Updated reformulation of architectural design via data acquainted by
sensors.
3. Design technology for easy-to-recyclable parts and structure.
4. Open supply chain of replacing parts.
5. Behavior mining, data mining, potential simulation, temporal human inter-
face of transportation.
6. Evaluation of maintenance and upgrading.
The 4+1 adaptable system is structured by four functional spaces and the
axis of temporal sequences as depicted at Fig. 4. The former constitutes four
functional spaces categorized by their characteristic functions: (1) self-descrip-
tive functional space, (2) self-sensing functional space, (3) self-searching
98 J. Yagi, E. Arai
Fig. 4 The 4+1 adaptable system of lifecycle management
functional space, and (4) self-imaging over history functional space. The latter
of +1 part is temporal integration of four functional spaces by circulating
four functional spaces in order of self-descriptive, self-sensing, self-searching,
and self-imaging over history.
Each categorical component of the 4+1 system will be described below in
order.
1. Self-descriptive
When a new component is introduced into the existing complex hierarch-
ical structure, the newly built-in component describes its own spatial and
temporal position within the hierarchy on its own, such as its subservient
spatial order for assembly and temporal order of durable period. The self-
filing mechanism can be realized by plug and play mechanism of radio
frequency identification tags attached to components over the tag network
which covers everywhere in the architectural system. Tags may contain
physical properties such as material property, stress characteristics and
distribution, archival manufacturing record, quality information, product
specification in addition to spatial and temporal characteristics.
2. Self-sensing
The tag network is superposed by the sensor network, most of whose
nodes are embedded on tags with various sensing devices for temperature,
moisture, ultra red, acceleration, wind loading, strain measuring, sound
loading, light intensity, smoke, olfactronics, electro-magnetic and other
sensors currently available or available in the future. Their position is either

static or movable within the network distributed over the whole hierarchies.
The sensor nodes function in either static or reactive mode. They function
either in individual mode or in collective mode, where the latter may make a
target object having hybrid sensing characteristics detectable with better
precision. A house thus becomes a sensing house device with sensory nerves.
It is not just sensing, but activating a series of proper events in reactive mode.
For an example the skeleton is exposed by various loadings such as seismic,
moisture, wind, light and other relevant loads constantly over lifecycle,
whose unavoidable consequence is its aged deterioration. When the sensor
network detects the accumulated loads beyond its threshold, extracting
needed information from the central database, it activates an event of warn-
ing locally. Some of sensor nodes may not be activated always, they may be
in sleeping mode and activated only when some events detected by other
nodes. Or some sub-network of sensor nodes may even be detached from the
whole network temporally, and works locally as an independent network.
3. Self-searching
The tag nodes are either passive or active tags. When a relevant tag is
positioned deep in the hierarchies, it becomes difficult to retrieve necessary
information from the tag. The electro-magnetic wave detection device on
tags is installed and then the hiding tags can be accessed as star dusts in the
sky can be detected in the same manner. Micro-robots play a role of supple-
mental detection for the shielded tags, which loiter around the network now
and then. Or let hiding tags establish peer to peer communication with
neighboring tags, and relay the contained information.
4. Self-imaging over history (memory)
Terabytes hard disks are already commercially available, and petabyte (10
to the power of 15), exabyte (10 to 18) will soon be available. AISTs
conception of world storage is no longer imaginary. Sensed data via the
sensor network will be stored piles by piles over time in even a single hard
disk. Scenario-based potential simulation, temporal transportation from the
past to future, and the other way around, behavior mining, data mining, and
pattern recognition are all made possible by the world storage. Memories are
prerequisite and basis for consciousness, which is produced by pattern
matching current sensory input with memories as in our brain [9, 10]. The
world storage is not just reservoir for electric contents like texts, images,
sounds, movies, patterns, but also computation mechanism is attached to
each data point whose value change activates events over some thresholds
and propagates signals through the whole network.
5. Self-awareness (+1: self-dialog over time)
A proper engineering decision can be made when one can freely access to the
entire history, basing on which one can simulate the future consequences. The
pattern matching to the past data produces engineering awareness. It is not once
100 J. Yagi, E. Arai
for all activity, but recursive dialog with the story of the past and future [11]. It is
something like continual communication with its own double in the past and
future, which immediately recalls Bohms self-recursive mirroring loops of the
spontaneous and unrestricted act of lifting into attention [12]. The technology
needed to realize the free self-dialog over time is a control mechanism of
compression, expansion, backward-flow, forward-flow of time, and creation
and sharing of time axis at will in a virtual world of space-time.
4 Experiments
4.1 Introduction
The lifecycle of a construction project consists of three phases, design phase,
construction phase, and maintenance phase as depicted at Fig. 5. The stake-
holders change their roles as the phase proceeds; the main players at design
stage are owners and designers, at construction stage, they are general contrac-
tors and traders, and at maintenance phase they are users and facility managers.
The holders of architectural documentations also change, as the stakeholders
change. The architectural models also change as they do. It is imperative to
develop an engineering mechanism to share the disintegrated information for
the lifecycle management.
Planning Stage
Conceptual Model
DESIGN PHASE
Detail Design Stage OwnerDesigner
Product Model
Design Document
CONSTRUCTION PHASE
To-be-Built General ContractorsSub contractors
Construction Stage
Process Model
Construction Plan
To-be-Built Document
Progress Management
MAINTENANCE PHASE
Model UserMaintenance Manager
Model
As-Built Construction As-Built
Experience Doc. Document
Document
As-Built Operation Stage
Usage Model Actual Usage
Document
Maintenance Stage
Maintenance Model Actual Maintenance
Document
Fig. 5 The lifecycle of an architectural product

Some in-situ experiments are conducted, one at construction phase (Sec-

tion 4.2), and another at maintenance or operation stage (Section 4.3) along
the line of the afore-mentioned views developed at the previous sections.
Various types of disturbance factor exist for the architectural lifecycle man-
agement. It is difficult to manage related information by using conventional
model. A new concept of Parts and Packets unified architecture [13] is
proposed. Data as information packets related to parts are carried by parts
themselves and can be handled to manage whole system of the hierarchies of
parts. It is integration of parts and packets with a control mechanism of
their dynamic flow throughout a perplexedly complex network of product
lifecycle in construction. It is a unified controller system which operates
parts and packets together.
Both experiments focus on the issue how to obtain the necessary information
from the architectural components and parts, and how to use this information for
construction and maintenance respectively. The same mechanism is employed for
both cases, using IC tags, sensor network, active database, 3D CAD representa-
tion and work-through, RT middleware for the network operation.
4.2 Construction Phase
Construction is surely production activity, but is better characterized as project-

type production, defined as a course of concerted actions intended to achieve
some chosen purpose for production, within a given period of time, under some
given constraints on available resource (man-power, material, and money), and
with responsibility for sustainment of environment. It is therefore of critical
importance to allocate production activities properly according to the well-
planned process model and scheduling. It is, however, more indispensable to
adjust the pre-assigned model and scheduling dynamically to the continually
changing production conditions and environment to concert the needed
actions. It is of primary concern for a project type production how to achieve
the dynamic equilibrium at every stage from the commencement of the project
to its completion which makes possible concerting the needed actions. The
combination of radio frequency identification and active database (or glue
logic) may open up promising that they might provide the mechanism which
achieves the required dynamic equilibrium for construction activity without
hindrance or halt of production at worst.
The dynamic nature of construction activities is exemplified at Fig. 6.
The in-situ experiment of construction control is conducted at some con-
struction site of a residential building of nine floors located in Tokyo. For the
construction site is located in a dense area of a town, the control room is located
at a room in a nearby hotel, 75 m away from the construction site. The
construction process is thus remotely monitored and controlled. The oriented
antennas set at both site and control room establish wireless connection
between them. The experimental settings are depicted at Fig. 7. Two types of
scheduling
mng. for design design
change
contact design for
design plan completion
execution
budget mng. new approval
for design by clients documents
drawing management
price
sales &
marketing negotiation
Modeling Engine
invoice &
callback Scheduling Engine
Glue Logic / Process Design Simulation Engine
Glue Logic / XML Server
Agents for
integration of ERP PDM Construction Site
existing
applications Glue Logic / Process Design
CAD Word/Excel PDF
Fig. 6 Dynamic process of construction
oriented antenna for

servers at control room wireless LAN
Remote control room
Construction Site
75m, elevation 35
oriented antenna for equipments at

wireless LAN construction floor
Fig. 7 On-site experiment setting

room tag
parts tag
Fig. 8 Distribution of tags on construction floor
IC tags are distributed on each construction floor as depicted at Fig. 8; one type
is fixed at each room for identification of location, and the other is tag attached
to each component (window sash) as control target.
Once parts are manufactured, chip-implanted parts are shipped to a con-
struction site, assembled, and installed on site. Despite that the parts are
dislocated spatially as well as temporally, or are reconfigured in succession as
they become more integral part of a building under construction, simple act of
reading the product URL on tag triggers to change the attributes of these parts
in the data management system of active database. As the parts attributes are
changed, they autonomously trigger to send messages to the pre-assigned
addresses with a simple logic attached to each data point. Every data point
therefore contains attribute, simple logic, and address. Millions of data points
are passing information each other. This very bulk of acts of passing and
receiving change the state of the whole data base, dynamically. Hence the
data base collectively behaves like an autonomous giant controller as well as
self-renewing data repository. Figure 9 shows the real time control for progress
of construction by this mechanism from manufacturing thru shipping to instal-
lation on site.
Fig. 9 Progress control of components installation
4.3 Maintenance Phase
The experiment for the operation stage is conducted at a warehouse of wines of

repute, located at Tokyo bay water-front. The temperature and moisture con-
trol is critical for the warehouse to keep the quality of wine intact. A sensor for
temperature and moisture is installed on IC tags of active type (Fig. 10) devel-
oped by AIST [14]. Some special IC tag is developed, which uses very weak
radio wave for communication to save battery. The I/O port is secured by 4 bit
on the tag, to which a sensor module is connected. The used sensor is shown at
Fig. 11 with specification.
Fig. 10 Active IC tag module

Fig. 11 Sensor module for temperature and moisture
The configuration of the installed network is shown at Fig. 12. Several

wireless access points are set up, one wireless LAN server station with three
wireless relay stations. The sensor installed tags distributed in a part of the
warehouse communicate with the relay stations via faint radio wave.
The monitoring of temperature and moisture is conducted for two weeks at
the warehouse, whose typical results are shown at Fig.13.
Fig. 12 Network configuration

(a) Temperature distribution (b) Moisture distribution
Fig. 13 Network configuration
5 Conclusion
The 4+1 adaptable system of lifecycle management is an engineering response

to scientific knowledge obtained by the recent development of physics and brain
research which suggest that a proper engineering decision can be made by
recursive dialog with the story of the past, something like continual commu-
nication with its own double in the past, or like Bohms self-recursive mirroring
loops of the spontaneous and unrestricted act of lifting into attention. The
free self-dialog mechanism over time must be engineered. A control mechanism
of compression, expansion, backward-flow, and forward-flow of time, and
creation and sharing of time axis at will in a virtual world is the basis for the
necessary engineering dialog with the past experience as well as its mirroring
image in the future, potential world simulation.
Some in-situ experiments are conducted from the views above mentioned.
Various types of disturbance factors and difficulty of decision makings exist for
the lifecycle management. Parts and Packets unified architecture is proposed
to open a way toward the solution. Data as information packets related to parts
are carried by parts themselves and can be handled to manage whole system of
the hierarchies of parts. It is integration of parts and packets with a control
mechanism of their dynamic flow throughout a perplexedly complex network of
product lifecycle. It is a unified controller system which operates parts and
packets together.
References
[1] A. Eddington: The Philosophy of Physical Science, University of Michigan, Ann Arbor,
1958.
[2] B.J. Hiley: Vacuum or Holomovement in the Philosophy of Vacuum, Clarendon Press
Oxford, 217-249, 2002.
[3] J. Yagi, et al: Action-based Union of the Temporal Opposites in Scheduling: Non-
deterministic Approach, J. Automation in Construction, Elsevier, 12, 321329, 2003.
[4] J. Yagi, et al.: Logics of Becoming in Scheduling: Logical Movement behind Tempor-
ality, Knowledge and Skill Chains in Engineering and Manufacturing; information
infrastructure in the era of Global Communications, ISBN: 0-387-23851-4, Springer,
111118, 2005.
[5] B.J. Hiley: Non-commutative Geometry, the Bohm Interpretation and the Mind-
Matter Relationship Proc. CASYS2000, Liege, Belgium, Aug. 712, 2000.
[6] A.N. Whitehead: Process and Reality, the Free Press, 1978.
[7] M. Hirose and K. Sakagami et al., Virtual Time MachineVision for Transcendence of
Temporal Limits via Information Technology, Report of National Institute of
Advanced Industrial Science and Technology (AIST), Japan, 2004.
[8] S. Hawikings: The Universe in a Nutshell, Bantam Press. 2001.
[9] M. Jibu and K. Yasue: Quantum brain dynamics and consciousness, John Benjamins
Publishing Company, 1995.
[10] G. Vitiello: My Double Unveiled, John Benjamins Publishing Company, 2001.
[11] D. Bohm: Wholeness and Implicate Order, London: Routledge and Kegan Paul, 1980.
[12] G. Vitiello: The dissipative brain, arXiv:q-bio.OT/0409037, v1, 30 Sep. 2004.
[13] J. Yagi, et al.: Parts and packets unification: Radio frequency identification application
for construction, J. Automation in Construction, 14 (2005) 477490, 2005.
[14] K. Ohara, K. Ohba, B.K. Kim, T. Tanikawa, S. Hirai, and K. Tanie, Ubiquitous
Robotics with Ubiquitous Functions Activate Module, Proceedings of Second Interna-
tional Workshop on Networked Sensing Systems, 2005.
A PLM Tools Taxonomy to Support
Product Realization Process: A Solar Racing
Car Case Study
D.A. Guerra-Zubiaga, E.D. Ramon-Raygoza, E.F. Rios-Soltero, M. Tomovic,

and A. Molina
Abstract Product Life-cycle Management (PLM) is a wide research issue

exploring the information and knowledge in the entire product lifecycle to
improve and increase innovation for a given product. The management of the
information and knowledge related to the whole life cycle is supported by
different Knowledge Management tools such as Product Data Management
(PDM) and Expert Systems. PLM receives support from PDM systems in the
many activities of the lifecycle from product design and creation, through
dissemination and after sales services, up to product dismissal and recycling.
Besides PDM systems, there are other knowledge management tools which
havent been integrated into the PLM. A particular shortcoming to integrate
tools and techniques for the design process within a PLM environment is the
lack of a PLM tools taxonomy. This paper presents a new PLM tools taxonomy
to support product realisation process
Keywords PLM tools PLM taxonomy PDM systems integration

Engineering Design Expert systems
1 Introduction
Product Life-cycle Management (PLM) is a research area adopted by compa-
nies, which can drive the life of a product by integrating their information,
knowledge, individuals, activities and software. The stages that a product
undergoes through its life are basically development, production, market activ-
ities, use, maintenance and disuse. Many authors and companies have defined
their own phases, but there is a general agreement that products are designed,
manufactured, distributed, sold and used; most of them are maintained, and
they are disposed or recycled when they need to be retired.
D.A. Guerra-Zubiaga (*)

Prolec GE S. de R.L. de C.V., Research and Development, Blvd. Carlos Salinas de
Gortari Km 9.25 C.P. 66600 Apodaca, N.L, Mexico
e-mail: david.guerra@ge.com

110 D.A. Guerra-Zubiaga et al.
The whole life cycle of a product is usually very long. This causes
a disagreement regarding the PLM definition, its stages and the tools it
comprises [1]. It also makes it too broad for a single tool to include all the
capabilities it requires which in turn causes a lack of integration among
several tools used throughout the PLM stages. Therefore, the products
knowledge, which most of the time comes in disparate formats, is not
always shared among its life-cycle [2]. For instance, there is no standard
regarding Computer Aided Design (CAD) formats, and the technology to
migrate from one format to another only handles geometry representations
of the product (Bill Of Materials are being included already but there is no
standard either).
The aim of this paper is to present some types of PLM tools that support
the engineering design stage in a PLM environment using a PLM tool
taxonomy.
2 PLM Definition, and Its Stages
2.1 Literature Survey of the PLM Concept
The concept of Product Life Cycle (PLC) has been addressed since the 1950s [3]
or 1960s [4]. However, it must be stated that there is a difference between PLC
and PLM, and that there are two points of view for the stages comprised in the
life cycle of a product. Some authors consider it as the different profit and sale
levels that a product has through its life. In this case, a products time in the
market is divided into introduction, growth, maturity and decline [5], [6]. This
concept is much older than the actual PLM, but is still used in the production,
sales and marketing niches. Researchers in this area use it to increase the
revenues by predicting the market behaviour, and taking actions towards
these predictions.
On the other hand, PLM intends to increase innovation [1], reduce errors [7]
and be more competitive [8]. One of the first public attempts to define PLM was
in the late 80s by Konstantinov [9]. His work was situated in the design manage-
ment area, and as it would be expected, the conception of PLM has changed
since then.
There are many researchers working in the PLM area. This situation has lead
to the proliferation of many representations and definitions of PLM stages.
Besides, they increase due to the fact that each company will have to develop a
life-cycle concept for its products[10]. However, an agreement can be found
about the underlying concepts in most cases.
Table 1 presents a summary of the different stages conceptualized by several
representative authors. Altings work [11] was selected for being a seminal
source. The definitions provided by Aca [12] and Guerra et al. [13] are an
A PLM Tools Taxonomy to Support Product Realization Process 111
Table 1 Representative PLC representations (Adapted from [8], [11], [12], [13], [14] and [15])
[11]
Need recognition Production Distribution Usage Disposal
Design development
[12]
Product Process Production Usage and
Facility Disposal
development development development maintenance
Sales
[13]
Ideation Product Process Facility Production Usage and Disposal
development development development maintenance
Sales
[8] After sales Production

Product Process Factory -Marketing -Sales Quality Production scheduling
design planning planning -Procurement -Distribution Maintenance planning & control
[14]
Requirement Concept Product Manufacturing Manufacturing Sales &
analysis and engineering & engineering Disposal
engineering and production Distribution
planning prototyping Recycle
[15]
Imagine Define Realize Use/support/
End of life
service
example of the evolution of the PLM concepts. Kovacs et al. [8] dissect their
representation to facilitate the identification of most subprocesses included in
the life cycle. Grieves [14] is a representative author in the industrial sector for
the PLM area. Stark [15] provides a condensed representation, comprehensive
definitions and several points of view of the PLC concept. A list of PLM
definitions is provided in the appendix.
Table 1 presents the PLC definitions by many authors. Acas work [12] is
partly based in Altings [11], and provides another cycle which is divided
into two major classifications: engineering and supply. These have three
stages respectively. Based on his work, Guerra et al. [13] add an ideation
stage at the beginning of the cycle. Similarly, Kovacs et al. [8] propose an
engineering stage subclassified into product design, process planning and
factory planning. They also present two other stages with subclassifica-
tions: support activities chain and operations chain. These three stages are
parallel. Grieves [14] proposes seven stages which are shown. In turn, Stark
presents different points of view about the life cycle, including the user and
manufacturer perspectives. The table shows a condensed version of these
two.
In the following sections, these concepts will be used in an attempt to provide
an integrative and more general definition of the PLC, which is depicted in
Fig. 1
Fig. 1 Stages of PLM
2.2 Relationship Between Concurrent Engineering and PLM:

The Development Stage
A look into the early stages of the PLC shows that product design and product
development are interpreted differently by many authors. According to Suh [16] ,
design is an interplay between what we want to achieve and how we want to
achieve it. He states that every designer follows these tasks: (1) know or
understand their customers needs, (2) define the problem they must solve to
satisfy the needs, (3) conceptualize the solution through synthesis, (4) perform
analysis to optimize the proposed solution, and (5) check the resulting design
solution to see if it meets the original customer needs. Likewise, Aspelund [17]
says that design is a plan of action, created in response to a situation or
problem that needs solving. In his opinion, there are seven stages in the design
process: conceptualization, exploration/refinement, definition/modeling, com-
munication and production. In this case, production is referred not as mass-
production, but to prototypes or single pieces. Conversely, development, as it is
defined by Otto and Wood [18] and Aca [12], include both design and manu-
facturing plans.
As it is depicted in Table 1, Alting [11] specifies that the development work
normally begins based on an assessment of a need recognized in the market [10].
In his definition of Product Development, Aca describes four subactivities
called Conceptualization, Basic Development, Advanced Development and
Launching, where Basic Development includes the gathering of customer
requirements information [12]. Guerra et al. [13] mention the ideation stage
and keep the market requirements given by Aca within Product Development
[19]. Grieves also takes into consideration that there is a need prior to the formal
definition of the product [14]. Finally, Stark mentions the previous existence of
a dream in someones head [15] which is later concretized.
In a PLM environment, the development stage should comprise a holistic
view of the whole PLC [1]. Concurrent Engineering (CE) fits and is used in this
phase, because its focus is on the integrated design of products, and its goal is
that outset developers consider all elements of the product life-cycle [20]. CE
intends to prevent adverse choices regarding a product by bringing together
different disciplines such as product design, manufacturing, procurement,
maintenance and marketing [21, 22]. The collaborative design of a product
and its manufacturing processes needs information on product features, man-
ufacture, services and costumer requirements while the design is simultaneously
going on [23]. This information includes part and plant layouts, designs and
manufacturing rules, as well as any other useful information for the product
life-cycle [23].
Note, however, that CE only happens at the development stage of the PLC.
For example, it doesnt comprise the actual production, but its planning and
design. In other words, the only physical results from CE could be partial or
complete prototypes.
2.3 Production and Market Activities
Production and market activities are related. For example, it is difficult to

distinguish the boundaries between materials requirements and procure-
ment. On the one hand, a brief examination of Table 1 shows that every
author coincides in the existence of a production or realization stage. It
should be noted that Stark refers to realize as the production phase [15].
Production basically refers to the manufacturing and quality assurance
tasks, i.e. everything that happens in the shop floor. Materials acquisition
is also considered as part of this stage. The squared boxes of Table 1 are
included in this phase
On the other hand, it can be seen by the arrow boxes in Table 1 that there is
an overlapping between production and what will be referred as market activ-
ities. According to Juttner et al. [24], the product life-cycle management needs
to balance fast response to changing consumer demands with competitive
pressure to seek cost reductions in sourcing, manufacturing and distribution.
In Acas work [12], production and sales are mixed into a single phase. This
representation is followed by Guerra et al. [13]. However, it can be seen that the
rest of the authors keep market, sales, or distribution separate from the man-
ufacturing or production stage.
Market activities include marketing and trade of goods. Therefore, supply
(or distribution), sourcing (or procurement) and sales are also part of this stage.
The market requirements are communicated to developers to include clients
needs into product specifications. Note that these are restricted by production
capabilities. The market behaviour also determines the strategy of the company
towards New Product Development (NPD) and investments. The supply chain
management requires both the business administration and engineering plan-
ning, and is determined by both the production volume and distribution
channels
2.4 Use and Maintenance
As it happens with production and market activities, use and maintenance have
a direct relationship. The use stage is customarily related to the customer or
operator [15]. As the product is used, it will eventually need support, repair,
service or upgrade to keep it in good conditions, change its application or
improve the way it works. From a life-cycle perspective, the manufacturer
plans this in advance. Furthermore, product-service supply is increasingly
becoming more relevant to the point that companies could solely provide
accompanying services for products manufactured by themselves or another
company [8].
Even when it is not shown in table 1, Alting specifies that there are companies
and user costs due to warranty service and maintenance during what he calls the
usage stage [10, 11]. In his graphic representation of PLM Grieves does not
include use or maintenance. However, in his definition of PLM, he states that
the product life goes from its design through manufacture, deployment and
maintenance-culminating in the products removal from service and final dis-
posal [14]. This clearly reflects that the product is obviously used at some point
of its life, and that it is prone to maintenance. In the case of Stark, he differ-
entiates use from support and service as it is perceived by the client and
manufacturer. In his words, when the user is using the product, the manufac-
turer will probably need to provide some kind of support [15]. All of the above
denotes that there is a concern about maintenance, and reinforces the argument
that it is strongly related to the use phase.
From a corporative perspective, there are basically two options for main-
tenance: outsourcing and in sourcing. In the first case, the buyer may come to an
agreement with the manufacturer as to receive maintenance from them, or may
sublease another company which is specialized on that service. In the in sour-
cing case, the company can direct its own resources to give support to the
product. The advantage of outsourcing is that it enables companies to concen-
trate on their core competences. Therefore, they dont have the need to train
and maintain a department or specialist for this task. Most of the time, this
requires that external entities involve in the daily activities of the assisted
company, which is not always beneficial. This is rather true for high-tech
enterprises that need to keep a low profile about their processes.
In Fig. 1, the use and maintenance stages are separated due to the fact that
the user and service provider most often deal with the product at different times
and locations, and they have different goals towards the product.
2.5 Disuse
The disuse of the product comes when it doesnt provide enough benefit to the
company or user, and needs to be recycled, refurbished for sale or disposed.
Stark [15] calls it end of life, Grieves [14] addresses it as disposal and recycling,
both Alting [11] and Aca [12] call it disposal, so do Guerra et al. [13]. Kovacs
et al. [8] include recycling as part of their PLM definition as well. The intention
behind the term disuse is to state that the product may or may not continue
existing after it is not used anymore, that there are many possibilities for what
happens afterwards, and that it is no longer used for the purpose for which it
was designed. It also reflects that its owner may change, but its lack of useful-
ness determines its retirement.
3 PLM tools taxonomy and techniques
Figure 2 presents a taxonomy of digital tools and techniques used in PLM. The
proposed classification takes into account that people within a company have
different profiles regarding their activities and the tools they use. The PLM
tools are divided into three major categories: tools for engineering, for knowl-
edge management and for business activities. Except for ambient intelligence
(see Fig. 2), the engineering digital tools are implemented for concurrent
engineering within the development stage (see Fig. 1) presented in Section 2.2.
Business tools refer to the software used in the market activities (see Fig. 1). In a
simplified way, Knowledge Management (KM) tools are used to transfer and
manage data throughout the stages of PLM (see Fig. 1) and their respective
tools (see Fig. 2), so they are used by both business and engineering tools.
Fig. 2 Taxonomy of the digital PLM tools and techniques

Representative examples are presented for most of those subclasses. Going

from specific to general, these tools are implemented in their respective stage
and are linked to the rest of the tools (view Fig. 1). For example, CAD software
is used for engineering design in the development phase.
3.1 PLMKM Digital Tools
Knowledge Management Systems (KMS) are represented in the middle of Fig. 2.

These are information systems which make knowledge available to people as
required [14]. They are a mean to create, share and use knowledge [15]. These
tools are divided into Groupware and Decision Support Systems (DSS). Gallupe
defines Groupware as Software and hardware that enables workgroups to
communicate and collaborate [25]. It is a new type of computer software and
work processes to collect, document, classify and organize knowledge [26]. The
definition provided by Gunnlaugsdottir is collaborative technology which
allows people to communicate with each other, co-operate on projects and
share information and knowledge [26]. Beyond this definition, she states that:
The groupware links employees together and connects them with the information and
knowledge base of the organisation, offering them the opportunity to use it and expand
it. Records are safely stored in an organised, central database where all authorised
employees have access to the latest versions of manuals and other documents and
records. The system offers also version control, that is how many versions were made
and who wrote each version. [26]
The most common Groupware used for PLM is the Product Data Manage-
ment (PDM) system which aids companies in managing all product-related data
throughout the product life cycle (view Figs. 2 and 3). The main role of PDM
systems, within a PLM environment, is to provide support to the many activ-
ities of the lifecycle, such as design, the tracking of information in changing
orders, the alternative designs management, and the product configuration
control. PDM systems keep data shareable, transportable, secure, accurate,
timely and relevant [15], The fist term, shareable, refers to data that can be easily
and instantly viewed or used by more than one person at any instant. Trans-
portable regards the capacity of data movement in an easy fashion. When data
is secure, it means that it can be protected from unauthorized destruction,
modification or use. Accurate data is reliable and precise. It can also be said
that timely data is current and up-to-date. Finally, it can be called relevant if it is
really useful for the decision in turn. Likewise, PDM systems provide the plat-
form in which every kind of user can access the information to share or modify
it. However, software such as CAE or CAM tools (see Fig. 3) should be used to
modify the integrity of the given data. In Fig. 3, the data is represented by NC
codes, marketing statistics, user manuals, repair reports, profit of recycling and
CFD data. This resembles the idea that PDM systems connect data throughout
many other applications, but doesnt modify it.
Fig. 3 PDM as the data bus

(left) and elements of a PDM
system (right)
An expert system is a Knowledge Based System (KBS) that emulates the

decision-making ability of a human expert. Remark that an expert can solve
problems that most people cannot or can solve them much more efficiently [27].
In the commercial world, however, there are systems that can effectively and
efficiently perform tasks that do not need an expert [7], for instance, Rule-
based systems [28].
Note that an expert system is limited to a determined activity. For instance,
an expert system for maintenance cannot be used for engineering analysis.
Some of the expert systems developed for product design are: ProPlanner1
and Expert system for product manufacturability and cost evaluation. Some
Rule-based systems are: Product audit tool1, and Boothroyd & Dewhursts
DFA and DFM Concurrent Costing1.
It is important to mention that the digital tools for business and the digital
tools for engineering are integrated by the knowledge management tools.
Furthermore, some of the business or engineering tools already have knowledge
management tools embedded. For instance, there is software for design and
manufacture that shares files indistinctly (e.g. CATIA1 and DELMIA1 from
Dassault Systemes1) and can create design rules for product families (e.g. the
Knowledgeware module from CATIA1).
3.2 PLM Business Digital Tools
The business administration, sales and marketing department are essential for
an extended enterprise. The market activities have not been fully integrated to
the PLM maybe because the PLM concept was coined within the engineering
niche. However, the people involved to what has been referred here as market
activities also deal with valuable information regarding the product life-cycle.
Therefore, the information they provide also drives the life of the product.
Some companies use different strategies to be more competitive. One of

the most comprehensive and detailed strategy tools is the Collaborative
Planning, Forecasting and Replenishment (CPFR) [29] (view Fig. 3). The
Voluntary Interindustry Commerce Standards (VCIS), which is responsible
for the adoption of the bar-coding, and the Efficient Consumer Response
(ECR) Europe claim that it helps retailers and manufacturers to get more
accurate forecasts, reduce stock-outs, increase sales, reduce inventories [29],
higher order fill rates, and faster cycle time [30]. The CPFR is Web-based,
and its goal is to coordinate trading partners on various activities such as
production and purchase planning, demand forecasting, and inventory
replenishment [31].
Figure 2 shows artificial agents as part of the sales tools. Firstly, it must be
understood that the Agency Theory studies the contractual relationship
between two parties [32]. An artificial agent is a Knowledge Based System
able to communicate, interact and negotiate with other agents, which can be
intelligent, human or artificial [33, 34].
Electronic Data Interchange (EDI) (see Fig. 2) allows a computer-to-com-
puter exchange of business data for a transaction without human intervention
[30]. EDI is used to reduce cycle time, improve inventory management, increase
productivity, reduce costs, improve accuracy, improve business relationships,
enhance customer service, increase sales, minimize paper use, and increase cash
flow [30].
Procurement and Strategic Sourcing tools are very common for business
practices such as Customer Relationship Management (CRM) or chain supply
design tools. This type of tools is used for the distribution management of the
product (view Fig. 2). An e-procurement B2B system is an open system that
enables the organisation to reach and transact with suppliers and customers in
virtual markets [35].
3.3 PLM Engineering Digital Tools
Four basic categories are presented for the engineering classification of the
PLM digital tools taxonomy. It can be seen from Figs. 1 and 2 that these
correspond to the development, production, maintenance and disuse stages
presented in Section 2.
From left to right, the first engineering tools classification is maintenance. In
most cases, the digital tools used for this activity are limited to the development
stage, where the disassembly process is modelled, for example. Even when this
approach is useful for servicing products, there is much more information about
what is actually done in the repair shop during the maintenance stage which
needs to be integrated into the PLM.
The term ambient intelligence was coined by Emile Aarts of Philips [36].
It is defined as the convergence of ubiquitous computing, ubiquitous
communication, and interfaces adapting to the user [8]. It is a field of informa-

tion systems that is rapidly increasing, and could be used to provide valuable
information about the current status of the product [8]. This information could
be sent and received from the shop floor to the KMS of the company to
integrate the support department with the other departments of the company
such as procurement (to get spare parts) or development (to improve designs
through Design for Maintenance). Emilini points out that the Maintenance of
ambient intelligence environments and of their components is also expected to
play a significant role with respect to health and security issues [37].
One of the most promising tools to incorporate the disuse (see Fig. 1 and to
the right of Fig. 2) stage within the PLM is the Life-Cycle Assessment (LCA)
[38]. This kind of software can evaluate the environmental impact of a product
(as well as processes or activities) throughout its life [39] by compiling an
inventory of relevant inputs and outputs [40]. Some of the most popular tools
for this assessment are GaBi4 and SimaPro 6 [39].
The most thoroughly digital tools that are used within PLM are the Virtual
Concept tools. These include the manufacturing and design software for the
development stage (see Section 2.2). The design tools are able to produce
virtual prototypes of products, reducing scrap and improving quality of
products [41]. These virtual prototypes are digitized versions of the real
products that can be tested and simulated in near-to-in-situ operation. The
Virtual Manufacturing tools support the design and simulation of process by
means of layout planning, factory flow simulation, ergonomics and DMU
kinematics [42].
3.4 Techniques and Methodologies Used for PLM Tools

The digital tools use several embedded methodologies or techniques to accom-
plish their goals. In the case of KM tools, two very common techniques are
Data Mining (DM) and Case Based Reasoning (CBR). DM is an advanced
technology that can process large amounts of information, discovering useful
information and knowledge to support decisions [43]. CBR is a technology that
provides KMS with some degree of intelligence by comparing the situation in
hand to past captured experiences to solve the present case [43].
Maintenance is a fertile area to develop for PLM. Reliability prediction is
one of the most difficult issues to address in this regard, because there are many
variables involved. However, failure can be prevented by techniques such as
Condition Based Maintenance (CBM; see the center of Fig. 2) which is used to
determine when preventive maintenance should be performed [44]. CBM means
maintenance tasks based on the current status obtained from in-situ, non-
invasive tests [45]. It requires data from products to determine when service is
to be done. Therefore, this technique is used by ambient intelligence tools to
infer whether maintenance is needed or not [8].
Another technique used for the maintenance of products is the Propor-

tional Hazards Modelling. It is a technique that can make use of oil analysis
results collectively to perform statistical control thresholds to asses the risk of
failures [44]. But again, it requires a knowledge source which could be popu-
lated by the ambient intelligence tools. This data would relate to (1) failure/
replacement data; (2) inspection data; (3) maintenance action data; and
(4) installation data [44].
There is a wide variety of techniques for design. Figure 2 presents four of
these. First, Quality Function Deployment (QFD) is a way to identify
customers requirements [1] to generate clear engineering specs thus, mini-
mizing the waste and changes in products [46]. TRIZ is a Russian methodol-
ogy that is primarily useful in identifying physical working principles to
solve technical problems by finding implicit contradictions in a problem.
Design for X (DFX) means to design a product towards a specific goal.
Design for X includes a set of initiatives to reduce products cost and
improve its quality in the early stages of design. DFA1, DFM2, DFMA,
DFE and DFR are just some of the most common types of methodologies
[47]. Every special case of DFX follows a set of rules, guidelines or proce-
dures to accomplish its ends All of the above (QFD, TRIZ and DFX) can be
integrated into the development tools to generate and specify designs in
some way (view dotted arrows in Fig. 2), for instance, by the integration of
expert systems and CAD software [48].
4 Case Study
The case study in this section shows how the tools presented in the PLM
taxonomy are applied on a redesign project. The case study was taken from
the PURDUE solar racing project. This project involves the construction of the
7th generation of a solar racing car, or in other words, it consists of improving
the current solar racing car. This project started on September 2006 and will
finish during spring 2009.
The scope of this experiment is the hub system redesign. DFX process
framework will be followed in order to achieve that, and will be supported by
the Methodology to support DFX process using a PLM framework [48],
which includes the next stages: (1) Defining product requirements, (2) Design
for Assembly application and (3) Creating new knowledge and expertise. Each
of those stages are matched with the property step from the Methodology to
support DFX process using a PLM framework.
1
Design for: manual, automatic and robotic assembly are special cases of DFA.
2
Design for: casting, injection molding, welding, machining, brazing, etc. will be considered
as special cases of DFM.
4.1 Defining Product Requirements
This is the first stage to be achieved in this case study. The product requirements
are provided mainly by the product customers. The costumer requirements are
then translated from common language into technical parameters to modify
using a QFD methodology. In this case study, the last solar racing car team was
considered as customers due that they have the use and building experience of
the last version.
After an interview with the customers they provide the next requirements
for the product:
Mechanical systems must be improved. Those systems include chassis & Roll
Cage, Uprings, Axles, Hubs, Brakes, Training Arm and Steering.
4.2 Design for Assembly Application
The main issue in this stage is the application of DFA tool designed by Boot-
hroyd in order to perform the DFA task. However, before that, is necessary to
collect the useful product information available. In addition, the PDM applica-
tion will be executed in order to support this stage. The methodology steps
developed in order to support the mentioned issues are the second and the third
ones, which are: Getting Design for Assembly (DFA) information and Product
data-information management through a PDM system.
4.2.1 Getting Product Information

The product information collected corresponds to the hub system only, due that
it was the section assigned to this case of the study (Fig. 4). The product
information collected consisted in CAD files, Drawing files, CAM application
files, report documents and pictures files. CAD files and drawings format was
pro-e; CAM application files include mainly finite element analysis. Report
documents include Vehicle impact analysis and mechanical systems analysis.
4.2.2 Product Data-information Management Through a PDM System

Once the product information is collected the next step is to put it into the PDM
system in order to manage it properly and provide it at the designers in the best
way. However, according with the methodology proposed, previously to PDM
application there are some tasks that need to be performed, which are: to define
the PURDUE solar lifecycle, to define the PLM actors and to organize the
product information collected.
Solar racing car has six stages into its lifecycle (view Fig. 5). First one is
project plan, which contain all the projections for the future, tasks definition,
resources definition, organizational plan, etc.; Second stage is product redesign,
Fig. 4 Hub Drawing (Top View)
which include last part analysis and parts improvement; Third stage refers to
manufacturing processes design or in some cases just to improve the past
processes; fourth stage correspond to the marketing issues, where publicity
and sponsors searching are performed; Next stage is maintenance, in this one,
solar racing car is put it in use by the customer and then it naturally will need
some kind of maintenance due failures or simple use; Finally, when solar racing
car wont be able to work it must be disposal and/or recycle.
4.2.3 Defining PLM Actors

Once solar racing car lifecycle has been defined, it is necessary to define who the
actors are at each stage in order to analyze what kind of information is
generates by each one as well as to decide which information must be shared
them (Table 2). For instance, product designers could need some material
properties information, while marketing is just interested in which will be the
final solar racing car shape in order to create some publicity about it.
Project Product Manufacturing Recycling

Plan Redesign Proceses Design Marketing Maintenance &
Disposal
Fig. 5 Solar racing car lifecycle

Table 2 Actors into solar racing car lifecycle
4.2.4 Organizing Information

The information collected was organized following the organizing product
information class diagram proposed in the methodology (Fig. 6).
4.2.5 PDM System Application

In this case study, Teamcenter1 Community is the commercial software hold-
ing the PDM system. The Teamcenter community is used for managing the
product information through the entire solar racing car lifecycle. Then, in order
to star using Teamcenter community, is necessary to create the Teamcenter
Community site for the users access. After to create the Teamcenter community
site is necessary to create and customize the users. Teamcenter community
offers four kinds of users, which are Reader, contributor, web designer, and
administrator. The privilege of each one are explained below.
Fig. 6 Organizing information through UML
A. Reader Has read-only access to the web site.

B. Contributor Can adds content to existing document libraries and lists.
C. Web Designer Can creates lists and document libraries and customizes
pages in the web site.
D. Administrator Has full control of the web site.
For this specific case study all the users have been defined as contributors
because all of them are supposed to create files that must be share with all the
other members. After to define the users privileges, is the time to define the files
privileges, which are the same that the users. Then, we are able to upload those
files with their own privileges also.
4.2.6 Boothroyd DFA Tool Application into PDM Environment

After applying Boothroyds methodology it is necessary to define the most
economic assembly method for the particular project. Boothroyd proposes
three categories of assembly: manual assembly, special purpose transfer
machine assembly and robot assembly.
The most economic assembly method can be determinated by three aspects,

the first one is the annual production volume which is the average number of
assemblies of all styles produced during the equipment payback period; second
issue is the number of parts in the assembly, which is the average number of
parts or subassemblies to be assembled on assembly system; the last one is the
total number of parts, which is the total number of parts or sub-assemblies from
which various product styles can be assembled. Considering all the aspects
mentioned above, the most economic assembly method for the hub is manual
assembly method.
The methodology to manual assembly method involves the next steps:
1. To get the information about the product or assembly.
2. Take the assembly apart (or image how this might be done).
3. Fulfill the design for assembly worksheet.
4. Begin re-assembling the product.
5. To estimate manual assembly time and manual assembly cost.
6. Finally, the manual assembly design efficiency is obtained by entering the
figures generated from the worksheet into the equation: EM 3XNM/TM.
The sub-assembly system consists of six parts, which are rim, hub, washer,
pipe, bearing, break disc and screws. The hub system assembly exploded is
shown in Fig. 7.
The worksheet filled with the information of the assembly is shown below as
Fig. 8.
After analyzing the assembly with the support of the worksheet above, it was
decided not to reduce the number of parts in the assembly because of constrains
in manufacturing. However, the hub was redesigned in order to make it lighter,
the new hub design is shown in Fig. 9.
4.3 Creating New Knowledge and Expertise
The final stage is the creation of new knowledge and expertise, and the third step
of the methodology supports this stage in order to capture those knowledge and
Break disc
Bearing
Rim
Pipe
Hub
Fig. 7 Hub system assembly Washer

exploded
Fig. 8 Worksheet for Hub system sub-assembly
expertise for then reuse it and improve the DFX process. The tasks involved in
the third methodology step are collecting the expertise and knowledge from
designers, structuring information in rules, programming the rules into the
Expert System and finally, fitting Expert system into the PDM system. The
expert system used in this case study was CLIPS.
4.3.1 Collecting Knowledge

The knowledge captured in this case study was about how to select the most
economic assembly method, which was the first stage in Bootroyd DFA tool
application.
4.3.2 Making Rules

After collecting knowledge is necessary to structure it in a rule way in order to
be able to programming into the Expert System software. To make a rule is
necessary identify the dependent and independent variables for then relate
them. The independent variables defined were, (1) annual production volume
measured in thousands, (2) number of parts in the assembly and (3) total
number of parts. While the dependent variables defined were each one of
the six assembly methods: (1) Special-purpose indexing, (2) Sp.-purp. Free-
transfer, (3) Single-St. one robot arm, (4) Single St. two robot arms, (5) multi-
station with robots and (6) manual bench assembly. The way in which those
Fig. 9 New hub design
variables were related is summarized in the chart below, where VA means

annual production measured in thousands, while NA means number of parts
in the assembly.
Programming Rules Into the Expert System Software

CLIPS software has its own programming language. However, it has also
designed for full integration with other languages such as C and ADA. A
print scream of the codec programmed in CLIPS is shown in Fig. 10.
After to run the program done, the Expert System advices the manual
assembly such as the most economic assembly method. The print scream of
those results is shown below.
4.3.3 Fitting Expert System into PDM System

Due that CLIPS is external software with no interface with the PDM system,
the only way to fit it is upload it as a file. However, each designer must install
CLIPS software in their computers in order to be able to use the Expert System
file when will be necessary.
4.4 Results
The taxonomy presented helps to clarify PLM tools implementation. The
methodology based on the PLM tools taxonomy was measured through a
comparison with the last solar racing car project. The most important benefits
Fig. 10 Expert Systems programming
of the application of this methodology were on the time and cost reduction for
the hub design. These reductions were produced mainly by less time in searching
information, less time spent in meetings, reduction in mistakes made during
modelling process, and prototypes cost reduction.
Less time by searching information and knowledge was achieved with
the use of the PDM system, which allowed to organize and to share all the
information related with the hub through all the designers. In addition, the
Expert System allowed to reuse the experience of the designers. Less spent
time by meetings was achieved through the use of Teamcenter community
platform that established a collaborative environment between the designers.
Reduction on modelling mistakes was achieved with the use of the PLM tools
(CAD) and correct information available in the PDM and expert systems.
Finally, prototypes and testing cost reduction was possible through the use of
Table 3 Summary of Results

Current
Last version version
Spent time by searching information and knowledge 96 hours/ 25 hours/
month month
Spent time by meetings 16 hours/ 2 hours/
month month
Spent time in reworks due at mistakes made during 41 hours/ 8 hours/
modelling process month month
Prototype cost & Testing USD$780.00 0
the CAM-CAE technologies available into the PLM tools. All the depicted
points are summarized in Table 3.
5 Conclusion
It can be concluded that the PLM tools classification into business, engineering
and knowledge management is a good first approach to identify gaps in the
integration of the software used for the different stages of the product life-cycle.
This article shows that it is important to improve the understanding between
the development stages of a product and the ways in which the information and
knowledge can be administrated within and among such stages. Therefore, it is
extremely important to define new frameworks and their respective taxonomies
to improve the implementation of new PLM tools and concepts. In this sense,
the present work has defined a taxonomy to clarify the use of those tools, and
broadens the vision to incorporate new concepts.
In this way, the definition of new taxonomies allows to reach a better
understanding and integration of the different stages that comprise the
integral product development, and the different tools they use. The integra-
tion is not only benefited in the product realization phases, but also in the
globalized collaboration of different organizations that could interact for
such realization.
The conception of the framework was leveraged by the PLM tools taxon-
omy. It is a novel way to bring together Expert Systems, DFX and PDM
systems. These are PLM tools which had not been fully integrated previously.
The value added up to now in this research is that even though a PLM
taxonomy has been defined to support the implementation of PLM concept,
additional research is needed to explore new scenarios of the interconnection
between the applications used within each product life-cycle stage. New PLM
taxonomies are required to develop collaborative environments to foster the
coordination and cooperation among global groups, supported by tools and
methodologies that enable intellectual capital sharing in real time.
Acknowledgments The authors acknowledge the support received from Tecnologico de

Monterrey, Monterrey campus through the Research Chairs in Autotronics, and Purdue
University through the PLM Center for Excellence. Additional support for this work has
been provided by the IBM SUR GRANT. Many thanks to Samantha Rodrguez Bueno &
Mayra Guadalupe Puentes.
Appendix: Several PLM Definitions

Thimm et al. [49]
Product Life-Cycle Management (PLM) is a strategic business approach that

consistently manages all life-cycle stages of a product, commencing with market
requirements through to disposal and recycling (see Fig. 1). PLM involves a
multitude of stake holders (e.g., customers, suppliers, and regulators), who
require various levels of detail and representations of information.
CIMData [50]
PLM is not just a set of technologies, but a strategic business approach that
applies a consistent set of business solutions in support of the collaborative
creation, management, dissemination, and use of product definition informa-
tion across the extended enterprise from concept to end of life integrating
people, processes, business systems, and information. PLM forms the product
information backbone for a company and its extended enterprise.
Kovacs et al. [8]
The PLM concept appeared in the second part of the 1990s. This concept
provides a platform to share product-related knowledge across an extended
enterprise, from product design and creation, through dissemination and after
sales services, up to product dismissal and recycling . . . PLM is defined as a new
integrated business model that, using ICT technologies, implements an inte-
grated cooperative and collaborative management of product related data,
along the entire product lifecycle, dismissal included.
Chiang et al. [51]
Amann et al. (2002) defined PLM as a strategic business approach that applies
a consistent set of business solutions in support of the collaborative creation,
management, dissemination and use of product information across the
extended enterprise from the concept to the end of product life integrating
people, processes, business systems and information. Extending from product
data management (PDM) scope, CIMdata (2003) defined PLM in 11 key

functions. They include (1) document and content management, (2) engineering
change process management, (3) collaborative product design, (4) bill of mate-
rials management, (5) supply chain integration, (6) part classification manage-
ment, (7) product service management, (8) program and project management
(PM), (9) product portfolio management (PPM) and analysis, (10) data author-
ing and analysis, and (11) digital manufacturing.
Juttner et al. [24]
Product life cycle (PLC) management as the integrated, information-driven

approach to all aspects of a products life, from concept to design, manufactur-
ing, maintenance and removal from the market, has become a strategic priority
in many companys boardrooms.
Turban [7]
(PLM) is an integrated, information-driven approach to all aspects of a

products life, from its design through manufacture, deployment, and main-
tenance, culminating in the products removal from service and final disposal.
PLMs goal is to streamline product development and boost innovation in
manufacturing. PLM has the potential to vastly improve a companys ability to
innovate, get products to market, and reduce errors.
Stark [15]
(PLM) is a new activity for manufacturing companies that opens up new

business opportunities. . .
PLM manages each individual product across its lifecycle from cradle to
grave; from the very first idea for the product all the way through until it is
retired and disposed of.
(PLM) enables the companys complete portfolio of product to be managed
in an integrated way. . .
PLM is a holistic business activity addressing many components such as
products, organisational structure, working methods, processes, people, infor-
mation structures and information systems.
Grieves [14].
Product Lifecycle Management is an integrated, information-driven approach

comprised of people, processes/practices, and technology to all aspects of a
products life. From its design through manufacture, deployment and main-
tenance-culminating in the products removal from service and final disposal.
By trading product information for wasted time, energy, and material across
the entire organization and into the supply chain, PLM drives the next genera-
tion of the lean thinking.
References
[1] D. A. Guerra, E. F. Rios, A. Molina, R. Parkin, M. Jackson and N. Elvira, Mecha-
tronics Design Methodology Applied at Manufacturing Companies, presented at the
Mechatronics 2006 The 10th Mechatronics Forum Biennial International Conference,
Malvern, PA, June 1921, 2006.
[2] R. Young, A. Cutting-Decelle, D. A. Guerra, G. Gunendran, B. Das and S. Cochran,
Sharing Manufacturing Information and Knowledge in Design Decision Support, in
Advanced Integrated Design and Manufacturing in Mechanical Engineering, Alan Bram-
ley, Daniel Brissaud, Daniel Coutellier, Chris McMahon (Eds.) Netherlands: Springer,
2005, pp. 173185.
[3] R. Ahmad and F. Yuqing, Managing Product Data and Design flow Process, Proceed-
ings of the sixth International Conference on Intelligent Systems Design and Applica-
tions (ISDA06) pp. 11911196.
[4] R. Polli and V. Cook, Validity of the Product Life, The Journal of Business, Vol. 42,
No. 4 (Oct., 1969), pp. 385400.
[5] P. W. Stonebraker and J. Liao, Supply Chain Integration: Exploring Product and Envir-
onmental Contingencies, Supply Chain Management, Vol. 11, No. 1, 2006, pp. 3444.
[6] B. Yang, N. D. Burns and C. J Backhouse, Postponement: A Review and An Integrated
Framework, International Journal of Operations & Production Management, Vol. 24,
No. 5, pp. 468487, May 2004.
[7] E. Turban, J. E. Aronson and T. P. Liang, Decision Support Systems and Intelligent
Systems 7th ed. New Jersey: Pearson/Prentice Hall, 2005.
[8] G. Kovacs, S. Kopacsi, G. Haidegger and R. Michelini, Ambient Intelligence in
Product Life-cycle Management, Engineering Applications of Artificial Intelligence,
Vol. 19, No. 8, pp. 953965, Dec. 2006.
[9] G. Konstantinov, Emerging Standards for Design Management Systems, in Proceed-
ings of the Computer Standards Conference, Washington, 1988, pp. 1621.
[10] A. Kusiak, Concurrent Engineering: Automation, Tools, and Techniques, John Wiley
& Sons: New York, 1993.
[11] L. Alting, Life-Cycle Design of Products: A New Opportunity for Manufacturing
Enterprises, in Concurrent Engineering: Automation, Tools and Techniques, A. Kusiak
(Ed.), John Wiley & Sons, Inc., 1993, pp. 117.
[12] J. Aca, Reference Model and Methodology to Configure/Reconfigure Integrated
Product, Process and Facility Development Processes, M.S. thesis, Division of Eng.
and Architecture, Tecnologico de Monterrey, Monterrey, Mexico, 2004.
[13] D. Guerra, M. Contero, P. Orta and A. Molina. Industrial Applications of PLM:
Automotive and Aerospace Sectors, Proceedings of Virtual Concept, Playa del Carmen,
Mexico, November 2006.
[14] M. Grieves, Product Lifecycle Management: Driving Next Generation of the Lean Think-
ing. McGraw-Hill: New York, 2005.
[15] J. Stark, Lifecycle Management: 21st Century Paradigm for Product Realisation. London:
Springer-Verlag, 2005.
[16] N. P. Suh, Axiomatic Design: Advances and Applications New York : Oxford Uni-
versity Press, 2001.
[17] K. Aspelund, The Design Process. Fairchild Books, New York. March 30, 2006.
[18] K. N. Otto and K. L. Wood, Product Design: Techniques in Reverse Engineering and New
Product Development. Upper Saddle River: Prentice Hall, 2001.
[19] E. Gonzalez-Mendivil, D. Guerra-Zubiaga, P. Orta and M. Contero, Cross Cultural
Issues on Globally Dispersed Design Team Performance: The PACE Project Experi-
ences. MUDD Design Workshop, May 2007, Harvey Mudd College, California, US.
[20] S. C. Skalac, Implementing Concurrent Engineering in Small Compa1nies. Marcel
Dekker, Inc. (2002) New York, NY
[21] L. Xu, Z. Li, S. Li and F. Tanga, A Decision Support System for Product Design in
Concurrent Engineering, Decision Support Systems, Vol. 42, No. 4, pp. 20292042, Jan.
2007.
[22] A. Kamrani and A. Vijayan, A Methodology for Integrated a Methodology for Inte-
grated Product Development Using Design and Manufacturing Templates. Journal of
Manufacturing Technology Management, Vol. 17, No.5, 2006.
[23] B. Prasad and J. Rogers, A Knowledge-based System Engineering Process for Obtain-
ing Engineering Design Solutions, in Proceedings Of IDETC/CIE 2005, Long Beach,
September 2005, pp. 2428.
[24] U. Juttner, Janet Godsell and M. G. Christopher, Demand Chain Alignment Compe-
tence Delivering Value Through Product Life Cycle Management, Industrial Market-
ing Management, Vol. 35, No. 8, pp. 9891001, Nov. 2006.
[25] B. Gallupe, Knowledge Management Systems: Surveying the landscape, International
Journal of Management Reviews, Vol. 3, No. 1, pp. 6177, Mar. 2001.
[26] J. Gunnlaugsdottir, Seek and You Will Find, Share and You Will Benefit: Organising
Knowledge. International Journal of Information Management, Vol. 23, No. 5,
October 2003, pp. 363380.
[27] J.C. Giarratano and G. D. Riley, Expert Systems: Principles and Programming, 4th ed.
Cambridge, MA: Course Technology, 2004.
[28] A. A. Hopgood, Intelligent Systems for Engineers and Scientists, 2nd ed. Boca Raton:
CRC Press, 2001.
[29] J. Smaros, Forecasting Collaboration in the European Grocery Sector: Observations
From a Case Study. Journal of Operations Management, Vol. 25, No. 3, April 2007,
pp. 702716.
[30] J. H. Park, J. K. Lee and J. S. Yoo, A Framework for Designing the Balanced Supply
Chain Scorecard. European Journal of Information Systems (2005) 14, 335346.
[31] W. J. Kelly, The Risk Managers Survival Guide Risk Management. New York, Apr 2007,
Vol. 54, No. 4; pp. 2026.
[32] H. P. Holzer and H. Norreklit, Management Accounting and Control Systems.
Tijdschrift Voor Economie en Management, Vol. XXXVI, No. 3, 1991.
[33] N. Zarour1, M. Boufaida1, L. Seinturier, P. Estraillier Supporting Virtual Enterprise
Systems Using Agent Coordination, Knowledge and Information Systems 8,
pp. 330349, 2005.
[34] W. Guth and H. Kliemt, Perfect or Bounded Rationality. Some Facts, Speculations and
Proposals. Analyse & Kritik, Vol. 26, No. 2, pp. 364381, 2004.
[35] D. Wright, The dark side of ambient intelligence. Info. Volume 7 number 6 2005 pp. 33-51.
[36] P. L. Emiliani and C. Stepahnidis, Universal Access to Ambient Intelligence Environ-
ments: Opportunities and Challenges for People with Disabilities, IBM Systems Jour-
nal, Vol. 44, No. 3. pp. 605, 2005.
[37] M. L. Socolof and J. R. Geibig Evaluating Human and Ecological Impacts of a Product
Life Cycle: The Complementary Roles of Life-Cycle product Assessment and risk
Assessment Human and Ecological Risk Assessment, Vol. 12, No. 3, p. 510, Jun. 2006.
[38] K. F. Pun Determinants of Environmentally Responsible Operations: A Review,
International Journal of Quality & Reliability Management, Vol. 23, No. 3,
pp. 279297, 2006.
[39] A. J. Russell, Human and Ecological Risk Assessment and Life Cycle Assessment:
Intersections, Collisions, and Future Directions, Human and Ecological Risk Assess-
ment, Vol. 12, No. 3, p. 427, Jun. 2006.
[40] D. Guerra-Zubiaga, R. Ramirez and L. Donato, Digital Manufacturing for Aerospace
Industry: Experimental Aircraft, Proceedings of Virtual Concept 2006, playa del
Carmen, Mexico, November 26th.
[41] D. Guerra-Zubiaga, L. Donato and R. Ramirez, Knowledge Sharing to Support
Collaborative Engineering at PLM Environment, U. Reimer and D. Karagiannis
(Eds.): PAKM, LNCS 4333, pp. 8696, Springer-Verlag Berlin Heidelberg 2006.
[42] D. Guerra-Zubiaga, A Manufacturing Model to Enable Knowledge Maintenance in
Decision Support Systems, Doctoral thesis, Loughborough University, March 2004.
[43] Tsang Data Management for CBM optimization Authors(s): H. C. Albert, W. K. Tsang,
Andrew K.S. Yeung, Jardine, Bartholomew P.K.Leung Journal: Journal of Quality in
Maintenance Engineering Year:2006 Volume: 12 Issue: 1 page: 37-51
[44] M. Bengtsson, Condition Based Maintenance System Technology? Where is Develop-
ment Heading? Euromaintenance 2004? Proceedings of the 17th European Maintenance
Congress, 11th & 13th May, 2004 (Spanish Maintenace Society), Barcelona, Spain, B-19.
580-2004.
[45] F. L. Ramos da Silva and K. L. Cavalca, Combined Application of QFD and VA Tools
in the Product Design Process, The international Journal of Quality & Reliability
Management, Vol. 21, No. 2/3, p. 231, 2004.
[46] K. T. Ulrich and S. D. Eppinger Product Design and Development, McGraw-Hill, 2003.
[47] G. Q. Huang, Design for X: Concurrent Engineering Imperatives, London: Springer-
Verlag, 1996.
[48] E. D. Ramon, Expert System Application into a PLM Framework in Supporting DFX
Process, M.Sc. thesis, Division of Eng. and Architecture, Tecnologico de Monterrey,
Monterrey, Mexico, 2007.
[49] G. Thimm, Towards unified modelling of product life-cycles. Computers in Industry,
Vol. 57, No. 4, pp. 331341, May 2006.
[50] CIMData, CIMdata Reports PLM Market Growth Exceeds Forecasts, CIMdata
Press Releases, April 5, 2006.
[51] T. A. Chiang and A. J. C. Trappey Development of Value Chain Collaborative Model
for Product Lifecycle Management and its LCD Industry Adoption, International
Journal of Production Economics, Vol. 109, No. 12, pp. 90104, Sept. 2007.
Social Issues of Product Lifecycle Management:
Developing Cross Cultural Virtual Teams;
Supporting Todays Green Manufacturing
Imperative; Educating and Preparing
Tomorrows Workforce; and Impacting
Inter-Organizational Relationships in Supply
Chain Management
Cynthia Tomovic, Hannah Anderson, Alyssa Anglin, Ligia-Varinia Barreto,

Sharron A. Frillman, Tanner J. Georgiades, Scott R. Homan,
Jonathan F. Kochert, Magdalena B. Lech, John E. Sukup, Kari L. Wilde, and
Michael Wisma
Abstract Product Lifecycle Management, as an enterprise-wide, information-driven

business approach to reducing wastes and reallocating captured resources in
support of product and process innovation, is not without its challenges. Major
social issues are considered in this chapter, concluding with an example of a
PLM-like application. The first topic for consideration addresses the context in
which PLM is practiced, namely, the virtual environment. In particular, the
challenge of working in a cross-cultural virtual team environment is presented
and a theoretical model in support of creating a successful and productive team
is suggested. The second topic addressed in this chapter proves to be timely
given the current debate on how best to mitigate and adapt to the effects of
global climate change. In this section, PLM is presented as a strategic business
initiative that simultaneously drives green manufacturing while supporting
organizational sustainability. The third topic addresses how best to prepare
tomorrows workforce for a PLM environment. The issue of education is
discussed in terms of a practitioner-based, competency model. Finally, an
example of a PLM-like application in support of the development of inter-
organizational relationships in supply chain management in higher education is
presented.
This chapter represents the collective work of Dr. Cynthia Tomovic and named students.
Each section of this chapter was pressented previously as separate papers by students at the
International Conference on Comprehensive Product Realization, 2007 in Beijing, China.
C. Tomovic (*)
Darden College of Education, Old Dominion University, Norfolk, VA, 23529, USA
e-mail: ctomovic@odu.edu

Keywords PLM virtual teams cross-cultural green manufacturing

competency model PLM education inter-organizational relationships
1 Introduction
Globalization of business is forcing managers to grapple with complex
issues as they seek to gain or sustain a competitive advantage. With
expanding world markets and increased international business competi-
tion, comes a corresponding demand for organizations and individuals to
be knowledgeable about and prepared to operate in the global environ-
ment. Innovation, adaptation, and adoption have determined, to a large
degree, the economic fate of nations, business, and industry working in
a global economy. The key to survival for many organizations is the ability
to capitalize on the potential of advancing new knowledge/technology.
In spite of its promise, however, technology transfer and assimilation
demands substantial changes in the way in which organizations operate.
While technology helps to eliminate a number of barriers to globalization,
many significant barriers remain, notably those involving people and the
organizations we build around them.
The benefits of adopting a new knowledge/technology are undeniable,
however, human factors and organizational relationships either enhance or
constrain potential quality, productivity, and financial gains. Successful orga-
nizations know that they cannot put their faith in new knowledge/technology
alone, as human issues are equally important and often demand even greater
attention than technology.
Product Lifecycle Management (PLM) is an integrated, digital formation-
driven approach to conducting business comprised of people, processes/
practices, and technology affecting all aspects of a products lifecycle, from its
design through manufacturing, deployment and maintenance, culminating in
the products removal from service and final disposal. Be definition, given that
PLM is driven by digital information, users must become somewhat electronic
savvy. However, just as it is important for people to be knowledge about the
technical side of PLM, it is equally important that they be knowledgeable of its
human side. The aim of this chapter is to explore social issues associated with
PLM, namely, the virtual environment as the context in which PLM is con-
ducted; the practice of PLM as a strategy in support of green manufacturing in
the fight against global warming; the education preparation of tomorrows
workforce for a PLM environment; and to illustrate the impact of information
sharing on the development of inter-organizational relationships in supply
chain management. In essence, this chapter suggests that the success of todays
Social Issues of Product Lifecycle Management 137
high-tech initiatives, like PLM, are dependent on high-touch people-driven

considerations.
2 Virtual Environment Context of PLM: A Model in Support

of Cross-Cultural Team Development
The importance of a well functioning cross-cultural virtual team (CCVT) is

imminently important in todays global environment. Having a successful
CCVT can be the difference between being the market leader or struggling to
survive in todays global business arena. The diversity and flexibility of virtual
cross-cultural teams, however, can be difficult to achieve. The introduction of
collaborative PLM tools may prove to be critical to organizational success. From
the components of virtual teams to the cultural elements needed to create a
CCVT, this section of the chapter covers many aspects of CCVT. The Cross-
Cultural Virtual Team Model (CCVTM) presented can be used to create a virtual
team in a multinational corporation and may be tied back to the integration of a
PLM system. A successful PLM system is the key element within the CCVTM.
Conversely, the CCVTM supports and strengthens the PLM system.
2.1 Cross-cultural Virtual Teams

In todays business world, leaders must pay greater attention to communica-
tions and the social system among their employees. It is critical that organiza-
tions control how employees collaborate and coach them in how to interact
with one another. It is challenging to establish Virtual Team Management since
technology is a substitute for human face-to-face interaction. While time and
space constraints are concerns of virtual team environments, companies
launching a global virtual team with PLM support must address issues of
cultural differences among the team members. PLM enables easy access to
information, knowledge, and data sharing; yet some of its users fail to maintain
well organized teams because they have not considered cultural distinctions of
its members. This section provides suggestions for creating cross-cultural vir-
tual teams, as well as, provides examples of difficulties concerning the imple-
mentation of cross-cultural teams.
2.1.1 Components of Virtual Teams

Individuals tend not to consider cultural differences and think that people are
the same across cultures. However, culture has a major impact on the effective-
ness of teams. According to Gannon, team members from the same cultural
background are said to be of average effectiveness; while multicultural teams
Fig. 1 Model of cultural impact on team effectiveness (Adapted from Adler (1983))
tend to be either highly effective or ineffective (Gannon 2001). Citing Kovach,

Gannon suggests that the level of effectiveness is directly related to the leaders
role in choosing team members so as to avoid the risk of low effectiveness. This
paradox is displayed on Fig. 1 that shows the relative productivity of 800 teams
consisting of 4 to 6 members (Gannon 2001). This figure illustrates that certain
cultural factors may aid or present team effectiveness. While single-culture
groups are limited in their effectiveness level, multi-cultural teams have the
potential to increase productivity.
This model suggests that multi-national teams are an asset in todays global
environment as they provide high productivity and different approaches to
tasks. However, individuals from different cultures vary in terms of their com-
munication and group behaviors, including the motivation to seek and disclose
individuating information and the need to engage in self-categorization, iden-
tify with a particular group (Gudykunst 1997). According to Hofstede, one of
the dimensions of cultural variability is individualism-collectivism. Individua-
listic culture members tend to value the needs, values, and goals of the indivi-
dual over the needs, values, and goals of the group. In collectivist cultures there
is an opposite tendency (Hofstede 1980). Hofstede suggests that members of
individualistic cultures are likely to be less concerned with self-categorizing, are
less influenced by group membership, can easily adjust to changing groups, and
can involve themselves in more open and confident communication than their
counterparts from collectivist cultures (Hofstede 1980). Members from both
types of cultures represent a vast array of experience and training which, if
managed appropriately, could collectively lead to greater team performance
due to the richer synergy created based on a broader spectrum of viewpoints.
When having such diverse cultural backgrounds in one team, there are essential
factors that link all members together and enable effective management of the
team. These important factors, presented by Hastings (Hastings et al. 1994)
include:
Trust team members expect their partners to deliver to agreed times and
specifications. Creating trust among team members that represent different
attitudes may not be easy, knowledge and a willingness to work together to
overcome cultural differences for mutual benefit is required (Cartwright 2003).

Members reaction to particular stimuli in the first communication events often
produces patterns that may prove to be long-lasting in a team. Thus, it is
important to encourage team members to be tolerant, to learn about and from
one another, and to be willing to adapt and change communication behavior
accordingly. Regarding the electronic environment, it has been suggested that
electronic communication, without the ability to send non-verbal cues, e.g.,
gestures, accents, dress codes, etc., may actually enhance team member trust
and hence impact productivity (Jarvenpaa and Leidner 1999).
Control choosing the priorities for control by a team and leaving out some
elements for an individual to decide on priority. A teams responsibility lies
mostly in meeting objectives and remaining within budget. Teams need to be
focused, especially in a virtual environment. The tactic of filling in the subject
line on everyday messaging helps to maintain focus on the task (Lipnack and
Stamps 2000).
Motivation motivation is a critical element for a high performing project
team, especially in a virtual environment. Virtual team members need to
overcome the sense of isolation (Nauman and Iqbal 2005). There is no
presence of physical contact and so the team needs to look for substitutes.
Chat rooms and video communications helps to support team bonding
(Hastings et al. 1994).
Communication differences in culture and language hinders communica-
tion. When co-workers are dispersed, socializing is significantly reduced.
They rely heavily on information technology to communicate (Nauman
and Iqbal 2005). Cartwright describes one virtual team he worked with in
which members prepared cue cards that were held up during the videocon-
ference. They read: LAUGH, RUBBISH, WHAT?, and other replies
and were found very effective (Cartwright 2003).
Nauman and Iqbal (Nauman and Iqbal 2005), in their article on challenges
of virtual project management, present some recommendations for improving
the virtual processes:
Members in virtual team environments often have expertise in a specific area,
so there is great need for knowledge sharing via effective communication and
knowledge management techniques.
Face-to-face communication in the beginning of forming a team is an
essential condition in establishing higher levels of trust and motivation
among specialists working from geographically dispersed locations.
Managers or team leaders must be communication mediators between the
members of virtual teams in order to reduce conflict.
A single communication point is a must to avoid idleness and conflict.
The suitable use of telephones, video-conferencing, and face-to-face meet-
ings should be considered vital for effective communication.
Leaders should effectively implement roles and responsibilities which are
essential.
2.1.2 Creating Virtual Teams

In order for a well performing group to be formed it needs to go through several
stages. There are five steps: forming, storming, norming, performing, and
adjourning. Some groups go through the stages in a different order or omit
some of them, but this model is good at describing a groups development
(Tuckman, 1965). The forming stage is best accomplished if the team members
meet physically if possible. This effort is said to be expensive yet it pays off when
the team starts working on the project. In this stage, trust is already present owing
to the fact that the members have already been acquainted with one another.
Many companies initiate the project by bringing people together and allowing
them to interact socially. In the case of conflict where it is hard to define an
attitude of an individual on the other side of instant messenger, teams need to rely
on the individual making their feelings known to the group (Cartwright 2003).
According to Duarte and Snyder (Duarte and Tennant-Snyder 2001) there
are six essential competencies in creating virtual teams:
1. Performance Management and Coaching (team, individual performance,
compensation)
2. Appropriate use of Information Technology
3. Managing Across Cultures
4. Aiding in Team Members Career Development and Transition.
5. Building and Maintaining Trust and Networking
6. Developing and Adapting Standard Team Processes.
2.2 Introduction to Cross-cultural Studies
Since virtual teams are such an integral part of the PLM process, it is important
to create them in such a way as to achieve a high level of communication, trust,
and coordination. When forming virtual teams across cultures, rather than
groups that are formed within the same culture and/or country, some variables
come into account that can dramatically affect group performance. In order to
understand these variables, two different cultural studies are considered; Geert
Hofstedes study of Cultural Variables and the Project GLOBE study. After a
thorough review, the most important lessons will be selected and compared with
what was learned about effective virtual team construction. The overlaps
between these studies form the basis upon which the present cross-culture
virtual team model was constructed. A discussion of the two studies and their
relevance to the development of the model follows.
2.2.1 Hofstedes Value Dimensions

Geert Hofstede is a Dutch expert on the interactions between cultures, be it
national cultures or organizational cultures. He developed his system over the
course of several years from 1966. In his book, Cultures Consequences he

posits that his research is critical to understanding not only organizational
continuance, but to understanding mankind itself.
Hofstede believes that the survival of mankind will revolve around people
learning to work together and accept others cultural differences. He goes on to
state that within the confines of human nature, understanding the thinking of
a foreign culture is not an intellectual luxury. Instead, he believes that gaining a
heightened understanding of vague cultural differences is one of the main con-
tributions the social sciences can make to practical policy makers in governments,
organizations, and institutionsand to ordinary citizens (Hofstede 1980).
It is important to note that Hofstedes book was written in 1980, over
27 years ago before the rise of globalization as it exists today. Awareness and
interest in global business has flourished since Hofstedes writings. His exten-
sive work with IBM and later, the work of other researchers, led to the
construction of a four dimensional framework (with a later fifth added) for
assessing cultural differences. Hofstedes five dimensions are:
Power Distance This dimension addresses how less powerful organizational
members both accept and/or expect the distribution of power to not be equal.
It is defined from the perspective of the subordinate and not the superior of
that subordinate. What this represents is whether or not a society endorses and
accepts a disproportion of power by subordinates as well as superiors. While
most societies differ in this dimension, sometimes the difference is extreme (i.e.
some cultures Power Distance is so great that questioning authority in any
way can lead to major consequences, whereas in others it is so small that the
concept of superior and subordinate are vague or non-existent).
Individualism The individualism dimension shows the level of group
integration within an organization. Highly individualistic cultures value
personal responsibility, e.g., work by ones self, emphasis on and the
taking care of ones own needs without extensively relying on other
people. On the other hand, cultures ranking low in individualism are
more prone to be bonded through group activity (collectivism). Collec-
tive cultures support the development of a collectivist attitude from
birth and it continually reinforced in the society. Collective cultures
often base much of their society on rewarding the actions of the group,
where as individualistic cultures reward individual effort.
Masculinity Masculinity is a gender-based dimension in which behavior is
described as being more male or female-like in expression. Hofstedes study
revealed that there is a difference across cultures in the degree to which
expectations of males and females are differentiated. In less differentiated
cultures, males and females tend to act similarly, and both tend to express
values and behaviors that western cultures associate with females, e.g.,
modesty, caring. In highly differentiated cultures, there is a greater degree
of differentiation in male and female behavior. Both males and females tend
to express values and behaviors that western cultures associate with male
behaviors, e.g., assertiveness, competitive, though females in these cultures

tend to express these tendencies less than their male counter parts.
Uncertainty Avoidance Some cultures are based on highly structured ways
of doing things and are intolerant of change while others are more open to
accepting change and the inherent risk of the unknown. This dimension can
also describe a societys aversion to accepting values that differ from their
own (e.g. religion). Cultures that rank high in uncertainty avoidance tend to
develop highly structure rule-bound societies. People within these cultures
tend to hold a believe that there is little they can do to shape their future, they
express a high locus of control. In contrast, societies ranking low in this
dimension are more open to pursuits that have never been undertaken and
have no specific set of rules to follow, people tend to think they create their
opportunities, and as such, express a low locus of control.
Long-Term Orientation (only recently applied to 23 countries) As the name
implies, long-term oriented cultures tend to base their decisions on outcomes
in the future. Values such as perseverance, future ambitions, and fulfilling
ones life goals are expressed in cultures with long-term orientations. Con-
versely, short-term oriented cultures (or ones that rank low in the long-term
orientation dimension) are based more on fulfilling obligations that will have
a more immediate or near term impact (Hofstede 1980).
It is fairly easy to see how Hofstedes cultural variables can assist in the
development of creating cross-cultural teams. For example, should an indivi-
dual from a country with high power distance be placed in a team where the
group leader is from a country where power distance is low, the possibility that
the leader finds the subordinate lacking in initiative is great.
2.2.2 Project GLOBE

Project GLOBE (Global Leadership and Organizational Behavior Effective-
ness) is a major long-term multiphase, multi-method research project that
studied cross-cultural leadership differences and similarities among countries.
The studys objective was to quantify data and score it based on nine cultural
attributes and six major leadership behaviors. The 62 countries involved in
Project GLOBE amassed nearly 17,000 responses from managers in 951 orga-
nizations (House et al. 2004). The ongoing study of Project GLOBE has been
very successful in defining human behaviors across different cultures. Questions
asked include the following:
1. What qualities affect leadership?
2. What behaviors are universally accepted?
Each member in the project is referred to as a country co-investigator (CCI) and
is responsible for their particular country. CCIs are responsible for studying specific
cultures in which they have a certain level of expertise, meaning that they often are
native to the region or have lived there long enough to understand local cultures
thoroughly. CCIs collect both qualitative and quantitative data which they use to
check the validity of their questionnaires, write descriptions of their specific coun-
trys culture (usually in context of the data), and give personal insights into their
experiences within the culture. The CCIs are usually natives of the culture in which
they collect information and most cultures contain 2 to 5 CCIs (House et al. 2007).
With the information collected by these research teams, Project GLOBE has
formed nine measurable dimensions for differentiating between societal and
organizational values (Javidan et al. 2006):
Performance Orientation level to which a society encourages and rewards
people for performance excellence.
Assertiveness extent to which a culture endorses speaking ones mind and
being confrontational versus being quiet and modest.
Future Orientation way in which a culture rewards future oriented behavior
such as strategic planning and investing.
Humane Orientation extent to which a society encourages and rewards
individuals for being compassionate towards others.
Institutional Collectivism level of loyalty felt towards the organization and
its goals rather than ones own individual goals.
In-group Collectivism level of pride found in belonging to small groups
within a society (i.e. work groups, family, religious group, etc.)
Gender Egalitarianism way a culture endorses gender role differentiation.
Power Distance level of equality distinction between superiors and their
subordinates in terms of power, responsibility, and authority.
Uncertainty Avoidance societys reliance on social norms to avoid unpre-
dictable future events.
2.3 Cross-cultural Virtual Team Success Model
The CCVTM below illustrates that the four major factors of virtual team
success are clearly dependent on some type of collaborative tool, in this case,
PLM and all of the technologies that accompany it. Without PLM these factors
would not be achieved easily and in the same token these factors fuel PLM itself.
For example, open communication between all departments one aspect of
PLM functioning most certainly depends on trust, motivation, communica-
tion, and control.
As described in the previous sections, these factors are measured based on
cultural aspects. When creating a cross-cultural virtual team these dimensions
should be considered when selecting team members, or when developing cross-
cultural training for potential team members. Consideration of these dimen-
sions will likely increase the probability of team success within a cross-cultural
virtual team environment.
The four major factors of virtual team are extremely dependent upon one
another. For instance, if there is a lack of trust then open communication will
Fig. 2 Cross-cultural virtual team success model
most certainly falter. When just one of the four factors is compromised the
cross-cultural virtual team will be in danger of self destruction. By addressing
cultural dimensions as described, cross-cultural team productivity is likely to
increase in a PLM environment (Fig. 2).
2.3.1 Trust
Creating trust among team members representing different cultural back-
grounds is not easy (Catwright 2003). Building trust among team members
may be facilitated by knowledge of power distance, in-group collectivism, and
institutional collectivism. For example, representatives of a culture with strong
in-group collectivism take pride in membership in small groups such as their
family and close circle of friends, and the organizations in which they are
employed. By definition, they reveal less trust to people from the outside. It
may prove to be more difficult for members from a strong-in-group collective
culture to embrace new assignments at first, but later they are likely to prove
quite loyal. In essence, trust arises with knowledge gained about the members
on the virtual team (Jarvenpaa and Leidner 1999).
2.3.2 Control
Depending on the kind of task, team members may be responsible for control
over the task elements; however, objectives and meeting budgets are controlled
by the team as a whole (Lipnack and Stamps 2000). Power distance and
institutional collectivism are dimensions which when properly interpreted in
the beginning and used accordingly, reveal prompts for team forming. Some
team members may originate from high power distance and high institutional
collectivism cultures, which means they may be reluctant to gain control over
any aspects of team work since they prefer a leader to be in charge and set rules
and regulations. Members of low power distance culture would be enthusiastic
to have an opportunity to manage particular parts of a project without too
much control from authorities.
2.3.3 Motivation
In a virtual environment, motivation is a significant element for a successful
virtual project team. No physical contact is present in the process of coopera-
tion and depending on the backgrounds, the virtual team members handle this
problem differently. To be more aware of these distinctions, dimensions such as
in-group collectivism, institutional collectivism, performance orientation, and
future orientation need to be interpreted properly. For instance, in individua-
listic cultures with high performance, society rewards individual members for
performance excellence and encourages them to be more confident. This
method would not harmonize with individuals from collective cultures that
have lower performance orientations in which society rewards member and
encourages them to attend to family issues over job performance.
2.3.4 Communication
In the virtual team environment with no physical contact, communication is
dependent on information technology (Nauman and Iqbal 2005). The dimen-
sions that facilitate better understanding of the communication differences
include power distance, assertiveness, and gender egalitarianism. For instance,
members from countries that pay more attention to gender role differences,
where there are relatively few women in positions of authority, might find it
hard to cooperate with women who come from an environment with a society
that provides males and females equal opportunity. In order to improve com-
munication among the team members, behavior rules and explanations may
need to be introduced as a matter of protocol.
2.4 Implementation of the Cross Cultural Virtual Team Model

in Fictitious Environment
The CIO of a multinational technology firm was assigned the task of creating a
virtual team consisting of representatives of the companys offices in different
countries. This cross-cultural virtual teams purpose was to document the IT
systems of all company divisions. About a year before this team was to be
created, the company had decided to introduce Product Lifecycle Management
(PLM). Manager responsible for creating a cross-cultural virtual team may
likely take the following actions.
First, the manager may order a demographic analysis of the employees in his/
her company, with special interest in the following cultural influences:
Power Distance
In-group Collectivism
Institutional Collectivism
Performance Orientation
Future Orientation
Assertiveness
Gender Egalitarianism.
Once the results of this demographic analysis were tallied the manager used
them and other personal measurements and past records to best choose indivi-
duals that were likely to promote the four factors of a successful virtual team;
trust, communication, control, and motivation.
Once the members of the cross-cultural virtual team were chosen, the man-
ager started working on building the four factors into the framework of the
team. The manager set up a four day conference off site. The goal of this
meeting was to introduce team members face-to-face and to jump start the
bonding process. Team building exercises were conducted to encourage trust,
motivation, control, and communication among the team members.
After the conference the team members went back to their respective coun-
tries and divisions and began working as a virtual team. The PLM technology
was in place and was available to support cross-cultural communications within
the team. The communication between members was facilitated by video con-
ferencing, e-mails, instant messaging, and phone conversations. Video confer-
encing helped the virtual team members feel as though they were in the same
room as the other individuals.
As the team matured, benefits of PLM began to improve due to the fact that
the communication, trust, control, and motivation levels were higher than when
the team first formed. On the occasion when lack of communication occurred, it
was repaired by the trust, motivation, and control that the cross-cultural virtual
team created among its team members. The feedback loops in this model helped
the company create a cross-cultural virtual team that could successfully com-
plete the task and more; the team not only created a complete documentation
system per its original assignment, but it also went on to other company wide
improvement projects.
Attention to cross-cultural dimensions is important when Implementing
PLM. When using both CCVTs and PLM a multinational corporation will be
able to compete in all areas of their business. When implemented correctly
following the CCVTM both the CCVTs and the PLM approach should grow
stronger over time. An early investment in the time required to create a
potentially successful CCVT will allow, later on, multinational corporation to

concentrate on other business objectives.
3 Product Lifecycle Management in Support of Green

Manufacturing: Addressing the Challenges of Global Climate
Change
The rapid consumption of natural resources and the growing interest in global
warming (Kharin et al. 2007) have motivated more and more companies to
change their manufacturing strategies and processes in order to become more
ecologically sound. In general, there is a desire to develop innovative manufac-
turing systems that align with green manufacturing goals.
Green manufacturing reduces or eliminates the use and generation of hazar-
dous substances throughout all phases of a products lifecycle (University of
Alabama 2006). In addition to being environmentally friendly, green manufac-
turing has a positive effect on the bottom line; its implementation lowers costs,
improves production lead time, and increases product quality (Noci 1995).
Green manufacturing is seen as a competitive advantage by companies who
can efficiently use financial resources, technological knowledge, and operations
to implement green manufacturing practices.
In this section of the chapter, we investigate why green manufacturing
processes are beneficial to companies and which green activities are currently
in place. A summary of financial impacts, current sustainability practices, and
regulations and policies associated with green manufacturing is presented. An
argument is made that Product Lifecycle Management (PLM) can support
green manufacturing strategic initiatives.
3.1 Background
A pilot study was conducted in which green manufacturing practices were

investigated. Six companies were selected from the Global Round Table on
Climate Change (GRTCC). According to the GRTCC website (Global Round
Table on Climate Change 2006), they bring together stakeholders from all
regions of the world to discuss and explore areas of potential consensus regard-
ing core scientific, technological, and economic issues critical to shaping sound
public policies on climate change. This study investigates green manufacturing,
the adoption of environmentally-friendly standards, and influences to the PLM
process because of green activities. As background, we provide an overview of
these aspects as well as a summary of the companies used in this study.
Green manufacturing identifies methods that minimize waste and pollution
during product design and production. Its main goal is sustainability by
reducing waste on the spot and by promoting recycling such that every
company does it part in conserving natural resources. Green trends have

caused manufacturers business needs to change significantly over the past
and are projected to have an even greater impact given the challenges pre-
sented from global climate change. In fact, green technology has been identi-
fied recently as a key factor that will affect the overall success of a company
(Cimalore 2007).
Environmental standards and regulations influence organizations to follow
environmental procedures. Organizations must comply with governmental
regulations or face immediate and severe consequences (Grieves 2006). In
the past, multinational companies tended to operate with low standards
(Hassan, et al. 2002). However, more recently, some companies have joined
with others to self-regulate and to adopt polices that surpass US governmental
regulations (Christmann 2004).
Product Lifecycle Management is an integrated, information-driven
approach to improve a products life. It achieves efficiency by using a shared
information core system that helps a business to efficiently manage a products
life cycle from design to disposal (Grieves 2006). Thus, PLM support businesses
as they reduce the use of wasted materials and energy.
3.1.1 Our Companies

To obtain the six companies in this study, an initial set of 10 companies was
chosen at random. The companies were carefully inspected in order to secure
that enough information was available to conduct an analysis of the companies
green manufacturing practices. All research is based on publicly available data
e.g., (Bayer 2006), (Bayer Sustainable Development Report 2005), (DuPont
2007), (Florida Power and Light 2007), (General Electric, 2007), (The Dow
Chemical Company 2006), (The Dow Chemical Company 2005), (Toyota
2007). The companies include:
FPL Group, Inc. one of the largest providers of electricity-related services.
Its principal subsidiary, Florida Power & Light Company, serves more than
8 million people along the eastern seaboard and southern portion of Florida.
DuPont a company operating in more than 70 countries and offering a wide
range of innovative products and services spanning many markets.
General Electric (GE) a multinational company that participates in a wide
array of markets. GE is dedicated to turning imaginative ideas into products;
in 2006, it recorded revenues of $163.4 billion.
Toyota a multinational corporation and the second largest automaker in
the world. Toyota Corporation encompasses Toyota, Lexus, Scion, and
parts of Daihatsu brands, divisions and companies.
The Dow Chemical Company incorporated in 1897 and has since expanded
into over 175 countries, employs 43,000 people worldwide, and has annual
sales of $49 billion.
Bayer a worldwide company that employs 106,000 employees and in 2006

had 29 million EU in net sales. Bayers corporate mission statement includes
the slogan, Bayer: Science for a Better Life.
3.2 Financial Impacts
It is of importance to keep up with a products functionality and cost during the

life cycle of the product (Kimura and Kato 2003). Two of the biggest benefits
for a company to pursue green manufacturing are additional cost saving and
production increases. This is possible because a good plan enables a company to
spend less money by preventing unnecessary waste rather spending additional
money to dispose waste (e.g., zero-emission strategy). Alternatively, the com-
pany may reuse components or materials which represent more savings. Man-
agers consider improvements in environmental performance one of the basic
competitive priorities along with low operational costs and production lead
time for higher quality (Azzone and Noci 1998). Moreover, developing efficient
(and green) product and process technologies constitutes the basic cost and
ecological impact parameters. The processes drive the types of raw materials
used, workers health and safety, ecological risk, materials efficiency, and waste
treatment.
In order to achieve best sustainable practices, partners have to cooperate and
tap into the know-how of all parties at all stages of a product. To minimize the
risks and to secure the maximum result, all of them should be part of the value-
adding processes (Westkamper et al. 2001).
3.3 Current Sustainable Practices
A summary of the sustainability practices is provided. Sustainability has sig-

nificant potential for improving environmental friendliness while simulta-
neously driving improvements in productivity and costs.
3.3.1 Recycling and Disposal

A comprehensive recycling effort and disposal practice is crucial for reducing
waste. The emergence of product take-back legislation in Europe and current
directives in the US has forced companies to think about how to dispose their
products through efficient recycling, disassembly, and reuse. There is also a
focus on safe disposal of hazardous materials or substituting those materials for
more environmentally friendly materials (Grieves 2006).
Our companies recycling and disposal activities are summarized (Table 1).
An entry in the table indicates a green activity is performed; when known,
additional details are included in the cell. Dow Chemicals main recycled
Table 1 Recycling and Disposal. This table indicates the presence of recycling activities and
active green disposal practices.
Energy
Companies Wind generator Landfill gas Conserv. Reduction in use
FPL X X X X
DuPont
GE X
Toyota
Bayer X
Dow X
products are their plastics. In fact, North America, Europe, and Asia are all
markets that recycle billions of pounds of plastics. Toyota uses a set of recycling
guidelines, Toyota Recycle Vision, to help minimize their negative environ-
mental impact. Toyota adjusted product design in the 2004 Sienna minivan and
the Camry Solara coupe to replace substances of concern with greener
materials.
3.3.2 Energy Consumption

Energy consumption is clearly important in green manufacturing (Table 2).
The electricity industry is the single largest source of industrial pollution in the
world, and one of the largest sources of greenhouse gas emissions. Over 80% of
the worlds electricity and 90% of US electricity comes from nonrenewable
fossil and nuclear sources (Starrs 2005). FLP is the US largest developer,
owner, and operator of wind powered generating plants. Half of the energy they
produce comes from natural gas and only 10% comes from oil.
General Electric has pursued developing engines that use an effective and
environmentally-friendly energy generation method from landfill gases. When
organic substances decompose, methane, carbon dioxide and nitrogen are
created. Landfill gas is a high-quality alternative for gas engines.
Bayer uses organic methods to create valuable energy. A new project
launched in 2005 treats sewer sludge so that bio-gas can be extracted from the
Table 2 Energy Consumption. This table indicates energy sources used and green practices
are active.
Energy
Companies Wind generator Landfill gas Conserv. Reduction in use
FPL X X X X
DuPont
GE X
Toyota
Bayer X
Dow X
organic content and used for energy generation. The European Commission
is supporting the development of this process as part of the Life environment
program.
Since 1994, Dow Chemical has reduced energy consumption by approxi-
mately 900 trillion BTUs. Their current goal is a 20% reduction in the energy
needed to fabricate one pound of product.
3.3.3 Water and Air Management

Water and air management has received considerable attention (Table 3).
DuPont strives to reduce and conserve water at all international and domestic
sites. Moreover, DuPont commits to reduce water use by at least 30% over the
next ten years at global sites that are located where the renewable freshwater
supply is either scarce or stressed as determined by the United Nations. For all
other sites, DuPont holds water consumption flat on an absolute basis through
the year 2015, offsetting any increased demand from production volume growth
through conservation, reuse, and recycle.
GEs Ecomagination program, involves financial and environmental perfor-
mance to drive the companys growth. It has played a key role in advancing
water reuse and purification technology. GEs improved membrane materials
and spiral-wound membrane element configurations has revitalized its water
reuse and purification system to address fouling, temperature, pH-value, and
contaminants.
Toyotas green complex in California conserves more than 11 million
gallons water annually. The plant uses special pipelines that supply recycled
water for cooling and landscaping.
Bayer reduced water usage from 2.2 million cubic meters per day
(19942004) to 1.2 million cubic meters per day in 2005. The companys
wastewater treatment plant treats contaminated water and reuses it.
Dow Chemical has made a reduction in water use by 38% since 1994. They
eliminate wastewater at the source. Dow states they are above the goal to
reduce wastewater intensity by 50% and are committed to continue source
reduction and recycle efforts to reduce fresh water consumption. Not only
Table 3 Water and Air Management. This table indicates green activities in water/air
management.
Water Air
Companies Conserv. Reuse Recycle/Treat Lower emissions
FPL X
DuPont X X
GE X
Toyota X X
Bayer X lower green house gases
Dow reduce waste water X lower green house gases
does Dow focus on eliminating water at the source, but they have developed
activities to recycle wastewater wherever feasible:
Groundwater Water created streams from underground water via wells,
Fresh Surface Water water captured from canals, lakes, and rivers,
Rainwater captured or stored rainwater,
Site Level Recycle on-site water is reused, avoiding new supply,
Seawater & Brackish surface or groundwater is captured from brackish or
seawater sources, and
Purchased Steam and Condensate Water use purchased water on site.
3.3.4 Products and Processes

Products and operational processes of the companies investigated have pro-
duced noteworthy environmentally friendly characteristics (Table 4). For
example, FLP joined forces with the US Climate Action Partnership
(USCAP) to ask the US Government to endorse mandatory policies to reduce
carbon dioxide emissions. FPL endorsed the Joint Statement of the GRTCC
committing itself to global action to alleviate global warming risks while meet-
ing the need for energy, economic growth, and sustainable development world-
wide. Furthermore, FLP is ambitious to become a leading green company by
using natural gas as fuel to produce 50% of their electricity. Currently, nuclear
sources constitute 20% and oil sources only produce 8% of total energy
production. FLP has added wind sources which are soon expected to bring
significant financial benefits.
Since 1994 DuPont has reduced their gas emissions measured as CO2 equiva-
lents by 72% and have reduced global air carcinogen emissions by 92%. By
2015, they plan to reduce their emissions at least 50% from a base year of 2004.
DuPont will introduce fleet vehicles which use the leading technologies for fuel
efficiency and fossil fuel alternatives. All together, this will bring DuPonts total
reductions since 1990 to 96%.
Toyota leads the field in lowering emissions and improving fuel economy
in gasoline-powered vehicles. Toyota not only developed and produced the
Table 4 Products and Processes. This table indicates which companies have active green
products and operational processes
Product/Processes
Companies Energy efficient Fuel efficient Operational Efficient
FPL
DuPont X X
GE X X X
Toyota X X
Bayer X X (info availability)
Dow X
worlds first mass-produced gas/electric hybrid car, but is also pioneering fuel
cell cars to further reduce air pollutants.
Bayer cut greenhouse gas emissions to 5.6 millions metric tons of CO2
equivalents in 2004 and 3.9 million metric tons of CO2 equivalents in 2005 as
explained by Bayers Sustainability Report for that year. In total, greenhouse
gas emissions throughout the Bayer Group decreased by over 70% from 1990 to
2005.
Dows production has increased by 32% since 1994, their total emissions of
CO2 have been successfully reduced by 32%. Dow has achieved this through
long-term energy efficiency improvements and by converting to more climate-
friendly technologies.
3.4 Regulations and Policies
Many companies are seeking to improve their environmental performance in

order to join consortia, associations, and projects and to comply with environ-
mental regulations (DuPont 2007). Satisfying the regulations and policies is not
easy. For example, the End of Life Vehicle (ELV) regulation requires that by
2006 all automakers in the European Union recover 85% of a vehicles weight
and 95% by 2015. In addition, no lead, mercury, cadmium, or hexavalent
chromium can be used after July 1, 2003.
Internal and external factors influence a company to follow environmentally
friendly standards. External factors include customers, competitors, and reg-
ulations. This last factor induces manufacturing to rapidly implement green
processes (Udomleartprasert 2004). Having environmentally friendly standards
creates a positive reputation with the public, saves cost, and promotes new
designs. Internal factors include ethical objectives and the desire to achieve
competitive advantage.
General Electric, Bayer, and Dow follow the Waste Electrical and Electronic
Equipment (WEEE) regulation when developing electronic devices. The WEEE
directive requires manufacturers of electrical and electronic equipment to
recycle their products at no consumer cost and prohibits the use of certain
electrical materials. In addition, both Bayer and Dow abide by the RoHS
Directive which prohibits the use of certain hazardous substances in electrical
and electronic equipment. The directive bans from the EU market any new
electrical and electronic equipment containing more than agreed levels of lead,
cadmium, mercury, hexavalent chromium, polybrominated biphenyl, and poly-
brominated diphenyl ether flame retardants (RoHS 2007).
DuPont follows the International Standards Organization (ISO) and agrees
on specifications and criteria that they will consistently apply to all stages of a
product lifein the classification of materials, in the manufacture and supply of
products, in testing and analysis, and in the provision of services. To maintain
effective management systems for health and safety protection, Bayer abides to
the Health, Safety, Environment and Quality (HSEQ). The Sarbanes-Oxley Act
of 2002 (SOX) requires all executives to certify the accuracy of their financial
statements and attest to the processes, controls, and systems used in financial
reports.
As the world moves to more standardized international regulations, organi-
zations will be required to track more information. Furthermore, integrating
stronger and efficient green activities complicates company planning. This
brings us to the next section on PLM. PLM will be an integral part of the
information systems executives will need to have in place in order to abide by
the required regulations and to develop sustainable green practices.
3.5 Product Lifecycle Management

Environmental changes and the implementation of green manufacturing
require a company to change its entire processes; from product planning and
procurement policies to production and logistics (Azzone and Noci 1998). The
PLM model, as described by (Grieves 2006) includes five major categories
(plan, design, build, support and dispose) centered by an information core
(Fig. 3). PLM is about product data, information and knowledge combined
with people, processes, and technology. PLM can benefit a company practicing
green manufacturing. For example, the information core will help a company
capture and organize information, avoid the use of illegal materials, and help
identify processes with negative environmental impact. As recently stated, The
Eco & PLM database manages data about all stages of the product lifecycle
from procurement to disposal or recycling. This enables the evaluation of
whether the product complies with the environmental laws. It also facilitates
the recording of environmental data and makes quick product recall possible
(Minami et al. 2004).
Fig. 3 Relation between

green initiatives/PLM. PLM
model taken from Grieves
(2006)
Similarities exist between environmental protection activities and the opera-

tions methods already in place throughout a products lifecycle. Programs
explaining how to keep pollution under control, reduce waste, or develop
green designs may also reinforce traditional operations management such as
statistical process control, total quality control and total quality management
or design for manufacturability (Burgos and Cespedes 2001).
It is essential to design the product as a whole from planning, through design
and manufacturing, to usage, maintenance and reuse/recycling/disposal. A
sound strategy for product maintenance and improvement during usage should
be established, and all the life cycle processes are to be well controlled. By such
approach, reuse/recycling activities are also rationalized, made visible, and
controllable. Such an approach, called Inverse Manufacturing, stresses the
controllability of reuse/recycling processes, closes product life cycles, including
maintenance which are also are pre-planned and controlled (Kimura and Kato
2003).
During the planning and design phases, specifications, requirements, and
functionality of a product need to be determined. The PLM information core,
not only would store data and design information to make it accessible, but also
a library of green directives and legal issues. When building a new product,
PLM can provide information on facility lay-outs and knowledge of existing
equipment. This information can pre-determine which facilities are best suited
to practice green manufacturing. The support of a product relies heavily on
sales and distribution. PLM would provide information on warranties, product
performance, hiring practices, and other product data that would make the
support function more effective.
The disposal phase is the final stage of a products life. PLM provides
information on product design and material make-up which is vital for the
recycle and disposal phase. Simultaneously, methods of efficient product dis-
posal and recycling are also related to the planning and design phases of PLM.
Thus, the accessibility and information sharing that PLM provides supports
green manufacturing because not only does it assist companies compliance
with environmental directives and regulations, but it also greatly decreases
wasted time, energy, and materials.
3.6 Observations and Conclusions

Green manufacturing changes are not easy and they may not represent an
economic benefit in the beginning, however, they do represent a good long-
term investment. The following observations are made about the companies in
the pilot study:
FLP, an important energy producer for the US, is engaged in substantial
green energy efforts. More than half of its energy comes from natural gas and
more green expansions are planned.
DuPont has committed that by 2015 it will increase revenues from energy
efficient processes and/or significantly reduce green house gas emissions.
General Electric has become a leading company in developing green pro-
ducts and processes through their Ecomagination program. Furthermore,
GEs belief that financial and environmental performance can work together
to drive company growth will enable them to continue to undertake the
worlds biggest environmental challenges.
Toyota leads green efforts in the automotive industry and is working
towards a Toyota Recycle Vision. Many of Toyotas goals have been
reached due to their global earth charter that promotes environmental
responsibility throughout the entire company.
The Dow Chemical Company implementation of energy efficient processes,
chemical recycling programs, waste water reduction, and lowered green
house gas emissions prominently demonstrate their commitment to green
manufacturing.
Bayer is working towards sustainable development by using energy effi-
ciently, increasing water conservation, more safely disposing hazardous
wastes, and lowering green house gas emissions. Their goals are sought
with efficient resource management and future-oriented climate protection
activities.
PLMs objectives emphasize not to waste time, energy or materials. Green
manufacturing also looks for the conservation of resources and reduction of
waste. Thus, PLM can support green manufacturing procedures by tightly
integrating such practices into the lifecycle of a product and yielding environ-
mentally friendly productions.
The environmentally-friendly practices identified indicate green manufactur-
ing is already a very actively pursued objective. However, to compete in a global
world and to comply with the many environmental protection regulations and
standards is not easy. The information core of PLM helps to lessen this burden
by capturing information and storing it for use by all stages of a products life.
This simultaneously leads to competitive advantages and environmental
friendliness.
4 Entry-Level Engineering Professionals and Product Lifecycle

Management: A Competency Model for Education
Recent decades have witnessed the burgeoning popularity of lean thinking and
its many theoretical offshoots as companies continue to look for ways to
strengthen their competitive edge. One of these, Product Lifecycle Management
(PLM), focuses on the goal of eliminating all forms of inefficiency from every
phase of the manufacturing cycle. By coordinating information technology with
organizational practices and processes, both within and across functional areas,
PLM is able to substitute information for wasted time, energy, and material
(Grieves 2006).
PLM has piqued the interest of corporate decision-makers across the
globe. As a consequence, companies that are implementing its guidelines
can be found on nearly every continent. Furthermore, as manufacturing
giants increasingly turn to PLM, so too do their supply chain vendors. One
of the beauties of this model is its provision for the following of manufac-
tured goods past the point of purchase, through disposal and disassembly, to
the eventual reuse and recycling of a products component elements. Thus,
although many organizations may have initially been drawn to PLM out of
the desire for improved profits, others also see it as perhaps the only solution
capable of helping them deal with the complexities of governmental regula-
tory changes.
4.1 The Competency Model Displaces the Task-Oriented

Approach
Faced with the changing regulatory climate and the rapidly evolving global
marketplace, organizations are also grappling with the need to bring their
workforces into the future. Todays employees are finding it necessary to
remain current with emerging technologies while simultaneously adapting
to profound shifts in the business aspects of manufacturing. Job descrip-
tions, once a corporate mainstay, are thus becoming outmoded as workers
find themselves called upon to tackle a host of new responsibilities. For
example, an editorial in the February, 2004, issue of Manufacturing
Engineering notes that manufacturing engineering (ME) professionals are
increasingly being asked to determine make or buy decisions; oversee
suppliers technologies, cost, delivery and quality; evaluate and improve
their suppliers core processes; integrate suppliers core processes with their
companies own processes; manage and deploy supply chain management
systems; and make improvements in their suppliers designs, processes, and
quality. The author therefore urges ME professionals to cultivate a wide
variety of new skill sets and to embrace a lifestyle of professional resilience
and personal development (Hutchins 2004).
4.2 Response from Trade-Interest Groups and Academia
Industrys call for versatile, cross-functional employees is being heard by a

growing number of professional associations. One of these is the Accred-
itation Board for Engineering and Technology (ABET) , which responded
by implementing, in 2001, a revised list of accreditation criteria for
engineering programs in colleges and universities. In addition to the time-

honored mandates for technical competence are new ones stating that the
graduate engineer must have the ability to engage in life-long learning, a
knowledge of contemporary issues, and the broad education necessary to
understand the impact of engineering solutions in a global and societal
context(Bell 2000).
The Society of Manufacturing Engineers (SME) is another industry trade
group that likewise recognizes the need for change. Since 1997 it has been
extensively researching what it refers to as competency gaps, specific capabil-
ities that companies insist are too often lacking among their workers. Based on
its findings the SME has developed its Manufacturing Education Plan (MEP) ,
an aggressive endeavor aimed at addressing the competency shortfall. Central
to the MEP is its list of competency gaps, which has undergone several revisions
since its inception in 1997. The most recent one has designated the following
fifteen items to be the most critical deficits between what education programs
provide for their students and what manufacturers need in terms of their
workforce:
Business knowledge/skills
Supply chain management
Project management
International perspective
Materials
Manufacturing process control
Product/process design
Quality
Specific manufacturing processes
Manufacturing systems
Problem solving
Teamwork/working effectively with others
Personal attributes
Written and oral communication
Engineering fundamentals. (SME and Competency Gap Research 2004)
The apparent shifting of paradigms from task orientation to one of com-

petencies in the workplace has not escaped the notice of academia. Faced with
the revised accreditation mandates from ABET, the groundswell of impor-
tance surrounding PLM, and impressive research data from interested groups
such as the SME, academe has begun test runs of innovative new curricula.
One example is a cross-disciplinary manufacturing engineering course
described by Chang and Miller of Purdue University (2005). The authors
state that the course was redesigned to meet industrys need for a PLM-literate
workforce, and that overall, it was very successful (p. 176). They acknowl-
edge, however, that at the time of publication their course had only been
offered once, and that further research will verify the degree of its success
among future students.
4.3 The Development of a PLM Competency Model for

Entry-Level Engineers
The purpose of this section of the chapter is to present a PLM competency model
that contains the elements necessary for the success of entry-level engineering
professionals who must function within a PLM environment. This project is
being ably assisted by seven highly credentialed PLM Subject Matter Experts,
all from Fortune 500 companies (i.e. Boeing, Rolls Royce, IBM, Caterpillar,
EDA, and Flexware), and all of whom are members of the Advisory Board of the
Product Lifecycle Management Center of Excellence at Purdue University. They
will define the skill objectives that are appropriate, and from these, a minimum of
200 derivative tasks will be specified. Then, from each task, thirty to forty key
skill objects will be isolated. The final list of skill objects will be used to construct a
Product Lifecycle Management competency model that also conforms to stan-
dards previously designated by the Society of Manufacturing Engineers.
The tool selected for this process is a software package titled SkillsNet,
developed in 1996. The program uses language descriptors that are common
to the Department of Labors Occupational Informational Network (O*Net)
in order to create a valid framework with which to define task, tools, and
knowledge requirements. These frameworks in turn are referred to by the soft-
ware as SkillObjects. The Trainers Guide defines SkillObject as a re-usable
detailed description of what people do in accomplishing work (SkillsObject
Trainers Guide, 19992004). A SkillObject contains the following elements: a
label or name, tasks, skills/abilities, tools/software/equipment/devices, unique
knowledge, resources, and performance standards.
In an industrial setting, the individual chosen to generate the criteria for each
SkillObject should ideally be a job incumbent who is recognized as a top-tier
performer by his or her peer group. Upon receipt of a list of SkillObjects, it then
becomes the task of the Work Element Editor to cull through each, and reduce,
according to the following guidelines:
Eliminate task, tool, and unique knowledge redundancies
Clarify tasks, tools, and unique knowledge so that they are understandable
to others in the occupation
Fix spelling and grammatical errors
Reduce the size of a task, tool, and unique knowledge list without losing
important information. (SkillsObject Trainers Guide, 19992004)
Following this initial editing process, each list will then undergo several more
revisions before its final submission to the managers or supervisors responsible
for implementation. For this study, the Subject Matter Experts will perform the
function of the top-tier job incumbent described in the previous paragraph. The
authors will then serve as the Work Element Editors. This project is still very
much a work in progress. At the time of this writing, surveys have been
distributed and the authors await the return of data.
5 Inter-organizational Relationships in Supply Chain Management:

A PLM-like Application in Higher Education Procurement
Changing business practices have altered traditional roles of purchasing profes-
sionals in many organizations. This is especially true in higher education where
purchasing professionals are becoming increasingly involved in most stages of
product development, in part because they can accurately predict the cost,
availability and suitability of items in question. By consulting with the purchas-
ing department in the early stages of product design, higher education organi-
zations are able to avoid, or greatly lessen, potential problems that may arise in
regards to pricing, availability, delivery, quality, and support. The need to
develop and implement effective supply chain management strategies is quickly
becoming a major factor in higher education resource allocation and a major
focus of their purchasing departments. The application of product lifecycle
management (PLM) could prove to be very beneficial in institutions of higher
education.
5.1 PLM in Higher Education

New practices have pushed higher education institutions into increasingly
complex, integrated supply contracts, making supplier selection critical due to
the increased scope and duration of these contracts. Integrated supply chains
incorporates all members of the supply chain including the supplier, transpor-
tation companies, and the university. A major responsibility of most purchasers
is to resolve problems that may arise with a supplier, since the success of the
relationship ultimately affects the buying firms productivity and performance.
Understanding the evolving relationship between supplier and buyer is a crucial
factor to success as it affects efficiency with regard to ordering and processing
orders that can be achieved in real time.
When an organization plans and executes in real time, based on a granular
view of items or case locations, speed and accuracy of information exchanged is
critical among the parties involved. By collaborating with key trading partners
in flexible planning cycles that address consumer requirements quickly, orga-
nizations become consumer-driven. Being consumer-driven places the needs of
the organization at the forefront of the supply chain with a focus on effective-
ness and efficiencies.
In contrast, the traditional supply chain is slow and unresponsive. Instead of

basing decisions on a finely tuned view of consumer demand, it uses data that is
less consumer driven, such as warehouse withdrawals or distribution center
receipts. This type of batch data provides visibility into brand or a case-level
volume that describes what happened yesterday or last week as opposed to
what is happening right now. Whats more, traditional supply chains operate in
a series of silos with little integration, collaboration and visibility among the
players, especially in institutions of higher education. The application of pro-
duct lifecycle management (PLM) could positively affect the supply chain
process at universities as they move toward and execute purchases based on
real time, driven by the immediate needs of the customer.
According to Grieves, systems that enable approaches such as PLM are
developed and find a place in organizations for a fundamental reason: with
these systems, we can substitute the use of information for the inefficient use of
time, energy, and material (Grieves 2006). The focus on increasing efficiencies
and the application of technology affects the development of the supply chains
at many universities.
PLM is of interest in the buyer-supplier relationship because it allows for a
holistic approach to supply chain management that utilizes technology to
aggregate buying. It leads to better understanding the big picture in regards
to commodities. Below is a summary of Grieves main points of PLM:
PLM is about product data, information, and knowledge
PLM concerns itself with the entire life of the product, from inception to end-
of-life
PLM is an approach that is more than software and approaches
PLM crosses functional boundaries
PLM combines the elements of people in action, processes, and technology
PLM drives the next generation of lean thinking.
Due to the fact that PLM brings better products to the market faster, it is
important to have established relationships with suppliers to capitalize on these
opportunities. The fact that higher education organizations are governed by
numerous state and federal statutes makes transparency in business practices
essential. PLM awareness allows for the buyer/supplier relationship to develop
to the point that integration of solutions, rather than products, becomes the
norm. According to Stark, many companies now offer complete solutions
rather than products. This adds a new layer of challenges as solutions are
more complex to develop and support than single products (Stark 2005)
5.2 The Case for PLM Higher Education

Purchasing Departments in higher education organizations are moving away
from traditional single exchange transactions toward an emphasis on creating
and maintaining long-term buyer-supplier relationships. As strategic partners,

these relationships generate a supply chain that must be managed. They encom-
pass all the activities associated with the flow of goods and services, as well as
the flow of information. Information throughout the entire organization must
simultaneously flow both up and down the supply chain to leverage strategic
positioning and to improve efficiencies. For organizations to obtain sustain-
ability in the future they must develop a strategy for creating, maintaining, and
sustaining buyer/supplier relationships based on institutional requirements. As
stated by a Vice-President of the Boston Consulting Group, As the economy
changes, as competition becomes more global, it is no longer company versus
company but supply chain versus supply chain (Henkoff 1994).
Also of importance is the maintenance and development of the relationships
between the purchasing organization and the supplier of the goods or services
required. These relationships appear to be formulated from several different
attributes of purchasing in an organization. Duffy believes that there are seven
areas in which change is occurring that will affect the buyer-supplier relation-
ships: electronic commerce, strategic cost management, strategic sourcing,
supply chain partner selection, tactical purchasing, performance measure-
ments, strategic supplier alliances, and complexity management (Duffy 2003).
At Purdue University an Enterprise Resource Planning (ERP) initiative is
under way. The e-commerce mechanism installed by the software product
SAP allows for aggregation of buying and management of the relationship
attributes with an emphasis on partnering. This partnering effect is designed
to utilize organization wide procurement quantity pricing by including all
local divisions and regional campuses connected electronically for ease of
tracking and ordering. This makes information sharing an integral part of
the process by governing and monitoring all aspects of the procurement
process including relationship management with suppliers.
5.3 Model of Strategic Relationship
While many intuit that face to face communication will become less important
as technology enhances communication, one of the major finds of this study is
that strategic relationships require attention. Understanding the complexities of
the relationship as viewed by both the buyer and supplier is important.
In the establishment of a relationship between the buyer and supplier
there emerged several attributes that make up the core of this model. These
attributes are contextually different from the maintenance or future devel-
opment attributes that surround the establishment core attributes. They are
different in that they are specifically focused on each relationship and show
little, if any variability, as the foundation from which the relationship is
built. Whereas, the maintenance attributes surround the foundational core,
Fig. 4 Supply chain strategic relationships
establishment attributes will vary according to the industry, suppliers size,

and capabilities.
The size and capabilities of a supplier are not as important when it comes
to the establishment of the buyer/supplier relationship. Honesty, commu-
nication, integrity, professionalism, delivery, follow-up, competitive pri-
cing, and willingness to go the extra effort, are regarded by the participants
of this study as key attributes in the formation or establishment of a buyer/
supplier relationship.
Honesty is regarded by one buyer To be the ability for both parties to put all
of their cards on the table without fear of price inflation or other unethical
applications. Honesty is thus viewed as the foundation upon which commu-
nications are designed and established.
Communications rely on the honesty of both the buyer and supplier in the
development of the initial relationship between the two. This communication
must attend to both the buyers and suppliers needs with particular emphasis
on empathetic listening skills.
Integrity combined with honesty in communications is a fundamental ele-
ment by which both the buyer and the suppliers representatives conduct
themselves. The honesty at which communications are conducted is indicative
of the individuals integrality. The integrity, once developed, further personifies
the relations essence in regards to perception. This perception of honesty,

integrity, and a degree of professionalism in the communications are regarded
by participants of this research as fundamentally vital in establishing relation-
ships between the buyer and supplier.
By having this open channel of communication between the buyer and
supplier, the other core attributes of delivery, follow-up, competitive pricing,
and willingness to go to the extra effort are more effectively established through
mutual trust and information sharing.
Delivery in the core development stage of the model is primarily focused on
time and condition of product. It is essential that the supplier understand the
need of the buyer in regards to delivery.
Follow-up is the key attribute that ensures feedback is obtained and the
relationship expectations are being followed-up on. If there is a delivery pro-
blem then there needs to be an open and honest dialogue to locate the problem
and come up with a mutually beneficial solution.
Competitive pricing is the foundation from which the actual product builds.
For institutions of higher learning it is exceptionally important to ensure that
the buyers are being good stewards of the tax payers money. One important
way this is accomplished is through ensuring that the supplier with which the
relationship is being developed has competitive pricing.
Professionalism is another key attribute of the inner core of the model. The
sales representative and buyer both need to stay focused on the needs of the
institution and the overt objective of the relationship. By presenting ones self in
a professional manner, one builds the relationship on these characteristics that
have a profound impact on both the buyer and supplier.
The willingness of the supplier to go to the extra effort is a summation of the
key relationship attributes for developing the relationship between buyer and
supplier. Without this effort, the rest of the attributes will not get the desired
results for either the buyer or supplier, primarily due to a lack of effort.
5.4 Performance Metrics
In institutions of higher education, there is considerable focus on aggregating

products to leverage purchases and obtain the best price. This best price is
complex in that it must now take into consideration the quantification of a long-
term relationship with a supplier. This is difficult at best, primarily due to the
challenge of measuring the solution and innovative nature of the relationship.
How the value of such a relationship is measured determines the strategic
direction supply chain management takes in higher educational institutions.
The call for developing performance metrics has been sounded.
One way of measuring effectiveness of the supply chain relationships is
through standardization of processes. This, in turn, allows for the quantifica-
tion of time, materials and energy levels expended in addition to the cost of a
product or service. The formation and monitoring of the relationship elements

provide a mechanism for working with the suppliers to achieve maximum
effectiveness in aggregation of the purchasing function.
PLM supports the need for buyer/supplier relationships to meet the growing
pressures and demands felt by institutions of higher learning. Brown and
Duguid speak on the pressure colleges and universities feel to change to be
become more effective and efficient, the whole system of higher education is
under pressure to change. Moreover, despite appearances to the contrary in
some ivied exteriors, colleges and universities are changing significantly. Chan-
ging the direction of organizational vessels as large as universities, however,
provides a fair amount of thrashing and churning. Consequently, it remains
hard to see the ultimate directions that educational organizations might take.
Since power is historically centralized in an educational institution and tech-
nology seems to dissipate distance through more effective communication, a
path must be formed to provide strategic direction. This path must steer
between a universitys centralizing tendencies and be optimistic that technology
will overcome the distance aspect and create a devolved system. This system will
then rely on relationships to develop future strategic directions.
6 Summary
Product Lifecycle Management, though only in its early years, has proven to
have tremendous impact. Whether trying to be first to market, improves qual-
ity, or increase customer base, PLM drives out wastes and permits the realloca-
tion of captured resources in support of product and process innovation,
resulting in potentially new revenue streams. Where once organizations con-
sidered the potential return of PLM before committing an investment, organi-
zations increasingly consider the implementation of PLM as a matter of
survival. Nonetheless, as with any change, PLM is not without its challenges,
both technical and social.
References
Adler NJ (1983) Cross-cultural management: issues to be faced. Int Studies Manag Organ
13(12):745
Azzone G, Noci G (1998) Identifying effective PMSs for the deployments of green manu-
facturing strategies. Int J Oper Prod Manage 18(4):308335
Bayer (2006) Bayer Annual Report. www.annualreport2006.bayer.com/en/ Homepage.aspx.
Accessed 11April 2007
Bayer (2005) Bayer Sustainable Development Report. www.sustainability.bayer.com/en.
Accessed 14 April 2007
Bell T (2000) Proven skills: the new yardstick for schools. IEEE Spect 63-67 ieeexplore.ieee.
org/iel5/6/18754/00866286.pdf. Accessed 9 March 2007
Burgos J, Cespedes J (2001) Environmental performance as an operations objective. Int

J Oper Prod Manage 21(12):15531572.
Cartwright R (2003) Mastering Team Leadership. Palgrave Macmillan, Wales
Center for Green Manufacturing (2006) University of Alabama. bama.ua.edu/cgm.
Chang YI, Miller CL (2005) PLM curriculum development: using an industry-sponsored
project to teach manufacturing simulation in a multidisciplinary environment. J Manuf
Sys 24(3):171177
Christmann P (2004) Multinational Companies and the natural environment: determinants of
global environmental policy standardization. Acad Manage J 47(5):747760
Cimalore C (2007) ERP/PLM complement each other & facilitate a collaborative environment
in design, engineering and manufacturing processes. Electronics Supply Manufacturing
Duarte DL, Tennant-Snyder N (2001) Mastering Virtual Teams. Jossey-Bass, San Francisco
Duffy R (2003) The future of purchasing and supply: where we are headed. Purchasing Today.
9094
DuPont (2007) www.dupont.com. Accessed 14 April 2007
Fai KP (2006) Determinants of environmentally responsible operations: a review. Int J Qual
Reliab Manag 23(3):279297
Florida Power and Light (2007) www.fpl.com. Accessed 14 April 2007
Gannon MJ (2001) Working across cultures. applications and exercises. Thousand Oaks Sage
Publications, CA
General Electric (2007) GE Ecomagination at work. http://ge.ecomagination.com/site.
Global Round Table on Climate Change (2006) www.earthinstitute.columbia.edu/grocc.
Grieves M (2006) Product lifecycle management: driving the next generation of lean thinking.
McGraw Hill Companies Inc, New York
Gudykunst WB (1997) Cultural variability in communication. Commun Res 24 (4):327348
Hassan C, Olawoye J, Nnadozie K (2002) Impact of international trade and multinational
corporations on the environment and sustainable livelihoods of rural women in Akwa-
Ibom state, Niger Delta Region Nigeria.
Hastings C, Bixby P, Chaundhry-Lawton R (1994) Superteams. HarperCollins, Manchester
UK
Henkoff R (1994). Delivering the goods. Fortune 130:6478
Hofstede G (1980) Cultures consequences. Sage Publications, Beverly Hills CA
House R, Hanges P, Javidan P et al. (2004) Culture, leadership, and organizations: the
GLOBE study of 62 countries. Thousand Oaks, Sage Publications, Beverly Hills CA
House R, Hanges P, Ruiz-Quintanilla et al. (2007) Cultural influences on leadership and orga-
nizations: Project GLOBE http://www.thunderbird.edu/wwwfiles/ms/globe/Links/process.
pdf#search=Project%20GLOBE%20cultural%20dimensions Accessed 26 March 2007
Hutchins G, (2004) Manufacturing engineers must reduce competency gaps. Manuf Eng 132(2)
http://www.sme.org/cgi-bin/find-articles.pl?&04fem010&ME&20040201&&SME& Accessed
29 March 2007
Jarvenpaa SL, Leidner DL (1999) Communication and trust in global virtual teams. Organ
Sci 10(6):791
Javidan M, Dorfman PW,de Luque MS, et al. (2006) In the eye of the beholder: cross cultural
lessons in leadership from Project GLOBE. Acad Manag Perspect 20(1)
Kharin V, Zwiers F, Zhang X, et al. (2007) Changes in temperature and precipitation extremes
in the IPCC ensemble of global coupled model simulations. J Clim 20(8):14191426
Kimura F, Kato S (2003) Life cycle management for improving product service quality. CIRP
Life Cycle Engineering Seminar. Erlangen, Germany
Lipnack J, Stamps J (2000) Virtual teams: working across space, time, and organizations.
John Wiley & Sons, New York
Minami T, Kanemaki Y, Nemoto H et al (2004) Eco & PLM project for environmental
traceability of EEE. Hitachi Rev 53(5):243249
Nauman S, Iqbal S (2005) Challenges of virtual project management in developing countries.
Eng Manag Conf IEEE Int 2:579583
Noci G (1995) Accounting and non accounting measures of quality based performance in
small firms. Int J Oper Prod Manage 17(7):78105
RoHS (2007) What is RoHS? www.rohs.gov.uk. Accessed 21 April 2007
SkillsObject Trainers Guide (19992004) SkillsNet Corp. Waxahachie TX
SME and Competency Gap Research (2004) Manuf Eng 132(2) http://www.solartoday.org/
2005/jan_feb05/chairs_cornerJF05. Accessed 29 March 2007
Stark J (2005) Product lifecycle management. McGraw-Hill. NY
Starrs M (2005) The futures so bright. Solar Today. http://www.solartoday.org/2005/jan_
feb05/chairs_cornerJF05. Accessed 10 April 2007
Tuckman B, (1965) Developmental sequence in small groups. Psychological Bulletin 63(6)
384399
The Dow Chemical Company (2006) 2015 Sustainability Goals. http://www.dow.com/com-
mitments/goals/index.htm Accessed 9 April 2007
The Dow Chemical Company (2005) The Dow Chemical Company 2005 Global Reporting
Initiative Report
Toyota (2007) Toyota Environment. http://www.toyota.com/about/environment/index.html.
Udomleartprasert P (2004) Roadmap to green supply chain electronics: design for manufac-
turing implementation and management, IEEE International Conference on Asian Green
Electronics
Westkamper E, Alting L, Arndt G (2001) Life cycle management and assessment: approaches
and visions towards sustainable manufacturing. J Eng Manuf 215(5):599624
Product Design Optimization: An
Interdisciplinary Approach
Sergio Romero Hernandez and Omar Romero-Hernandez
Abstract As new markets arise and global competition takes place, there is a need
to come up with innovative-successful projects that present the best holistic alter-
native based on customer needs and market expectations. The present research
provides a high-end technical solution to determine the best product alternative in
terms of market expectations, product & process specifications, logistic and envir-
onmental performance. The integration of this framework is oriented towards
rapid generation and evaluation of innovative products. The methodology is an
engineering design framework based on the efficient use of design tools and
computer aided design and engineering (CAD/CAE). This methodology is illu-
strated with 2 case studies: the design and development of a commercial fan and
the envisioning of an energy efficient luminary for public lighting. The implemen-
tation of this methodology delivered a better ranked product in less time and lower
cost. The correct application of these engineering concepts can be easily extended
into almost any new product at its development stage.
Keywords Product design CAD CAE Life Cycle Assessment
1 Introduction
Higher degrees of competitiveness demand worldwide companies to reconsider
its perspective about traditional design schemes. It is necessary to come up with
innovative-successful projects that present the best holistic alternative based on
customer needs and expectations. Depending on the corporate strategy of the
company the breach between what the company produces and what the market
and customers are expecting can be overcome to result in innovative product
alternatives [1]. Furthermore, cost control and product quality only sustain a
level of competitive advantage, while product or service innovation is the actual
source of competitive advantage in the global market [2].
S.R. Hernandez (*)

Department of Industrial Engineering and Operations, Center of Technology
Development, ITAM: Instituto Tecnologico Autonomo de Mexico, 01010 Mexico
City, Mexico

170 S.R. Hernandez, O. Romero-Hernandez
Several aspects should be analyzed to reach a successful design. First of all, the
objectives of the product in development should concur with the strategical
objectives of the company. Second, the time expended in the design stage should
be benchmarked against the market standards, and the cost of development should
be as low as possible. In this sense, it should be carefully analyzed whether the
customer and market needs are fully considered in the product. The manufacturing
and logistics needs of the product should also be analyzed [3]. Considerations
should not be limited to the previously mentioned but can be expanded to include
the marketing performance and the environmental impact along the life cycle of
the product.
A hierarchy of attributes is imperative in order to consider the relative weight
of each aspect during the design process. This hierarchy is typically obtained from
interviews with several members of the marketing, sales, manufacturing, and
research areas of the company. As such, developments in the design of a product
must be defined and communicated in an explicit and hierarchical manner [4].
Nowadays, the behavior of global markets for goods demands a rapid and
efficient adaptation of the products so they can meet the requirements of the
supply chain, the customers and local legislation in record time. Hence, the need
to apply an efficient stage of design and development of products that takes into
account these aspects is emphasized. A design that does not take into account
these considerations probably will present problems in further stages [5].
For any type of product, the changes required to improve its performance
will have a higher cost the later in time they are implemented. This increase in
cost will be exponential along the life cycle of the product; hence a modification
at the production stage will be hundred times higher than one at the design
stage as shown in Fig. 1. Moreover, design processes usually include several
Modification cost at different development stages

$100.00
$100
Normalized cost for modifications
$80.00
to the original design
$60.00
$40.00
$20.00
$1 $10
$0.00
Design Testing Production
Phase development stage
Fig. 1 Cost of modifications along a product life cycle

Product Design Optimization 171
GENERATION OF DESIGN FOR

CORPORATE CUSTOMER/
PRODUCT CONCEPT CONCEPT PRODUCT MANUFACTURING DETAILED
STRATEGY COMPANY
SPECIFICATIONS GENERATION SELECTION ARCHITECTURE AND ASSEMBLY, DESIGN
NEEDS
(QFD) PRODUCTION, ETC
FUNCTION/
FUNCTIONAL
VOC- MKTNG CONCEPT
DECOMPOSITION
TESTING
Fig. 2 Traditional PDD process
trial-and-error attempts oriented to tackle fine-tuning issues that take place

during the manufacturing process design, where unforeseen dimensional errors
occur. These trial-and-error attempts can be fastened with the use of simulation
models and variability analysis [6].
In a traditional product design and development process (PDD), satisfaction
of customer needs is achieved by the development of engineering specifications
that aid the generation of several product concepts and the further selection and
refinement of one of them (Fig. 2). This traditional generation and selection
process is usually accelerated with the use of template-based environments were
design calculations, component search, machine selection and preliminary cost
are integrated into a relational database [7]. Usually the process concentrates in
developing the best solution based on the operational performance of the
product and its manufacture and/or assembly. Nevertheless, there are impor-
tant aspects that are often overlooked, such as the environmental impact that
the product will have along its life cycle or the logistic performance of the
production, distribution and final disposition of the product [8].
A significant problem arises in the PDD process: market dynamics demand for
fast and lean solutions. As such, the framework proposed in this work considers a
series of product innovations which account for specific customer needs. The
main objective of the proposed framework is to conduct rapid evaluations in
order to determine those alternatives that satisfy both customer and manufac-
turer requirements. Moreover, this framework helps determine the most ade-
quate environmental and logistic alternative for the design of a new product.
The interdisciplinary approach presented in this article receives inputs from
various areas such as marketing, production, logistics and R&D. An efficient
manner to deal with all this information lies in the utilization of efficient models
that conceptualizes both holistic and component views in compact packages. At
least two of these detailed models can be found in the literature [9].
2 Methodology
The Romero-Romero methodology (R-R) is based in a recursive process that
on each iteration develops a new model of the product. In order to apply it, the
corporate strategy has to be properly defined, as well as the mission statement
for the new product. Customer needs should be properly identified, and engi-
neering requirements of the product should be developed to satisfy them. The
previous tasks can be performed by aims of marketing studies, analysis of the
supply chain of the product, analysis of the processes affecting the product,
and finally a legal analysis of the normativity and intellectual property regard-
ing the new product. Once this information has been gathered, it is necessary to
conduct a priorization of the product features and requirements. Most compa-
nies have used diverse methods to successfully carry out this priorization, most
of them based in multiple-criteria decision tools [10]. The Quality Function
Deployment methodology [8] can be used to conduct this correlation. The
application of this methodology will result in a House of Quality that will
work as a roadmap for the design process. Then, a first proposal of the product
design can be developed by conventional means, such as the traditional PDD
process presented in a previous section.
The methodology shown in this article is an engineering design framework
based on design tools and computer aided design and engineering (CAD/
CAE) (Fig. 3). This framework integrates not only manufacturability and cost
issues but also two other fundamental aspects: environmental and logistic
performance of the product along its life cycle. Based on these two perfor-
mance aspects and with the use of analytical tools it is possible to envision
and develop advance product designs that comply with market and usability
objectives, while achieving lower environmental impacts and lower transpor-
tation costs.
For each iteration there will be a comparison between the last evolution of the
design and the one before. From those comparisons trouble areas and weaknesses
MARKETING
ASSESSMENT
COLLECT AND TRANSLATE

DISCRIMINATE PROPOSALS TO
CHANGE DESIGN
PROPOSALS CHANGES
PRODUCTION
ASSESSMENT
DISTRIBUTION
TRADITIONAL OF DESIGN TO
PDD SEPARATE
PROCESS EVALUATION IMPLEMENT
TEAMS CHANGES INTO
THE DESIGN
ENVIRONMENTAL
ASSESSMENT
SAME SUPPLY CHAIN
COST/ITERATION
DESIGN
ESPECIFICATIONS LOGISTICS LOWER THAN
NO HOLISTICALY
RE-ASSESMENT ASSESSMENT EXPECTED
OPTIMISED
IMROVEMENT?
YES
Fig. 3 Framework for product optimization

Fig. 4 Environmental assessment
in the product are identified. At the end of each iteration the environmental and
logistics team propose a wish list of improvements, which are usually presented
in a ranking of design modifications. At the end of each iteration there is a re-
assessment of the design requirements.
As part of this framework, Life Cycle Assessments (LCA) have to be per-
formed in order to identify: (i) the most adequate raw material, (ii) opportunity
areas for low energy consumption and (iii) the most convenient number and
shape of components in order to facilitate recyclability, remanufacturability
and reusability (Fig. 4). The LCA also considered transportation and logistic
issues such as (i) the effect of design size, weight and shape on storage capability
or (ii) the effect of storage capability on cost and environmental impact.
Furthermore, social aspects were also considered and integrated as design
constraints since the conception stage. As such, designs were subject to a
minimum use of hazardous materials as established in a risk assessment evalua-
tion (also performed by the authors), based on Mackay III fugacity models and
official US toxicity values. [11, 12, 13, 14]
The logistic assessment of product performance is based on an analysis of the
expected supply chain of the product (Fig. 5). In a forward logistics assessment,
the metrics for evaluation are the portability features of the product and its
packaging (weight, volume, and stacking capability), as well as the reliability,
cost and quality of the different suppliers of raw materials and parts. Moreover,
environmental and logistic issues can be evaluated along the supply chain in
order to identify process alternatives that lead to lower environmental impacts
and consequently, to greener products [15, 16].
Suggestions from different assessment groups are collected by a multidisci-
plinary team that discriminates them and translates a selected few into design
changes that will be passed on to the design team for implementation into the
product. In order to keep control of the changes the authors recommend limit-
ing the maximum number of changes per iteration to five.
Fig. 5 Logistics assessment
3 Case Studies
3.1 Case 1: Development of a Table Fan
The following figures and tables correspond to a case study developed by the
authors in order to design, develop and optimize a table fan. This product was
requested in the Mexican market, so the new design took into account (i) design
for manufacturability, (ii) design for the environment, (iii) design for logistics,
and (iv) design for sustainability under a PLM scheme [10]. The table fan was
designed from a series of inputs: market study, analysis of competitors, inter-
views with experts of various departments such as manufacturing, logistics,
CAD and environmental engineering.
At the top of the methodology lies the corporate strategy and the products
mission. Afterwards, a detailed examination of all characteristics and con-
straints of the new product was carried. This examination was based on market
analysis, an understanding of the supply chain and knowledge of legislation,
environmental and health restrictions. All attributes are ranked in order to
communicate its relative importance (Table 1).
This information is also associated to engineering requirements (perfor-
mance metrics) and translated into a priority matrix a house of quality, as
illustrated in Fig. 6 [8]. The house of quality correlates customer needs with
engineering requirements in a hierarchical manner. Several designs can be
generated in a first stage, and then an iterative process takes place to select
and develop one of them until a first generation of the product is obtained [16].
A series of technical constraints and specifications are considered mandatory
for the design:
1. Blade diameters must be 40 cm
2. Electricity input must be 110 V
Table 1 Priorization of design needs for the table fan

Attribute Percent Ranking
Low noise 11.50%
Good Packaging 8.00%
Lightweight 8.00%
Multi-speed 7.50%
Oscillation 7.50%
Assembly time 7.50%
No vibration 7.00%
Stackable 7.00%
Recyclable 7.00%
Built In-house 6.50%
Low manufacturing energy 6.50%
Tilting 6.00%
Reduced material 6.00%
Re-usable 4.00%
Fig. 6 House of quality for a table fan

3. Fans size must be under 60 cm from the base to the end of the blades house.
4. Fans weight must be under 7.5 kg
5. Fans base must not exceed 50 cm in diameter
6. Engine must supply at least 60 W and must be provided by a qualified
supplier.
The first two constraints are based on the Mexican standard NMX-J-016,
while the rest of the specifications were provided by the marketing and the
design departments. It should be kept in mind that not meeting these specifica-
tions will lead to a less attractive product.
The first design was developed by aims of computer aided engineering
software used in the company for the design of all its family of products:
Unigraphics NX3. The first design is presented in Fig. 7.
It is important to note that the first iteration came up after undertaking a
traditional methodology for product design. First of all, clients needs were
identified and expressed in terms of engineering specifications, which in turn led
to concept generation. These inputs represented the base to the following
activities: concept selection, product architecture and detailed design imple-
mented in CAD [5].
The R-R methodology was then applied in order to improve the initial design.
This iterative methodology compromises the new design in terms of aspects such
as manufacturability, environmental performance, and logistic implications.
A wish list of improvement ideas based on these three aspects was generated
and used as a list for trial and evaluation attempts. Furthermore, design para-
meters were reviewed in order to narrow the universe of new possible changes.
Conflict on the scope of improvement decisions during the iterative process were
solved based on the hierarchy of product attributes presented on Table 1.
Fig. 7 First generation of table fan

HDPE
LDPE
PP
Fabrication Final Disposition
Process Fan Use
Aluminium
Transportation Transportation Transportation
Cardboard
PS
CONSUMPTION OF MATERIAL RESOURCES
RAW MATERIALS ENERGY CONSUMPTION
Fig. 8 System for environmental assessment
The following sections illustrate the environmental and logistic assessments

to improve the performance of the product as evaluated for each iteration.
3.1.1 First Iteration-Environmental Performance

For the environmental evaluation of the first iteration, a life cycle assessment
(LCA) was performed. The methodology of the life cycle assessment is inte-
grated in an international standardized frame as specified in the family of
standards ISO 14040. [14]. LCA is a methodology commonly used in environ-
mental engineering for the environmental evaluation of products and processes
taking into account their whole life cycle: extraction of raw materials, manu-
facture, distribution use, recycling process, and final disposition of the product.
This methodology is composed by (i) definition of the system, objectives and
scope, (ii) an inventory of emissions obtained by energy and material balances,
(iii) a stage of environmental evaluation that takes into account a series of
aspects like global warming and resource consumption, among others, and (iv)
identification of improvements to the system [16]
Based on the aforementioned standards, in the practical application pre-
sented in this article the following system was defined (Fig. 8): First, raw
material is extracted and processed (High density polyethylene, HPDE; low
density polyethylene, LDPE; polypropylene, PP; Aluminum, Al; cardboard
and polystyrene) so it can be ready for production; afterwards, it is transported
to the plant, where it is transformed by means of diverse injection processes
which are needed to manufacture some pieces of the fan; after this, the assemble
and packaging is done; then, the fan is transported to the point of sale, from
there to the consuming place, where it has a specific operational life (approxi-
mately 5,000 hours); finally it is transported to its respective place for final
disposition (landfill or recycling). The developed equations for the material
and energy balances, as well as for the correlations that determine the inventory
of total emissions and the calculations for the relative environmental impact are
available upon request.
Mass and energy balances were performed with the use of a calculus spread-
sheet and GaBi 4.0, a commercial software for life cycle assessments. These
activities led to an inventory of emissions, with information on the type and
amount of materials and energy produced and consumed. Main results are
presented in four different categories:
1. Total: addition of the other three categories.
2. Fan production: includes emissions of the extraction of the raw material, the
transportation of the raw material to the plant, the manufacture of the fan
and the distribution to the sale points.
3. Fan use: includes exclusively emissions from the useful life of the fan by the
costumer.
4. Fan disposal: includes emissions from the final disposition of all parts that
constitute the fan (including packaging material)
This LCA had two fundamental objectives: first, to determine if an emission
of pollutants that was not allowed by regional law existed in the process;
second, to identify the process stage in which the highest environmental impact
existed; in order to direct modifications on products design, especially on those
activities associated to the highest environmental impacts.
The main indicators related to environmental impact were the following:
Potential contribution to greenhouse effect (Global Warming Potential,
GWP), Depletion of natural resources (Abiotic Depletion, AD) and Human
Toxicity (HT). Fig. 9 illustrates the relative GWP associated to major trans-
formation processes. As such, the X-axis represents a series of processes, such
as (i) energy consumption, (ii) production and transformation of materials
(LDPE, HDPE, PP and Aluminum) and the rest of processes involved in the
LCA. It can be seen that the stage that contributes the most to the GWP
environmental impact is the fan use by the consumer, which is basically due
to energy consumption (electricity). This contribution represents approxi-
mately 95% of the total emissions related to the life cycle of the product.
In other words, the energy consumption that is required during the useful life
of the fan is responsible for the largest environmental impact related to the
product.
Percent Contribution to GWP
% 95.5
100
10
2.11
1.28
1 0.459
0.32 0.331
0.1
Fig. 9 LCA results for first Energy LDPE HDPE PP Aluminium Other
generation of the design Consumption processes
As such, efforts on a new design were oriented to obtain a product of less

electrical consumption, and in particular to the implementation of a new engine
with lower requirements of consumption and power.
It is important to mention that other obvious changes like a design of less
material would also have a positive impact. Nevertheless the LCA showed that
this change represents a small percentage of the global emissions and does not
justify its incorporation into the first iteration of the process.
3.1.2 First Iteration- Logistic Performance

This section presents a supply chain analysis related to the fan from raw
materials procurement to end product delivery. Weight logistic variables related
to product design are presented in Table 2.
The ranking of the suppliers was obtained after performing a weighted
average of the following criteria: price of raw material, average lead time, and
quality of raw materials. A mathematical model was developed for this in order
to understand the effect of each variable on the general ranking and conse-
quently find the most convenient provider.
The mathematical model for the weighted average was devised as the addi-
tion of the weighted values for distance, cost and historic performance of
suppliers. The resulting equations lead to a comprehensive and clear set of
values that can be easily interpreted by all members in the design team. The
following formula represents suppliers ranking:
Weighted ranking A:fx B gy C hz (1)
where
A weight value assigned to distance
B weight value assigned to cost
C historic weight given on previous rankings.
Table 2 Logistic aspects First iteration

General aspects Units
Net total weight 5874 Grams
Number of parts 20 Pieces
Logistics
Package size
Length 500 Centimeter
Width 500 Centimeter
Height 350 Centimeter
Maximum Stacking 6 Boxes
Package weight 950 Grams
Total weight 6824 Grams
Suppliers rank 8.0
Weighted values assigned to the constants in this work are based on current
hierarchies as reported previously on Table 1. As such: A 0.25, B 0.45,
C 0.30
The functions f(x), g(y) y h(z) were proposed in order to obtain rankings on
the same order of magnitude (between 0 and 10) so they could be compared.
The function f(x) decreases at a decreasing rate. Three points define the
parabolic shape of this function:
1. 0 km equals 10 points (none of the suppliers reaches this value)
2. 1000 km equals 0 points (suppliers must be based within this distance)
3. 160 km equals 7.5 points (average distance considering all potential suppliers).
fx 6:6964 106 x2 0:0167x 10 (2)
where x distance.
The function g(y) is also a function that decreases at a decreasing rate. Three
points define this function:
1. A cost of zero equals 10 points (none of the suppliers reaches this mark)
2. There is an asymptotic limit on 0 points (i.e., the cost reaches an infinite value)
3. A cost of $ 13.21 equals 7.5 points (13.21 is the average price).
396:3
gy (3)
y 39:63
y cost of raw material as given by each supplier

The function h(z) is a function that increases at a decreasing rate. Three
points define this function:
1. 50 points equals a ranking of 9 (currently, this is the maximum rank)
2. There is an asymptotic limit on 10 points (none of the suppliers reaches this
mark)
3. 0 points equals a ranking of 0.
55
hz 10 (4)
z 5:5
where z total points based on suppliers history.

Results presented in Tables 1 and 2 suggest that logistic improving is directly
related to product packaging. Thus, a series of modifications were proposed in
product design in order to reduce package size and net weight.
Design improvements in terms of weight, resource depletion and energy con-
sumption were incorporated into the fans second generation. There was only one
constraint that could not be modified: the fans engine. The reason was that the
supplier had previously negotiated a long term supply contract. Moreover, engines
are provided by a qualified supplier. The new design is illustrated in Fig. 10.
Fig. 10 Second generation of table fan
3.1.3 Second, Third, and Fourth Iterations

The new generation performed better in terms of materials consumption
(resource depletion). Similar conclusions to those presented in the first genera-
tion were reached in the second generation. This is due to the fact that changes
adopted in other categories outweighed environmental performance (see
Table 1). Figure 11 presents the relative GWP impact of production and
transformations of materials (LDPE, HDPE, PP), Fuel consumption (diesel),
and Energy evaluated for the third generation of the fan design.
Logistic changes adopted for the second generation were performed. Major
improvements were concentrated in four areas: package weight, fans net
weight, packages height, and stacking number. The changes on the design
%
50 47,8
40
30
20
11,7 10,4 11
10,1 9,07
10
0
Total
LDPE B PP B Heat diesel B250 Energy South America HDPE B Remaining processes
Analyzing 1 p life cycle 'Ciclo de vida Ventilador 2'; Method: CML 2 baseline 2000 / World, 1990 / characterization
Fig. 11 LCI of the third generation of table fan, percentage contribution to GWP
Table 3 Overall improvements to table fan

1st 4th % of
Analyzed attribute Generation Generation Difference Improvement
Net Weight 5,915 g 4,319 g 1,596 g 27 %
Packaging weight 950 g 815 g 135 g 14 %
# stacking 6 boxes 11 boxes 5 boxes 83 %
Volume Packaging 87,500 cm3 63,000 cm3 24,500 cm3 28 %
In House Fabrication 75 % 75 % 0 0%
Pondered ranking of suppliers 7.70 points 7.89 points 0.19 points 2.5 %
Energy consumption of fan 70 W 66 W 4W 5.7%
Equivalent emissions of 266 kg 225 kg 41 kg 15%
CO2 (GWP)
Equivalent consumption of 1.67 kg 1.40 kg 0.27 kg 16%
Sb (AB)
allowed for a reduction of 8% in energy consumption due to a change in the

electric motor. Moreover, there was a reduction of 5% in the energy required to
produce the fan, due to the changes in mass to the components.
The iterative process presented in this section was conducted until a fourth
generation that compromised major attributes was reached. A final design for
this product involved significant advantages such as a new record on design
time, lower material consumption, lower weight and higher stacking number.
Table 3 presents a comparison of total improvements along design generations.
The final design is illustrated in Fig. 12.
From the results presented in Table 3, it can be appreciated that all logistic
and environmental aspects assessed suffered an improvement. The most sig-
nificant result from the logistics evaluation is the increase on the number of
Fig. 12 Final generation of table fan

items that can be stacked for transportation, it went from 6 to 11, which
represents an 83% increase. Design changes on the base and fan guards allowed
for an important reduction in the net weight of the fan, as well as a reduction in
material consumption to produce those parts. These modifications are reflected
on a lower expenditure on raw materials, a reduced amount of energy required
for its processing, and consequently a lower level of emission of GWP gases.
3.2 Case 2: Development of a Lamp for Street Lighting
The R-R methodology was also applied in the development of a highly-energy-

efficient lamp for street lighting. As was presented in the previous section, the
first step in the design process includes the definition of the mission statement of
the product. The lamp mission statement declared that the product was envi-
sioned to work either connected to the main electrical grid or (due to its high
energy efficiency) in an independent manner connected to a solar panel, con-
troller, and set of batteries.
There were several technical constraints regarding the development of the
lamp: (i) the assembly of the lamp should be easy, not relying on complex
technologies, (ii) the lamp should be able to be mounted on all commercial
street posts, (iii) the lamp should be able to operate under conditions of low
solar radiation (December and January in the northern hemisphere), (iv) the
lamp should be able to stand adverse environmental conditions (humidity,
salinity, wind, extreme temperatures, etc.) during its life cycle, and (v) the
lamp should meet the Mexican illumination standards NOM-001.
A typical supply chain for public lighting systems is shown in Fig. 13; each
member of the chain has different needs and requires different features of the
product to fulfill them. It should be noted that in accord to the holistic design
philosophy of the R-R methodology, all members are considered customers of
the product.
According to a series of interviews with different participants of the supply
chain the needs were identified as features that the product should have regard-
less of the technological solution. These needs were first ordered according to
Table 4.
The needs listed in Table 4 can be grouped in two large categories: those
related to the interaction with the lamp and those related to the performance of
the product during use. A new series of interviews and focus groups were
Producer /
Integrator of
Major Construction Installer Final User
public lighting
Distributor Company
systems
Fig. 13 Supply chain for public lighting systems

Table 4 Needs of different participants of the supply chain

Needs
Producer / The availability and continuity in the market of third party components
Integrator should be ensured
Parts should be efficiently stored
Major Distributor Easy to transport
Stock availability with the producer
Availability of credits and discount policy
Construction Financial justification of the investment
Company Ensure warranty and maintenance after sell
Meet international standards
Installer Comfortable and rapid installation
Avoid extra costs during installation
Protect the equipment against rough use and accidents
Final User Obtain enough luminescent performance compared to traditional
lighting technologies
Reduce maintenance and avoid replacements
Avoid blackouts
Avoid the stealing of the equipment
Reduce the electrical consumption and the costs associated to operation
and maintenance
conducted in order to further detail the needs, especially those relating to the
performance of the product. These needs were ordered according to their
importance, where level 3 needs are critical for the customer, level 2 needs are
of medium importance, and level 1 needs are desirable features (Table 5).
Engineering requirements were developed to satisfy the needs the customer is
expecting from the lamp. This was performed in a similar manner as for the
table fan presented previously. Once the requirements were established, a
benchmark was conducted by analyzing the existing products in the Mexican
Table 5 Performance needs for street lamp

Need No. Need description Hierarchy
1 The lamp is on all night 3
2 The lamp works even in cloudy days 2
3 The equipment is installed high from ground 3
4 Long life of components 3
5 Affordable prices 3
6 The illumination meets the Mexican standards NOM-001 2
7 Attractive and modern appearance 1
8 Low maintenance 3
9 Easy installation using standard tools 2
10 Detailed technical information of components 3
11 Clear manuals of operation and maintenance 3
market that were competitors to the street lamp. Since none of these solutions
were satisfactory for the design team it was decided to develop a new product
concept.
A product concept is an approximate description of the technology, opera-
tion, and shape of a product: this concept represents how the product will meet
the needs of the customer. In order to develop a concept, it is necessary to vision
the product as a series of connected functions (or sub products) that work
together. The process of defining these functions is known as functional decom-
position and helps identify the critical functions that the product has to perform
to translate them later on into features of the product. For the solar street lamp
the functional decomposition is depicted in Fig. 14.
From the functional decomposition three critical functions were identified,
namely: conversion of solar to electric energy, storage of electric energy and
conversion of electric to luminescent energy. From the definition of the
product presented above, the conversion of solar to electric energy was selected
to be by means of a set of photovoltaic (PV) panels commercially available. The
options of PV panels available in the market vary depending on the price,
conversion efficiency, light sensitivity and warranty. Several potential suppliers
were identified offering different technologies such as mono-crystalline, poly-
crystalline, or amorphous PV panels. The selection of the best suited PV panel
was not based only on the previous characteristics but also on its integration to
the whole system, since it was found that although the amorphous panels
provide the best energy performance, they require a larger area than the other
options hence complicating the installation. Moreover, it was also necessary to
Protect and fix the lighting system to a post

Street Lighting
Solar Energy
Storage Conversion of
Conversion of
of electric electric to
solar to electric
energy energy luminescent
energy
Management of the electric fluxes in the system
Critical flux
Fig. 14 Functional decomposition of the lamp

consider the reliability of the suppliers; consequently the selection was to use
crystalline PV panels.
A solar powered system for lighting presents an interesting problem in
terms of energy management; during daylight the PV panels will generate
electric energy that needs to be stored for its use during the night. The best
available technology for storing electric energy is the use of batteries; hence,
like in the PV panel selection, a reliable battery was selected. The selection
criteria for the batteries were the expected lifetime (directly linked to main-
tenance cycles) and the energy capacity. The development team decided to use
sealed batteries with a large energy capacity in order to reduce wiring and
connection complexity.
The third function, conversion of electric to luminescent energy, was con-
sidered critical for the development process of the system. It was found that
the competitors used either gas discharge lamps (specifically sodium oxide) or
fluorescent tubes. Although the gas discharge lamps offered good luminescence
to energy ratio, the light emission is omni directional and distinctively yellow in
color; in other words not all the light produced goes to the street surface, and it
is not possible to have a good color differentiation. The fluorescent tubes do
not have a good light penetration nor meet the Mexican standards for street
lighting. Since none of these alternatives were satisfactory for the design team it
was decided to develop a new solution based on Light Emitting Diodes
(LEDS). A LED is a source of highly directional light with an opening angle
of only 78, in other words a single LED at a height of 8 m will illuminate a circle
of 1 m diameter. Clearly this presented an issue since in street lighting systems
the area to be illuminated is a circle of 10 m radius (314 m2). The design team
developed a solution to orient an arrangement of LEDs in a semi-spherical
manner in order to cover the working area. This development was registered
with the World Intellectual Property Organization (WIPO) and is sketched in
Fig. 15.
Fig. 15 Concept sketch of

the orientation of LEDs (By
courtesy of Grupo ECOS
Innovations 1)
Fig. 16 Physical prototype

of LED based lamp (By
courtesy of Grupo ECOS
Innovations 1)
The LED arrangement was mounted in a commercial enclosure to test its

lighting efficiency, ease of manufacture and assembly, energy consumption, and
selection of materials. This resulted in a physical prototype as shown in Fig. 16.
The R-R methodology was applied to this product concept in order to
optimize its operational, environmental, and logistic performance. After
4 cycles of optimization there were improvements in the ranking of suppliers,
the energy to luminescence ratio, the amount of recyclable materials in the
lamp, the energy consumed during production, and the mounting arrangement
for the system, among others. Due to confidential purposes, specific technical
data cannot be presented in this paper.
The design was optimized to fit a cobra-head type of enclosure that presented
the best protection against adverse environmental conditions (rain, humidity,
dust, etc.) while improving the ease of assembly and installation. Figure 17
presents a close up of the final LEDs-enclosure arrangement.
Fig. 17 Final lamp close up

(By courtesy of ECOS
Innovations 1)
Fig. 18 Schematic of the

mounting system for the
street lamp (By courtesy
of Grupo ECOS
Innovations 1)
Once the three critical functions were solved it was important to integrate
them into a system that could be mounted into most commercially available
posts. This is a design requirement since the system is intended to be used in
several geographical locations within Mexico. A schematic of the mounting
system is presented in Fig. 18.
The system for street lighting was selected by the government of Mexico
City to be used in a linear park. More than 150 lamps were installed and have
been functioning with no malfunctionings or maintenance for over two years.
Figure 19 shows a view of the park.
Fig. 19 Solar powered street

lamps at linear park in
Mexico City (By courtesy of
ECOS Innovations 1)
4 Conclusion
Developing concepts in order to satisfy consumer needs does not represent a

sustainable competitive advantage. It is mandatory to innovate and generate
flexible concepts within record timing and minimum cost.
Innovative and integrating engineering designs and environmental and logis-
tic issues within a single framework represent a useful alternative to come up
with optimum product designs at a faster speed than with a traditional trial and
error scheme. Multivariable decision making can be accomplished with the use
of constrained parameterization. Hence, as long as manufacturing, environ-
mental and logistic interaction are expressed in terms of equations, the innova-
tive design process is facilitated.
This framework was successfully applied into various product-process sys-
tems. A comprehensive diagram of both frameworks was presented. The main
results showed that an optimized design can be achieved in the early stages of
the design process, by combining computer aided tools for design, analysis,
and predictions in a feed-back process loop. This methodology allows for the
conception of a successful prototype in a do it right the first time manner.
Furthermore, the mathematical framework showed trade-offs between higher
inputs (electricity, steams, and steel) and lower discharge limits (considering not
only the production process but also the need for a waste-treatment plant).
The conceptual design of a commercial fan serves as a means to illustrate the
methodology and the advantages of a holistic design. After reviewing results
from the first to the fourth iterations it is clear that design parameters vary
within a range. As such, trade-offs within parameters occur, which in turn are
evaluated with the QFD matrix.
The methodology was also applied in the design and development of a solar
powered system for street lighting in Mexico City. Four cycles of optimization
were performed resulting in improvements in the ranking of suppliers, the
energy to luminescence ratio, the amount of recyclable materials in the lamp,
the energy consumed during production and the mounting arrangement for the
system, among others.
More specific results are available upon request. So far, the application of the
aforementioned methodology allowed for the design, test, manufacture and
launch in almost half the time (and 40% cheaper) compared to traditional
schemes. Furthermore, the application of this methodology, which is fully
based on PLM concepts, delivered a better ranked product in less time and at
a lower cost. The correct application of these engineering concepts combined
with PLM can be easily extended into almost any new product at its develop-
ment stage.
Acknowledgments Asociacion Mexicana de Cultura A.C. and Consejo Nacional de Ciencia y

Tecnologia (CONACYT) are kindly acknowledged for their support. Grupo ECOS Innova-
tions 1 is kindly acknowledged for supporting this initiative.
References
[1] Rein G L (2004) Creating Synergy between Marketing and Research and Development.
Journal of Product Innovation Management 21: 3343
[2] Levesque J, Fred Walker H (2007) The Innovation Process and Quality Tools. Quality
Progress 40: 18
[3] Thomas G, Scott K (2003) A Research Agenda for the Nexus of Product Development
and Supply Chain Management Process.Journal of Product Innovation Management 20:
333336
[4] Graedel T E, Allenby B R (1996) Design for Environment. Prentice Hall, New Jersey
[5] Ulrich K T, Eppinger S D (2004) Product Design and Development. McGraw Hill
Higher Education, New York
[6] Ceglarek D, Huang W, Zhou S, Ding Y (2004) Time-Based Competition in Multistage
Manufacturing: Stream of Variation Analysis Methodology Review. International
Journal of Flexible Manufacturing Systems 16: 11
[7] Kamrani A, Vijayan A (2006) A methodology for integrated product development using
design and manufacturing templates. Journal of Manufacturing Technology Manage-
ment 17: 656
[8] Romero-Hernandez S (2002) Ingeniera total asistida por computadora: menores tiem-
pos, menores costos, innovacion efectiva. Direccion Estrategica 2: 3942
[9] Grieves M W (2005) Product Life Cycle Management: The New Paradigm for Enter-
prises. International Journal of Product Development 2: 71
[10] Ayag Z, Ozdemir R G (2007) An Analytic Network Process-based Approach to Concept
Evaluation in New Product Development Environment. Journal of Engineering Design
18: 2092007
[11] Romero-Hernandez O (2004) To Treat Or Not To Treat? Applying Chemical Engineer-
ing Tools and a Life Cycle Approach to Assessing the Level of Sustainability of a Clean-
up Technology. Green Chemistry 6: 395400
[12] Romero-Hernandez S, Romero-Hernandez O (2003) A Framework of Computer Aided
Engineering and LCA applied for Life Cycle Management. In LCA/LCM International
Conference, Seattle
[13] Romero-Hernandez S, Gigola C, Romero-Hernandez O (2004) Incorporation of Effec-
tive Engineering Design, Environmental Performance and Logistics Planning for Pro-
ducts LCM. Proceedings of the 2nd World POM Conference. Production and
Operations Management Society, Cancun, Mexico
[14] International Organization for Standardization (ISO) (1997) ISO 14040 Environmental
Management Life Cycle Assessment Principles and Framework ISO 14040 E
[15] Michelsen O, Magerholm Fet A, Dahlsrud A (2006) Eco-efficiency in Extended Supply
Chain: A Case Study on Furniture Production. Journal of Environmental Management
79: 290
[16] Romero-Hernandez O, Munoz Negron D, Romero-Hernandez S (2006) Introduccion a
la ingeniera. Un enfoque industrial. Thomson, Mexico City
A Comprehensive Business Approach to Product
Realization Using Service Oriented Architecture
Vijay Srinivasan
Abstract A comprehensive business approach to product realization involves

tackling both business and technical issues. As business executives focus on
innovation both in products and in business models to realize these products
as a means to increase revenue and profit, they have identified collaboration
and partnership with external players as keys to their success. They want
flexibility and responsiveness in sharing and exchanging sensitive product
information across a global network of partners, but feel that they have not
yet been able to achieve their goals. A technical solution to this problem may
well be provided by the convergence of three recent developments: maturity of
standardized product models and processes, emergence of service oriented
architecture for information sharing, and availability of middleware to imple-
ment them. This paper describes the results of a recent business and market
research in identifying the problems facing business executives, and an informa-
tion sharing architecture with service orientation for product realization to
solve some of their problems.
Keywords Innovation Product realization Product development Product

Lifecycle Management Information technology Standards Services
Oriented Architecture (SOA) Middleware
1 Introduction
Business leaders make tough decisions every day, usually without the benefit of
all the information they would like to have. For example, lets assume that in an
industrial firm the engineering department has proposed a change to a top-selling
product. The change will make the product better, but at what cost? How many
parts in inventory will be made obsolete? What new tooling will need to be
V. Srinivasan (*)
IBM Corporation and Columbia University, New York and University of North
Carolina at Charlotte, USA
e-mail: vasan@us.ibm.com

192 V. Srinivasan
designed and developed, and how long will it take? What customers will be
affected, and how are they likely to respond? In short, is there more risk than
reward in making the change? Such questions arise in any comprehensive
approach to product realization.
Traditionally, answering these questions has required information from Pro-
duct Lifecycle Management (PLM), Enterprise Resource Planning (ERP), Cus-
tomer Relationship Management (CRM), Supply Chain Management (SCM),
and in some cases plant floor Manufacturing Execution System (MES). This
information, however, was usually available only through complex manual data
compilation and analysis. Finding the answers could take weeks, time that is not
often available or that could give a competitor an advantage in the marketplace.
But profound changes in information technology infrastructures are chan-
ging this paradigm. A powerful new approach to Information Technology (IT)
known as Service Oriented Architecture (SOA) is promising to break down the
barriers between IT systems and applications to deliver actionable business
intelligence in ways never before possible. The combination of SOA with
PLM seems to be especially powerful, enabling greater insight into the business
impact of engineering decisions, facilitating collaboration with external part-
ners regardless of the systems and applications they have chosen, and leveraging
legacy PLM investments while reducing the cost and complexity of deploying
new or updated applications. As CIMdata recently observed, SOA can sig-
nificantly reduce a (users) exposure to costly upgrades and deployment
expenses and will preserve their tailored implementations [1]. Moreover,
SOA offers the potential for companies to empower the investment they have
already made in PLM, transforming it from a tool used primarily by product
engineers into a powerful business decision support enabler that allows users to
react quickly to sharp changes in market direction or to initiate changes the
rest of the market must scramble to match.
To understand how and why PLM is poised to become a key driver of
competitive business advantage, it is important to first understand the forces at
work in the market, the forces at work on PLM, and how the two are converging
to give business leaders the required power to innovate their businesses. In
Section 2 we focus on the business drive for innovation. This leads to the
importance of a global network of partners in Section 3. SOA is discussed briefly
in Section 4 as a technical solution to the information sharing problem. Section 5
addresses the application of SOA to PLM. The resulting business benefits are
outlined in Section 6, before Section 7 summarizes and concludes the paper.
2 Business Drive for Innovation
Today, Chief Executive Officers (CEOs) are concerned with the same business
objectives that have consumed them through the ages, including revenue growth,
cost reduction, asset utilization, and risk management. But how CEOs expect to
A Comprehensive Business Approach to Product Realization 193
achieve these business objectives has made a radical shift. For several years,
CEOs have focused on product innovation as a driver of competitive advantage,
recognizing that innovation is a means to achieving all those other goals. We will
fight our battles not on the low road to commoditization, said Howard Singer,
Chairman and CEO of Sony Corporation, but on the high road of innovation
[2]. Jeffrey Immelt, Chairman and CEO of GE Corporation, recently remarked
Constant reinvention is the central necessity at GE. . . Were all just a moment
away from commodity hell [3]. Ed Zander, Chairman and CEO of Motorola
quipped All I have done since I got here is focus on one word: innovation [4].
Recently, however, their understanding of what enables innovation has
changed, leading to a new focus not just on innovating products and services,
but also on innovating the business processes and business models that influ-
ence the creation of innovative products. Simply put, innovation looks for new
ways to make money, because old ways may not work anymore. This new focus
on innovating not just products, but how products are created and commercia-
lized, is a reaction to rapidly accelerating change. In a world where competitors
can emerge overnight from anywhere on the globe and breakthrough innova-
tion can fundamentally shift the competitive landscape, the ability to continu-
ally scan the horizon for threats and opportunities and then change directions
quickly to meet or seize them has become paramount.
In a recent IBM Global CEO Study published in 2006 [5], 87% of respon-
dents said their organizations will require fundamental change to succeed in
driving innovation in the next two years (see Fig. 1). This survey population
included 765 CEOs, business executives, and public sector leaders from 20
different industries and 11 geographic regions, both from mature markets and
from important developing markets such as China, India, Eastern Europe, and
Latin America. More than 80% said their organizations traditionally have been
largely unsuccessful in managing change in the past, and only one in ten CEOs
Fig. 1 Extent of
fundamental change needed
over the next two years
194 V. Srinivasan
Fig. 2 Importance versus

extent of business and
technology integration (%
of respondents)
believed their organization had the ability to respond to rapidly changing

market conditions.
The CEOs also recognize the important role that the integration of business
and technology plays to improve the flexibility and responsiveness of their
organizations. Nearly 80% of CEOs responding to the IBM survey rated
business and technology integration of great importance but only half said
they are executing at the levels required, leading to an integration gap illu-
strated in Fig. 2. This failure is costly: Extensive integrators reported three
times the revenue increases of less integrated companies. IBMs own financial
comparisons estimate that extensive integrators grow revenues five percent
faster than their competitors.
3 A Network of Partnership
In the same IBM Global CEO Study of 2006 [5], major strategic partnerships
topped the list of significant business model innovations. As global connectivity
reduces collaboration and transaction costs, companies are taking advantage of
the expertise and scale beyond the boundaries of their organizations. They are
assembling groupings of specialized capabilities, combining, for example,
internal expertise and scale through shared services centers with the capabilities
of specialized partners to create innovative business models and processes.
When asked which sources their companies relied on for their most signifi-
cant innovative ideas, CEOs responses held some surprises. Business partners
were near the top of the list, just behind the general employee population.
External sources are not only prevalent in the ranking of CEOs most significant
sources of ideas, they also comprise a substantial portion of the overall quantity
of ideas. This trend was particularly evident among financial outperformers.
CEOs believe collaboration is absolutely critical, but there is a problem:
Although collaborative aspirations were high, actual implementation was
Fig. 3 Importance versus

extent of collaboration and
partnering (% of
respondents)
dramatically lower. Only half of the CEOs interviewed believed their organiza-
tions were collaborating beyond a moderate level. See Fig. 3 for an illustration
of this collaboration gap. Citing a lack of the skills and expertise needed to
partner externally, many CEOs refer to partnering as theoretically easy but
practically hard to do. The message is clear: whether it involves crossing
internal or corporate boundaries, collaboration requires serious intent. The
upside of collaboration is underscored not only by qualitative CEO feedback,
but also by the financial performance of companies with extensive collabora-
tion capabilities. The Global CEO Study 2006 illuminates the degree to which
strong collaborators enjoyed healthier revenue growth and average operating
margin over their competition.
The challenge of meeting these collaboration objectives for product realization
is compounded by the need to operate effectively within a value network. Today,
delivering the product the market wants requires cooperation among a complex
ecosystem of players, from the customer-facing Original Equipment Manufac-
turer (OEM) to its design partners, their suppliers, and a host of manufacturers.
Simultaneously, distributors and retailers join the network, delivering the pro-
duct to customers and providing in-field service after the sale (see Fig. 4).
Finance and business controls govern revenues and costs. But because
product development processes extend across multiple companies, much of
the information required to establish these financial targets is known only by
the partners, suppliers and manufacturers who design and source the compo-
nents. Their activities, in turn, rely on information in systems they may not fully
control, such as inventory and manufacturing scheduling. The OEMs sales and
marketing organizations, meanwhile, need visibility into all of these activities to
gather market requirements and condition the market to demand the product.
In short, in a comprehensive business approach to product realization, there is a
dire need to share and exchange information across a complex network of
partners, such as the one shown in Fig. 4, that is spread globally and whose
topology can change quickly.
196 V. Srinivasan
Fig. 4 Product realization takes place within a complex network of partners
The truth is that until recently, the IT tools and strategies available simply
were not up to the challenge of sharing information across such a complex
network of players in an efficient way. In fact, hard-wired links between enter-
prise applications compounded the very challenges they were meant to address.
Those links were also difficult, expensive and time-consuming to build because
developers had to compensate for the incompatible architectures of the systems
involved. This resulted in redundant and contradictory data and a hopeless
jumble of connections that were costly and difficult to manage and maintain.
And when business priorities changed, changing the links could take so long that
companies were late in responding to opportunities or missed them entirely.
The way business leverages information technology must therefore change
radically if enterprises are to garner the insights and achieve the agility CEOs
require to respond to business conditions. Fortunately, the development of a
Service Oriented Architecture (SOA) approach to building information tech-
nology systems promises to overcome the challenges of inefficient and inflexible
architecture through adoption of an architecture specifically designed to
accommodate rapid and frequent changes.
4 Service Oriented Architecture
SOA is an approach to enterprise computing that works in a way that resem-

bles LEGO, the popular childhood construction toy. LEGO can be assembled
into one structure today, then broken apart and reassembled into something
else tomorrow. The blocks do not change, but their modular structure allows
each block to play many different roles, depending on how it is assembled

with other blocks.
SOA works much the same way, allowing an IT analyst to break business
functions and processes into small chunks known as services and then reassem-
bling them to support different business models and processes. Services are
functions that when invoked accomplish some specific task. They expose a well
defined interface, hide their implementation details, and are scalable through
open standards mechanisms. Services can be used to perform a wide variety of
tasks, such as portfolio and program decision support; requirements, config-
uration, or engineering change management; supplier and OEM collaboration;
commonality and part reuse; analysis and simulation; or system integrity vali-
dation, to name a few.
These modules can be assembled together on a standards-based framework
to support one business model today, and a different business model when
market conditions change. SOA enables a business-centric view of the enter-
prise, orchestrating functions in terms of people, processes, and information.
SOA allows one to integrate these services with customers, partners, and
suppliers when everyone in the chain has different applications and computing
platforms. See Fig. 5 for an illustration of how the information, process, and
people-related issues can be organized in a framework to support the technol-
ogy-business integration and collaboration needs identified by the CEOs.
Fig. 5 A framework to support business needs identified by the 2006 IBM CEO survey
198 V. Srinivasan
At the bottom layer of Fig. 5 are various engineering and business applica-
tion developed and supported by different Independent Software Vendors
(ISVs). The information generated by them can be integrated using standar-
dized data models [6], web services, and middleware to implement them [7].
These services can be composed to provide flexibility and responsiveness in
various work-flows to model and execute business processes. This results in the
role-based people collaboration and provides business decision support needed
at the top in Fig. 5 as desired by the CEOs.
Just as businesses are adapting to changes in the environment, so must their
supporting systems like PLM. The consistent growth of PLM is proof of its
ability to deliver business benefits specific to design, engineering and manufac-
turing, such as providing a 3D model-based development approach and
enabling concurrent engineering to reduce time-to-market and development
costs across a wide range of industries.
SOA is emerging as a key technology for enabling such growth in PLM.
Combining PLM with SOA can deliver the flexibility vital to enabling innova-
tion and achieving desired outcomes. This could provide the much sought-after
breakthrough for integrating applications around the product realization pro-
cess and breaking down the silos that traditionally have limited PLM. By
integrating PLM with the rest of the enterprise, SOA can transform a formerly
engineering-centric solution into a federated source of all product information,
including pricing, market demand, portfolio costs and more.
Together, PLM and SOA can enable flexible, standards-based access to
product information regardless of which software applications or hardware
platforms are in use throughout the enterprise and out into the value network.
The result is total product information visibility for product and portfolio
planners, support engineers, sales and marketing even business executives,
as envisioned in the following section.
5 Applying SOA to PLM
SOA could address the needs of CEOs, delivering an entirely new approach that
rises above the complexity of current IT systems to give enterprises the insight
and ability they need to thrive in todays competitive reality. SOA allows
organizations to more easily link and share PLM product data with informa-
tion from other enterprise systems, effectively supporting approaches that
require multiple-team collaboration, such as functional design, design-for-
compliance, design-for-cost and service after sales.
Using SOA to combine the value of PLM and ERP, for example, allows
users to understand costing and inventory levels of existing components, as
well as their associative sizes and tolerances, to make better upfront design
decisions. This lowers the cost to develop new products by leveraging existing
components, which results in less excess inventory, fewer design iterations,
and faster time to market.
By combining PLM with CRM, on the other hand, SOA allows organiza-
tions to showcase products in different configurations, allowing them to better
illustrate the features in a virtual reality mode that could be highlighted in their
sales initiatives. This is but one example of how product development could be
tied more closely to market introduction processes. Similar benefits are avail-
able by federating PLM information with Requirements Engineering and Sys-
tems Definition, as well as with other downstream enterprise systems, such as
Manufacturing Planning and In-Field Service Management.
Traditionally, most PLM environments utilize multiple applications, each of
which has its own database. Each of these databases contains not just informa-
tion, but knowledge about the relationships between the information the
context in which the information makes sense. When these databases attempt
to share information through an enterprise PDM (Product Data Management)
system, a traditional engineering tool, the knowledge embedded within the data
relationships specific to each application is lost.
Attempts to overcome this limitation by integrating proprietary applications
require hard-coded, difficult-to-change links. This defeats the goal of creating
agile, flexible business models. This challenge is compounded when an OEM, for
example, attempts to work closely with multiple partners and suppliers. Now
each organization in the chain must deal not only with its own application
complexity, but the complexities of its partners as well. Barriers between dispa-
rate systems make it difficult for people in interrelated functions to collaborate.
To get around these walls, people must schedule meetings, send e-mails or
make phone calls to share information that cannot be shared by the organiza-
tions disparate systems. This ad-hoc system of collaboration is fraught with
problems, however, including failures to include key people in critical decisions,
or difficulties determining whose data to trust. This further stifles the goals of
collaboration, innovation, and flexibility.
SOA-enabled PLM can remove the problems inherent in duplicating data
from individual applications into an enterprise PDM system by creating a
federated information mechanism that all applications access and share. Busi-
ness processes exist independent of specific applications and can be viewed and
accessed by all companies participating in a product realization value network.
Portals provide access and visibility into all business processes relevant to
particular user roles.
By eliminating traditional information silos and making vital product infor-
mation visible throughout an enterprise, SOA-enabled PLM could transform
PLM from an engineering application into a source of all product informa-
tion. SOA-enabled PLM provides business decision support, increases flexibil-
ity and responsiveness and improves integration with the value chain, enabling
CEOs to continually innovate their products, their business processes and their
PLM infrastructures.
SOA can also create an ecosystem in which multiple solution developers,
integrators and IT consultants can collaborate, replacing competition with
cooperation a profound benefit to the client. By permitting heterogeneous
200 V. Srinivasan
hardware and software to operate together smoothly through a shared commit-

ment to open standards, SOA can deliver on the long-held client dream to mix
and match best-of-breed applications to achieve a system uniquely suited to
their special goals and challenges. The result is tailored client solutions for the
electronics, aerospace and defense, automotive, consumer products, and fabri-
cation and assembly industries.
6 Business Benefits
The SOA-enabled PLM approach could deliver new capabilities that would not
be possible without this flexible infrastructure. These capabilities enable a
comprehensive approach to product realization processes used within the entire
value network. Some of the benefits of this approach were mentioned in the
previous section. Listed below are a few more of the new capabilities and their
benefits enabled by SOA.
6.1 Enterprise Asset Management and Service after Sales
Information systems used for enterprise asset management and service after
sales track asset data including location, work, and cost history to control costs
and operational condition of capital equipment. Collaboration between engi-
neering and field services made possible by connecting mobile field service
technicians to PLM and asset utilization data instantly speeds the communica-
tion of field issues back to the engineering department, allowing OEMs to
identify design issues earlier, design fixes and communicate them to the field
to reduce warranty claims and improve product reliability. Linking PLM data
to asset usage in the field can improve customer service, account for proper
asset usage, and lengthen useful life.
6.2 Enterprise Integration and Collaboration

Tools for enterprise integration and collaboration link engineering disciplines
with product development stakeholders in commercial and operational functions
by connecting disparate CAD and PDM systems to other enterprise systems,
enabling an enterprise-wide view of product data that even extends into the
partner network. When connected applications are accessed by users throughout
the value network through a web-based portal, each user gets role-specific, secure
access to the critical business applications they need to do their jobs. Information
is presented to each user in a manner consistent with their role and in a way that
they can understand. These flexible access tools help to ensure better business
decision support and lower product development time and cost. Improved
integration within the enterprise and into the supply network ensures flexibility
and collaboration with people throughout the product realization process.
For example, Volkswagen AG has used such portals to improve productivity
in its procurement department by improving access to product information.
The portals have shortened order-to-delivery cycles while making the procure-
ment staff 20% more productive and improving their ability to focus on high-
value-add activities. Meanwhile, another leading OEM is using portals to
streamline the communication of design changes between its own development
teams and its suppliers, cutting the design/engineering cycle by more than a
third, reducing development costs by 25% and replacing manual distribution of
time-sensitive change orders with an automated, real-time system.
6.3 Product Performance Simulation
A modeling and simulation system simulates the behavior, flexibility and

performance of a virtual product along all development stages and across
multiple design partners. It includes desktop analysis/simulation plus digital
mock-up (DMU) to improve product quality and shorten development time by
detecting design problems early; Enterprise Simulation/CAE (Computer-Aided
Engineering), which provides complex analysis and simulation; and IT
Resource Optimization for Engineering, which optimizes the supporting infra-
structures used to run compute-intensive analysis applications.
For example, Daimler-Benz Truck Division used mechanical analysis solu-
tions earlier in its engineering developing cycle, improving its cycle time to
product introduction by 60%. Meanwhile, a leading automaker reduced the
time needed for engineers to compile their reports by employing a systematic
approach to simulation data management for CAE that managed many terabytes
of simulation data spread across several million files. The solution not only saved
time; it also improved user confidence in the resulting data. And Magna Steyr, a
Tier One automotive supplier, applied grid technology to its clash detection
environment, reducing run times from 72 hours to four hours, which reduced
costs, improved time to market and resulted in higher quality products.
6.4 Supply Chain Collaboration
Supply chain collaboration enables extended value chain creation and manage-
ment of the virtual product into the supply chain, improving collaboration and
integration through an optimized infrastructure, improving the management of
data and development processes by suppliers, reducing program risk, improv-
ing business model and process innovation support, and reducing administra-
tive costs for activities such as manual exchange, checking and data translation.
Suppliers also benefit through reduced IT resource requirements, pre-defined
202 V. Srinivasan
solutions with industry templates that reduce implementation time and

time-to-productivity by as much as 30% and a 1020% reduction in overall
engineering costs. Other likely benefits include a 3035% reduction in engineer-
ing time and a 1520% increase in profitability.
For example, to build its Falcon 7X business jet in four years rather than the
traditional seven, Dassault Aviation assembled a team of 27 design partner
firms scattered around the globe. The company employed a fully digital process
to eliminate physical prototypes, and built a collaborative workspace that
allowed all of its partners to share data online in real time. The resulting
productivity gains have allowed the aircraft company to decrease its time-to-
market plan by 30%. Similarly, an automotive OEM identified and eliminated
bottlenecks in its collaboration processes after establishing a supplier portal as a
single data exchange source with its suppliers.
6.5 Software and Systems Development
Software and system development tools accelerate and improve the design,
development, implementation and management of the delivery of software and
systems. To create virtually any new product today, it is necessary to synthesize
the engineering disciplines of mechanical, electronic, and software development.
Engineers must understand, simulate, and validate a broad set of factors that
influence product success. Customer requirements must be captured, understood,
and allocated to functions and system architectures, which can then be analyzed
and simulated again, to enable the development of an optimal design that
balances risk, cost and time-to-market constraints. Software and Systems Devel-
opment integrates mechanical and electronic design to create a systems-level view
of PLM applications and databases on a SOA, regardless of their engineering
domain (for example, electronic-CAD, mechanical-CAD, software, hydraulics,
simulation, verification) or their internal development cycles and rules. This
allows for efficient reuse of system-level design know-how to drive lean, effective,
global and innovative product realization initiatives.
These and other examples such as integrated change management, pro-
duct information management, and product and portfolio management
demonstrate just some of the benefits real companies are achieving or could
achieve with an SOA-enabled PLM approach to manage their product life-
cycle data and business processes.
7 Summary and Concluding Remarks

This paper describes a comprehensive business approach to product realization
using service-oriented architecture. The emphasis is deliberately on business
issues that can be addressed by such an architecture the technical details to
design and implement SOA can be found in various technical literature, for
example in [7]. Some of the early successes in applying SOA for product
information sharing across an enterprise can be found in [8, 9]. While much
remains to be done to fully realize the comprehensive business approach out-
lined in this paper, these initial projects have yielded encouraging results that
point to a promising future. In essence, SOA allows engineering and business
processes to be built using modular chunks of software in the form of services
that can communicate with each other and be used across different parts of a
business. As observed recently by The Wall Street Journal the approach can
save companies time and money because the software modules can be reused
and reconfigured in new ways [10]. This is perhaps the most recent and best
endorsement of SOA by the business community.
Acknowledgments This paper is based extensively on IBMs business and market research on
innovation [5] and SOA for PLM [11]. The author gratefully acknowledges these sources and
numerous IBM colleagues who contributed to these researches and their documentation.
References
1. CIMdata Program Review, September 2006. http://www.cimdata.com/php/download_
reports.php
2. Sonys Revitalization in the Changing CE World. Howard Stringers remarks, CEA-
TEC, Tokyo, October 4, 2005. http://www.sony.com/SCA/speeches/051004_stringer.
shtml
3. E. Schonfeld, GE sees the light by learning to manage innovation: Jeffrey Immelt is
remaking Americas flagship industrial corporation into a technology and marketing
powerhouse. Business 2.0. July 1, 2004.
4. At Motorola, a Hop in Everybodys Step. BusinessWeek Online. August 22, 2005,
http://www.businessweek.com/magazine/content/05_32/b3946107_mz063.htm
5. IBM Business Consulting Services, Expanding the Innovation Horizon Executive
Summary of Global CEO Study 2006, publication G510-6258-00, March 2006, http://
www.ibm.com/ibm/ideasfromibm/us/enterprise/mar27/ceo_study.html
6. V. Srinivasan, Standardizing the specification, verification, and exchange of product
geometry: Research, status and trends, Computer-Aided Design, 2007.
7. IBM Systems Journal, Special Issue on SOA, Vol. 44, No. 4, 2005.
8. V. Srinivasan, An integration framework for product lifecycle management, under
review in Computer-Aided Design.
9. V. Srinivasan, L. Lammer, and S. Vettermann, On architecting and implementing a
product information sharing service, ASME Journal of Computing and Information
Science in Engineering, Special Issue on Engineering Informatics, March 2008.
10. C. Lawton, SOA stirs a computing buzz modular software lets services be used
across businesss parts, The Wall Street Journal, July 17, 2007.
11. IBM Product Lifecycle Management, Empowering Product Lifecycle Management with
Service Oriented Architecture, publication 07-000772, March 2007, http://t1d.www-03.
cacheibm.com/solutions/plm/doc/content/bin/07-000772_final.pdf
Integration of Collaborative Engineering Design
Using Teamcenter Community in Mechanical
Engineering Curricula
Xiaobo Peng, Ming C. Leu, and Qiang Niu
Abstract This paper presents a collaborative project conducted by Prairie

View A&M University (PVAMU) and University of Missouri-Rolla (UMR)
to jointly develop collaborative engineering design instructional projects,
utilizing the Teamcenter Community and NX3 software provided by UGS
PLM Solutions, and to implement these collaborative design projects in the
teaching of two CAD courses in the Fall 2006 semester at the two universities.
The collaborative design projects have been implemented in both the freshman
level course MCEG1213 Creative Engineering at PVAMU and the senior/
graduate level course ME363 Principles and Practices of Computer-Aided
Design at UMR. This paper describes how to effectively prepare such a colla-
borative effort and how to implement the Teamcenter Community software. It
was observed in this project that thorough communication among the team
members is the key to guaranteeing the success of collaborative design. This
paper also discusses the challenges in conducting a collaborative engineering
design. The planned improvements for the future implementation are discussed.
Keywords Collaborative Design Teamcenter Community Curriculum

PLM
1 Introduction
The globalization of the economy has significantly changed the way the industries
do business. Many companies have outsourced their design and manufacturing
factories overseas. Therefore, global collaboration has become necessary in
industry. Product Lifecycle Management (PLM) has made the collaboration
possible. As defined by Grieves [1], PLM is an integrated, information-driven
approach comprised of people, process/practices, and technology to all aspects of
X. Peng (*)
Mechanical Engineering Department, Prairie View A&M University, Prairie View,
Texas 77446, USA
e-mail: xipeng@pvamu.edu

206 X. Peng et al.
a products life, from its design through manufacture, deployment and main-
tenance culminating in the products removal from service and final disposal.
With such a PLM system the members of a design team in geographically
dispersed locations are able to interact with each other effectively in the colla-
borative design process.
One major goal of the higher education is to produce high quality students
who shall have the skills working in the changing global economic environ-
ment when they graduate. Since the last decade, the CAD/CAM/CAE has
been widely integrated into mechanical engineering curricula due to the
demanding needs from the design and manufacturing industry. CAx has
been considered as the mandatory training for engineering students. Today
the growing interest in PLM in the industries has demanded college education
to impart engineering students the necessary skills for collaborative design in a
distributed environment.
To meet this emerging need, Prairie View A&M University (PVAMU) and
University of Missouri-Rolla (UMR) have conducted the pilot collaborative
project to jointly develop collaborative CAD engineering design instructional
projects, utilizing the Teamcenter Community and NX3 software from UGS
PLM Solutions. These collaborative design projects were implemented in the
teaching of two CAD courses, i.e., MCEG1213 course at PVAMU and ME363
course at UMR in Fall 2006. The students at both universities teamed up to
work on the collaborative design projects, by dividing up the project tasks
among different team members. The members of each team consisted of stu-
dents from both universities. This project is funded by PACE program (Part-
ners for the Advancement of Collaborative Engineering Education).
In addition to the demanding needs from the industry described above, many
research papers have reported that integration of collaboration activities using
on-line tools can enhance learning, reduce the sense of isolation, increase partici-
pants learning motivation, and improve students social interaction skills [2, 3, 4].
From the social and psychological perspective, Hughes et al. [4] provided a
comprehensive survey of literatures to address how to establish the most effective
on-line collaboration. The paper summarized four aspects of effective collabora-
tion, including: (1) getting students to understand the value of collaboration;
(2) establishing comfort with the technology used in the collaborative environ-
ment; (3) building comfort and trust among students and between instructor and
students; and (4) creating a rich on-line social environment.
Several universities have developed a program of collaborative design and
product data management in their curricula. Purdue University has developed a
framework to integrate design using CAD, rapid prototyping, team-based
collaboration projects and realistic constraints and budgets in a computer-
aided design and prototyping class [5]. However, their collaboration was con-
ducted within one university. Tomovic of Purdue University [6] introduced the
PLM experience to the senior design students with integration of part design,
design optimization, and design of manufacturability. Georgia Tech and
University of Maryland College Park implemented a collaborative product
Integration of Collaborative Engineering Design 207
development project in their traditional engineering course using PTC Projec-

tLink system [7]. In 2004 and 2005, Virginia Tech and the Technique University
Darmstadt jointly offered a new course on collaborative engineering and pro-
duct data management [8]. The transatlantic term project was conducted using
Teamcenter Community. In 2004, Brigham Young University (BYU) and
Virginia Tech (VT) integrated PLM tools and collaboration into a capstone
student project to design a concept vehicle. The students from three disciplines
comprised the design teams to collaboratively complete the project [9]. In 2005
and 2006, BYU and VT expanded the collaborative project to a global scope
with eleven other participating universities around the world. The requirements
needed to either host a global collaboration project or be a participant were
discussed [10]. Michigan State University and ITESM implemented collabora-
tive student design projects utilizing Teamcenter Engineering and Teamcenter
Community tools in the International Networked Teams for Engineering
Design (INTEnD) program [11]. Anderson et al. [12] discussed the issues on
the implementation of Teamcenter Community and collaborative environment
in academia. Most of the above projects were implemented in the senior/
graduate level and two-semester long courses. By deploying the pilot project,
we are seeking to develop guidelines for integrating collaborative engineering
design in regular one-semester engineering curricula.
This paper is organized in the following manner. First, we introduce the
background of PACE program and two participating universities. In Section 3,
details are described on how to establish the infrastructure for collaborative
engineering design and how to implement the PLM tool Teamcenter Commu-
nity. Next, the issues on implementing the collaborative project in engineering
curricula are addressed in Section 4. Challenges, lessons learned, and benefits in
conducting a collaborative engineering design are discussed in Section 5. At the
end, the conclusion and planned improvements for the future implementation
are discussed.
2 Participating Universities
Both participating institutions, Prairie View A&M University and University of

Missouri-Rolla, are in the PACE program. PACE (Partners for the Advance-
ment of Collaborative Engineering Education) is a consortium established by
leading industries GM, EDS, Sun Microsystems, UGS, HP, and their global
operations. Since it was established in 1999, PACE now is partnering with over
forty (40) academic institutions worldwide to develop the Product Lifecycle
Management (PLM) teams of the future [13]. In a cost-effective manner, PACE
and its contributors provides hardware, software, training, and funding to
PACE institutions to educate globally engaged engineering and design students
with necessary skills to be successfully in a global economy. The software
available for PACE institutions contains prestigious CAD/CAM/CAE and
208 X. Peng et al.
PLM tools currently used by leading industries. These include Alias Maya
Unlimited, Alias AutoStudio, Altair HyperWorks, E-factory Toolkit, FLU-
ENT, LS-DYNA, MSC.ADAMS, MSC.Nastran, NX (previously Uni-
graphics), Teamcenter Community, and Teamcenter Engineering, etc.
Founded in 1876, Prairie View A&M University (PVAMU) is the second
oldest public institution of higher learning in the state of Texas located 30 miles
away from Houston. As part of the Texas A&M system, PVAMU is one of the
Historically Black Colleges and Universities (HBCU) in which approximately
96% of the students are classified as ethnic minorities. Prairie View A&M
University is known as one of the nations top producers of African-American
engineers. At PVAMU, the MCEG1213 course Creative Engineering is a
mandatory freshman course in Mechanical Engineering. This course introduces
students the basic knowledge of engineering drawing and computer-aided
design and teaches student how to use CAD software Unigraphics NX3 for
engineering design and analysis.
Founded in 1870, the University of Missouri-Rolla (UMR) has been one of
the top among technological research universities. UMR was recognized as a
top 50 Engineering School by US News & World Reports Americas Best
Colleges, 2006. At UMR, the ME363 course Principles and Practices of
Computer-Aided Design is offered to graduate and upper-level undergraduate
students every year. This course introduces the fundamentals of computer-
aided design and engineering with emphasis on geometric modeling. Uni-
graphics NX3 software is used by students to gain practical experience as well
as to help them grasp the theories of computer-aided design.
Both ME363 and MCEG1213 at the two universities have an emphasis
on developing students hands-on experience on applying NX3 to solve
engineering problem. In the old curricula of both courses, final team
projects were assigned to students to design a product. The final projects
aim is to provide students the opportunities to apply solid modeling,
engineering drafting, and assembly skills to attack a real design problem.
The projects develop in students the capabilities of report writing, giving
presentations, and working as a team. Because both universities have
similar final project assignments and have common interests in integrating
PLM in the curricula, the integration of collaborative design in the final
project has become very smooth.
3 Implementation of Teamcenter Community
Teamcenter Community (TcC) is the major tool used in our collaborative

design project. Provided by UGS, Teamcenter Community is a secure encrypted
platform with collaboration functionality that enables real-time synchronous
and asynchronous engineering collaboration between geographically dispersed
teams. As the host University of the collaborative project, PVAMU is in charge
of installing the software, Teamcenter Community. The collaborative project

has served as a catalyst for PVAMU to build the infrastructure to support
collaborative engineering design. A state-of-the-art collaborative design lab is
under development with the support of PACE and College of Engineering at
PVAMU.
3.1 Teamcenter Lab at PVAMU
A PACE Teamcenter Lab is under development at PVAMU since the

beginning of 2006. The objective of the establishment of this lab is to provide
a state-of-the-art collaborative engineering design environment including
hardware and software to support the education and research activities
related to collaborative projects. The hardware and software are available at
PVAMU including the following:
Hardware
Two High-end Sun servers (2100)
One HP large format printer
Twenty five high-end engineering workstations
Tandberg video conference system
Software
Unigraphics NX3
Altair Hyperworks 7.0
ANSYS
Teamcenter Community 5.2
Video and audio conferencing software.
3.2 Deployment of Teamcenter Community
The Teamcenter Community (TcC) V5.2 software was donated by UGS. We

obtained two Sun servers which were granted by PACE. One faculty and one
technician were fully dedicated to the implementation of TcC at PVAMU.
Teamcenter Community runs on top of Microsofts Windows SharePoint Ser-
vices (WSS). MS SQL Server 2000 is required as backend database for TcC.
Windows Server 2003 is required as the Operating System. There are many other
software and services that are prerequisite to run TcC, namely, Internet Informa-
tion Services 6.0 (IIS 6.0), ASP .NET service, Microsoft Active Directory (AD),
and Domain Name Service (DNS). It is recommended to separate the TcC Web
Server from the Database Server. As the nature of the prerequisites is complex, it
is quite challenging to deploy the Teamcenter Community.
The Teamcenter Community Installation and Upgrade Guide [14] has
provided great instruction to the deployment and installation of TcC. In the
210 X. Peng et al.
case that trouble was met and no solution could be found in the guide, the UGS
technical support has been an excellent resource in providing help. Normally we
were contacted promptly by the technical person. The trouble was able to be
resolved within one or two days. The implementation was relatively smooth
although no training course has been taken.
Before the installation of the TcC, we needed to select the proper deployment in
determining the number of servers needed and how the software should be
deployed on the servers. The enrollment of two courses generally were 2030
students in ME363 (UMR) and 1525 students in MCEG1213 (PVAMU). The
estimated concurrent users were under 100. Therefore, the deployment indicated in
the guide [14] which can support 900 concurrent users was adequate for this project
and can still provide enough room for expanding the scope of the collaboration.
Two Sun servers were configured as Web Server and Domain Controller
respectively as illustrated in Fig. 1. The software installed on each server is listed
as below.
Web Server
Windows 2003 Server
Internet Information Server (IIS) 6.0
Windows Share Point Service (WSS)
Teamcenter Community V5.2
Domain Controller and SQL Server
Windows 2003 Server
Domain Name Service (DNS)
Active Directory (AD)
SQL Server 2000.
Web Server
http://teamcenter.pvamu.edu
Windows 2003 Server
PVAMU IIS 6.0 UMR
Windows Share Point Service (WSS)
Teamcenter Community
Internet Client
Client
Domain Controller Client

Client Windows 2003 Server
Domain Name Service (DNS)
Active Directory (AD)
SQL Server 2000
Fig. 1 Deployment of Teamcenter Community at PVAMU

It has been indicated that it is challenging to grant the access of the Team-
center Community to the outside of the university firewall [8], [10], [12]. If there
is no entry of the firewall, virtual private network (VPN) has been an alternative
to allow outside user accessing of TcC. However, it has been demonstrated that
using VPN has been a problematic strategy due to the blocked internet port [10].
The IT department has full control over the network setting in the academia
environment. Due to the security concern, the IT normally is very restrictive on
granting the entry of the firewall. Fortunately the IT department at PVAMU
has been very supportive in granting a firewall entry to the Web Server when the
significance of the project was explained to them.
3.3 Teamcenter Community Website
Teamcenter Community (TcC) served as central location for the collaboration

and every team member were able to log in using a web browser. TcC enables
team members geographically dispersed to share 2D and 3D part files, docu-
ments, and presentations, schedule calendars, post announcement and pro-
gress, and route review tasks etc. The features provided by TcC hosted at
PVAMU are described in detail below.
Two layers of TcC websites were constructed for this collaborative pro-
ject. The top layer was the homepage for instructors. The bottom layer
consisted of websites which were accessed by student teams. As shown in
Fig. 2, the instructors homepage was restricted to be accessed by the
instructors and teaching assistants. The major component of the homepage
was the Professors Announcement. The Announcement was made as a
pushdown list which can be automatically distributed to all the team sub-
sites. This was where the project assignments, tutorials, instructions, and
other timely announcements can be uploaded and then distributed to all the
team subsites. The Team Subsite Links on the homepage provided a
convenient way for the instructors to access each teams webpage and moni-
tor the progress of all the teams.
Team subsites were designed for each team as shown in Fig. 3, which were
logged in directly by designated team members. It contains the following
components:
1. Professors Announcement Students can get announcements, assignments,
tutorials and other project related material handed out by the instructors.
This is distributed from the parent site. The students can not edit it.
2. Task Lists Each team uses it to assign the design tasks to specific team
members and to keep track of work that each team member needs to
complete. Description of the tasks, assigned team members name, due
date, status, and completion percentage are able to be tracked using this
feature.
212 X. Peng et al.
Fig. 2 Teamcenter Community homepage of collaborative project for the instructors
3. Contact Information Members can exchange their names and contact infor-
mation (such as telephone number, e-mail address, and living address) here.
4. Announcement The students can post news, meeting announcements,
and other short information which they want to share with their team
members.
Fig. 3 Collaborative project TcC homepage for teams

5. Routing Slips Routing items for approval is a powerful feature of Team-

center Community. A team may frequently need to conduct reviews. Using
Routing Slips, the user can create a Routing Slip, select documents to route,
and make sure it gets to the right people for review. Team members can route
items such as documents, 2D or 3D data, visual issues, or requirements to
members for review and approval. When a user logs into TcC, all of the
current routing slips assigned to him/her are displayed. From the list the user
can open the routing slips, review the items, and take actions to either
approve or reject the slip.
6. Links The links are used to post hyperlinks to webpages of interest to the
team.
7. Shared Documents This document library is mostly used by students to
upload NX3 part files, 3D assemblies, and Office documents. Users can
organize different types of file by creating directories. The user can open
reports, presentations, spreadsheet, PDF file etc. directly in the TcC with-
out downloading the files to local computer. To prevent the conflict of
more than one user using the same file in the library, TcC provides a
Check in/Check out function. It is very convenient to load an assem-
bly CAD model in TcC. The user only needs to select the assembly CAD
file. All the associated sub-component CAD files will be uploaded
automatically.
8. Calendar Similar to MS Outlook, the calendar is used to post information
about dates that are important for the team as shown in Fig. 4. One unique
feature of the calendar in TcC is a workspace can be attached to an event.
The workspace is a site where the team can manage the meeting agenda,
minutes etc.
9. Iseries Viewer It is a tool embedded in TcC that provides 2D/3D visualiza-
tion, review and markup capabilities. It is enabled by the CAD-neutral JT
format which can be created from all major CAD applications. Students can
perform the synchronizing design, review and markup within the TcC by
open the JT file of an assembly or a single CAD model as shown in Fig. 5. It is
a very powerful tool for collaborative design.
3.4 Client Software Download
To fully utilize the functions of Teamcenter Community such as visualizing JT

3D models, uploading the files to the server, downloading documents, and
performing seamless editing of Microsoft Office documents with Microsoft
Office 2000/XP etc., the plug-ins need to be downloaded and installed on the
client computers. In the public computer labs at both PVAMU and UMR,
students do not have the Administrator privileges to install the software. The
Administrator must be involved in this process. This caused some delay for the
214 X. Peng et al.
Fig. 4 TcC calendar
Fig. 5 Visualize 3D CAD assembly in TcC using Iseries View

student to start using TcC. Some of the students found out it was more
convenient to use their personal computers.
Tutorials on how to install TcC client software and how to use TcC were
developed and distributed to students. Students were trained in the class on how
to use TcC. It was reported by students that most of the functions of TcC are
easy to learn such as Shared Document, Calendar, Tasking, Announcement etc.
Because students had little knowledge on project management, the functions of
TcC related to project management such as routing slips were seldom used.
Another reason was the scale of the collaborative project was small. The
advantages of the using those functions did not appear appealing to students.
4 Implementation of the Collaborative Project
4.1 Collaborative Project Assignment
After the construction of Teamcenter website, the instructors from two uni-
versities designed three pre-defined topics as the final project which required
students from UMR and PVAMU to team up as groups. Each group was asked
to choose one design subject to work on. The final project required students to
complete a conceptual design of a new product. The final design must be an
assembly. In addition, the students were requested to address material selection,
manufacturing concern, and cost analysis of the product. When the instructors
designed the project topics, the following aspects were taken into consideration:
1) The scale of the project should be appropriate such that students can
complete it within six weeks; 2) The products to be designed by students should
have adequate complexity which can give students challenges; 3) The products
can be easily broken down into sub-components so that team members can
collaboratively complete the design. The design topics were:
Backhoe design excavating equipment consisting of a digging bucket at the
end of an articulated arm.
Tandem bicycle design a bicycle to be built for two people sitting one behind
the other.
Folding bicycle design a bicycle which can be folded for easy transportation
and storage.
There were 22 students enrolled in ME363 at UMR and 14 students
enrolled in MCEG1213 at PVAMU respectively. Each group consisted of
four students. A total of nine groups were formed by one of the following
formats: 1) Five groups consisted of two UMR students and two PVAMU
students, and 2) Four groups consisted of three UMR students and one
PVAMU students. Each group had one group leader assigned by the instruc-
tors. The group leader can make final decision when there are different
opinions during the collaboration. Each team had a Teamcenter website:
216 X. Peng et al.
http://teamcenter.pvamu.edu/collaborative/teamX (X is from 1 to 9). For

example, for group one, their website is http://teamcenter.pvamu.edu/
collaborative/team1.
By the time the project was assigned to students, students had six weeks to
complete the project. Each team was required to submit a report and give a
presentation. All the documents including NX3 part files, the final assembly
file, reports and PowerPoint slides were required to be uploaded to the Team-
center Community website. To incent students to use TcC, the active and
sufficient usage of TcC was applied as one of the criteria to grade the project.
4.2 Collaboration Activities
It was found by Wegerif [15] that participants in a collaboration project are tend
to be anxious and defensive to collaborate unless trust and comfort are estab-
lished among the team members. It is observed in this project that thorough
communication among the team members is the critical key to guaranteeing the
success of the collaborative design. Various collaborative tools besides Team-
center Community such as Video Conference, Email, and Instant Messenger
etc. were utilized to assist the communication between the team members.
Teamcenter Community was used extensively by students, and was the
primary means of transferring NX3 part files and data. Utilizing TcC, each
group member was able to express questions and concerns, receive answers and
post their completed part files. In addition, TcC allowed group members to view
relevant project files and a project timeline, both of which assisted in the timely
completion of the final project.
It has been suggested that an initial face-to-face meeting will significantly
enhance the trust and familiarity among team members and a willingness to
collaborate [4]. However, it was not feasible to have a face-to-face meeting
for our students due to the time and cost constraint. On the other hand, the
basic idea of the collaborative design projects is to train students to be able to
collaborate with partners they never meet before. The video conferencing
system was used at the initial stage of the project to get the team members
from two campuses know each other and discuss the proposal. Figure 6
shows that two PVAMU students were discussing the project with team
members from UMR using video conferencing. In addition to video confer-
encing, Email, Messenger, and Phone etc. were used by students. The
instructors did not impose any use of collaboration tools or collaboration
activities other than TcC and first Video Conference meeting. Different
groups had different opinions on using the collaboration tools. Some groups
met weekly using Messenger. Some groups did not have regular meeting.
Instead they called each other when they found it necessary. There was one
group that used email exclusively for exchanging ideas without calling each
other. It was interesting that the communication in the groups which had
Fig. 6 Students discussing

the project via video
conferencing
two PVAMU and two UMR students were more active than those groups
which had one PVAMU and three UMR students.
4.3 PACE Competition and Results
PACE provides student design competitions in the courses to encourage

students to use digital data in the product development and analysis processes.
We have integrated the students final project presentation with the PACE
competition. Students presented their projects simultaneously and interactively
with their partners from two sites through video conferencing system as shown
in Fig. 7. WebEx was used to share and present PowerPoint slides
Fig. 7 Final presentation

competition using video
conferencing
218 X. Peng et al.
Stick
Boom Cab
Engine Hood
Self-leveling
Linkages
Rear Bucket
Stabilizer Legs Bucket
Fig. 8 Backhoe loader
simultaneously while students were doing the final presentation. The presenta-
tions were evaluated by the panel of judges based on three criteria including
solid modeling and assembly modeling, report and presentation, teamwork and
collaboration. The panel of judges consisted of two faculty from UMR and
PVAMU, two representatives from Altair Engineering Inc. (PACE contribu-
tor), and one representative from GM. Three top teams were awarded prizes.
All the students received certificates provided by PACE.
Over the nine groups, seven groups chose Tandem Bicycle as their projects.
Two groups chose to design a Backhoe. Figure 8 shows the top design, a
Backhoe Loader. The backhoe loader design was broken down by students
into four major subassemblies, i.e., Loader, Backhoe, Cab, and Body. Each
team member was responsible for creating one of the subassemblies. Figure 9
shows the Cab designed by PVAMU student.
Figure 10 shows a Miami Beach Tandem Cruiser designed by Group 4.
The tandem bicycles main functional requirement is to seat two people and
have arrangements so that both passengers can pedal it. The students made
some improvement on the bike to make it safer and faster, in order to market
to a younger market audience. The frame was elongated to make it more
aerodynamic and stable in turns. The wheels and frame were lightened by
using carbon fiber. Knobby tires were put on to give the appearance of being
tough and rugged, while also giving the rider the freedom to take dirt or
gravel trails. The disc brakes made the bike safer in wet conditions and the
handlebars were placed lower, so the rider has a more aggressive stance while
riding.
Fig. 9 Cab subassembly of a

backhoe loader
Cab Subassembly Cab Housing
Steering Wheel Instrument Panel Seat
Fig. 10 New generation of Miami Beach Tandem Cruiser

220 X. Peng et al.
5 Discussion
5.1 Challenges
Although the experiences in conducting collaborative/distributed design pro-

jects have been shared by many previous works [5, 6, 7, 8, 9, 10, 11, 12], it still
remains challenging to integrate collaborative design into engineering curricula.
The challenges rely on the following:
1. A solid infrastructure of PLM solution is required in order to conduct
collaborative design. Teamcenter Community is proved to be an efficient
collaboration tool. Its implementation needs comprehensive support of both
hardware and software as described in Section 3.2. To secure the adequate
manpower support and the commitment from the IT department is essential
to the implementation of TcC.
2. The collaboration needs the participating universities to have common
interests and to be willing to accommodate their schedule for each other.
It is important to identify the answers for the following questions before
initiating the collaboration: Which courses will adopt the collaborative
projects? Do the schedules at the participating universities fit each
other? What would be the appropriate design subjects for collaboration
without adding too much overhead burden on the current curricula to
the students?
3. Thorough communication among the team members is the critical key to
guaranteeing the success of the collaborative design in a distributed environ-
ment. Although the scale of our collaboration is small, being conducted by
students between two domestic universities, students all agree that the com-
munication between team members is the most challenging part of the
project even with many the collaboration tools available.
4. How to motivate students to be engaged in and to appreciate the colla-
boration project is challenging, given that engineering students already
have a heavy course schedule. It was observed that UMR students were
more engaged and active than PVAMU students in the collaboration.
Partly it is because UMR students, who are senior or graduate students,
are more mature than PVAMU students who are in their first semester in
college. The format of the groups caused different degrees of engagement.
Generally those groups with 2 UMR students and 2 PVAMU students
were more active than groups with 3 UMR students and one PVAMU
student.
5.2 Lessons Learned

Upon the completion of the collaboration projects, we have learned the follow-
ing lessons:
1. Although six weeks were given to students to complete the projects which is
longer than the time given for non collaboration project, students still felt the
time was very limited. We should to be able to initiate the collaboration
earlier next time with the TcC in operation.
2. Misunderstanding and miscommunication occurred in some teams during
the collaboration. Because no weekly review meeting or progress reports
were requested, the instructors did not perceive the problems promptly. This
caused a delay in solving the problems. In the future, regular review meeting
and progress reports should be required.
3. More training on TcC should be provided to students. The progress of the
project was slowed down because students had different level of technical
skills. This problem became more obvious when undergraduate freshman
collaborated with graduate students.
4. Some important functions related to PLM were not used in this project such
as the workflows, task distribution, and application sharing functions due to
the time constraint to implement them. Application sharing allows the users
to share someones desktop with any other collaborators. By joining a visual
conference in TcC, all participants can view the hosts desktop and even
control the remote desktop if the host transfers the control to any
participant.
5.3 Students Comments

The participating students have given positive feedback:
To make sure that all the students were working in parallel was challenging,
but by using the Teamcenter Community website, a timeline was established
and adhered to fairly closely.
The idea of collaborating with another University is good one. Apart from
sharing ideas, there is also a scope for socializing.
This project was a wonderful experience for us. This project helped us grow
as future professionals. While doing this project we learned to trust our
teammates. We polished our communication skills in a real world setting.
Regular and frequent communication between team members proved to be
an essential component of this teams success in assembling a complete
backhoe loader model.
In the end, teammates came to the conclusion that team work is important,
but it is hard work.
5.4 Benefits
We have built a collaborative design infrastructure at PVAMU. Teamcenter
Community has been successful deployed and applied in the distributed design
222 X. Peng et al.
projects collaborating by UMR and PVAMU students in two CAD courses.

The benefits of integrating collaborative design projects into Mechanical Engi-
neering curricula are:
1. By implementing the developed collaborative design projects in the teaching
of CAD courses, the students learned effectively how to perform collabora-
tive product design with different techniques and CAD systems in a distrib-
uted environment and to understand the collaborative nature of product
design and related issues.
2. Because the collaborative design projects are stimulating to students, we will
be able to encourage more minority students to pursue an engineering career.
With the skills trained by participating in the projects, our students will be
more competitive in the job market.
3. We have deployed the latest CAX/PLM technology at PVAMU and UMR.
The pilot project will help us to develop guidelines and best practices for
integrating PLM solutions into the design process upon completion of this
project in the future.
6 Conclusion and Future Work
In Fall 2006 UMR and PVAMU have successfully integrated collaborative

engineering design projects using Teamcenter Community in two CAD courses.
The implementation of Teamcenter Community is described in this paper.
Various collaborative tools besides Teamcenter Community, including Video
Conference, Email, and Instant Messenger etc., were utilized to assist the
communication between the team members. We have discussed the challenges
of integrating collaborative design into engineering curricula. By implementing
the developed collaborative design projects in the teaching of CAD courses, the
students learned effectively how to perform collaborative product design with
different techniques and CAD systems in a distributed environment and to
understand the collaborative nature of product design and related issues.
It was observed that students lack skills on communication and project
management which are critical to the success of collaboration. We will expand
the collaborative design projects by incorporating project management funda-
mentals into the design process. Dr. Benjamin Dow from Engineering Manage-
ment and Systems Engineering Department at UMR will bring his expertise on
project management to the collaboration and lead his students in the collabora-
tive engineering design projects. The students from two disciplines and three
different classes will team up to work on the projects. The students of each
school will be provided basic project management instruction and incorporate
these fundamentals into the design project using Teamcenter Project. The soft-
wares capabilities allow the design teams to break down complex design
projects into specific tasks, assign resources, and manage the workloads of the
resources. This project is funded by PACE program as well.
Acknowledgments The authors would like to acknowledge the generous support of the PACE
program/GM Foundation in providing funding, hardware, software and technical support
for this project. Dr. Milton Bryant, Dean of College of Engineering at PVAMU, and
Dr. Shield Lin, Department Head of Mechanical Engineering at PVAMU have provided
tremendous support to this project. We would also like to thank Mr. Chris Galvez in College
of Engineering at Prairie View A&M University for his technical support of implementing
Teamcenter Community.
References
[1] M. Grieves, Product Lifecycle Management: Driving the Next Generation of Lean Think-
ing. McGraw-Hill, 2006, p. 39.
[2] J. Clark, Collaboration Tools in Online Learning Environments, ALN Magazine, 4(1),
2000.
[3] S. R. Hiltz, N. Coppola, N. Rotter, M. Turoff, and R. Benbunan-Fich, Measuring the
Importance of Collaborative Learning for the Effectiveness of ALN: A Multi-measure
Multi-method Approach, ALN Journal, 5(2), 2000.
[4] S. C. Hughes, L. Wichersham, D. L. Ryan-Jones, and S. A. Smith, Overcoming Social
and Psychological Barriers to Effective On-line Collaboration, Educational Technology &
Society, 5(1), pp. 8692, 2002.
[5] A. Lee, D. Anderson, and K. Ramani, Toying to Learn for 21st Century Product
Development Environments: Computer-Aided Design, Collaboration, and Rapid Pro-
totyping, in Proceedings of the 2003 ASEE Annual Conference & Exposition, Nashville,
Tennessee, June 2225, 2003.
[6] M. Tomovic, Integration of PLM Experience in Senior Design Course, in Proceedings
of ASEE/IEEE Frontiers in Education Conference, Savannah, Georgia, October 2023,
2004.
[7] T. W. Dennis and R. E. Fulton, Undergraduate Distributed Collaborative Engineering
Project Using CAD, CAE and PLM Tools, in Proceedings 2004 ASEE Southeastern
Section Annual Meeting, Auburn, Alabama, April 46, 2004.
[8] J. H. Bohn and R. Anderl, A First Transatlantic Course on Collaborative Engineering
and Product Data Management, in Proceedings of PACE Annual Forum, Mexico City,
Mexico, July 2730, 2005.
[9] C. G. Jensen, C. J. Dean, H. Bohn, and J. Marshall, Using Collaboration and PLM
Tools to Link Remote Capstone Team Members, in Proceedings of PACE Annual
Forum, Mexico City, Mexico, July 2730, 2005.
[10] J. B. McCammon, B. W. Brooks, and C. G. Jensen, Hosting/Participating in Global
Collaborative PACE Projects, in Proceedings of PACE Annual Forum, Provo, Utah,
July 2429, 2006.
[11] T. J. Hinds and J. R. Lloyd, Teamcenter Tools Collaboration Project, in Proceedings of
PACE Annual Forum, Provo, Utah, July 2429, 2006.
[12] J. Anderson, M. Eastein, B. Nielsen, and T. Rochow, Virtual Engineering Design
Collaboration: A Journey into Uncharted Territory, in Proceedings of PACE Annual
Forum, Provo, Utah, July 2429, 2006.
[13] Partners for the Advancement of Collaborative Engineering Education (PACE), http://
www.pacepartners.org/.
[14] UGS, Teamcenter Community Installation and Upgrade Guide.
[15] R. Wegerif, The Social Dimension of Asynchronous Learning Networks, The Journal
of Asynchronous Learning networks, 2, pp. 3449, 1998.

Product Realization

Uploaded by

Copyright:

Available Formats

Product Realization

Uploaded by

Document Information

Original Description:

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

Product Realization

Uploaded by

Copyright:

Available Formats

Product Realization:

ISBN: 978-0-387-09481-6 e-ISBN: 978-0-387-09482-3

# Springer ScienceBusiness Media, LLC 2009

Printed on acid-free paper

Comprehensive product realization, or Product Lifecycle Management (PLM),

Old Dominion University, Norfolk, VA, USA Mileta M. Tomovic

Product Life Cycle and Assessment Parameters . . . . . . . . . . . . . . . . . . . . 1

Collaborative Optimization and Application of Active Suspension. . . . . . . 21

Back to the Future: Product Lifecycle Management and the

Measuring the Impact of Product Lifecycle Management: Process Plan,

Reliability-Based Collaborative Design Platform for Hydraulic

The 4+1 Dynamic Management System of Lifecycle . . . . . . . . . . . . . . . . 91

A PLM Tools Taxonomy to Support Product Realization Process:

Social Issues of Product Lifecycle Management: Developing Cross Cultural

Product Design Optimization: An Interdisciplinary Approach . . . . . . . . . . 169

A Comprehensive Business Approach to Product Realization

Integration of Collaborative Engineering Design Using Teamcenter

Keywords PLC PDM Lifecycle Assessment Product Validation Product

M.M. Tomovic, S. Wang (eds.), Product Realization: A Comprehensive Approach, 1

cybernetic model has transformation of material, information and energy. A

that enable an optimum solution in a given environment. It is the optimality of

The characteristic constraints of the proposed presentation of the product

2 Description of Individual PLC Phases

Specification of design requirements:

Product perfection during conquering phase

40% Very fast first part

Product perfection in use phase

'Normal D&DP and small lose product perfection in use phase

Product perfection in elimination phase

Natural material fast recycling process - dangerous material

Fig. 5 Phase of product elimination

Product perfection during life cycle

Fig. 6 Product life cycle through all its phases

3 Determination of Variables in Individual Phases

Fig. 7 Influence of individual criteria during individual phases of PLC

VP,U Profitability 3010%

recommendations. Their general range cannot be determined with a sufficient

depend from country to country and international geographical and

j index for particular parameter inside the different criteria

Product validation using different criteria

Fig. 8 Variation of product assessments (various criteria) during the PLC

(a) Product validation with technical FI technical

Fig. 9a Product validation with technical and commercial criteria

(b) Product validation with financial

Fig. 9b Product validation with financial and production criteria

(c) Product validation with user FI user

Fig. 9c Product validation with user and ecological criteria

4 Trends of Assessment According to Individual Criteria

When assessing individual parameters in a certain category (tec,com,fin,pro,

FSBP 1  x; tSEN (3)

x; t product validation function by typical criteria named x, defined in

Sensibilness of the FSBP function is mostly defined with potential coefficient

FSBP assesment at technical and

function and the designers, researchers, salesman, controlling department and

Hao Wang, Shaoping Wang, and Mileta M. Tomovic

Abstract This paper addresses the current state of collaborative optimization

Keywords Collaborative Optimization Compatibility constraint

M.M. Tomovic, S. Wang (eds.), Product Realization: A Comprehensive Approach, 21

Multidisciplinary design optimization (MDO) is a field of engineering that uses

FSBP 1 x; tSEN (3)

x; t product validation function by typical criteria named x, defined in

l0d ld cos lu ; ld ; lk (6)

l ld lp l0d l0k l0p h0 ld lp