421

PROTOTYPING CONTRIBUTIONS FOR PRODUCTION

MANAGEMENT IN CONSTRUCTION
Fernanda A. Saffaro1 , Marcel G. Trescastro2 , Alana A. Rodrigues3 , Fábio K. Schramm4 ,
Carlos T. Formoso5 and Luiz F. Heineck6

ABSTRACT
The uncertain nature of construction requires the use of tools that generate useful information
and support decision making in each phase of the product development process (PDP).
Experimentation techniques have been adopted by other industrial sectors with the aim of
supporting product development, especially due to the existing uncertainty and risk. However,
those techniques have been poorly used in construction, especially in the study of production
processes at building sites.
This paper describes three studies that were carried out in construction companies with the
aim of investigating the role of two experimentation techniques, FRS and on-line learning,
and also the requirements for their implementation.
The results suggest that these techniques can be used to understand work methods and to
establish standard operation elements. The production constraints did not allow the standard
work to be fully adopted and interrupted the continuity of the cycle observation–reflection–
action that could create in-depth knowledge on work methods. The paper discusses the
requirements for establishing an appropriate environment in order to increase the prototyping
contributions for creating an in-depth knowledge on work-methods.

KEY WORDS
Prototyping, first run study, standard-operation

1
M.Sc, Assistant Professor at the Civil Engineering Department, State University of Londrina. Ph.D. Candidate
at the Postgraduate Program in Civil Engineering, Federal University of Santa Catarina, Brazil. saffaro@uel.br
2
Architect, M.Sc., Building Innovation Research Unit - NORIE, Federal University of Rio Grande do Sul, Brazil.
3castro@via-rs.net
3
Civil Engineer, M.Sc. Candidate at the Building Innovation Research Unit - NORIE, Federal University of Rio
Grande do Sul, Brazil. alana.rodrigues@gmail.com
4
M.Sc., Assistant Professor at the Federal University of Pelotas, Ph.D. Candidate at the Building Innovation
Research Unit – NORIE, Federal University of Rio Grande do Sul, Brazil,, \o “Este link externo ir? abrir em
nova janela” \t “_blank” fabioks@ufpel.edu.br
5
Associate Professor, Building Innovation Research Unit – NORIE, Federal University of Rio Grande do Sul,
Brazil, formoso@ufrgs.br
6
Professor, Postgraduate Program in Production Engineering, Federal University of Santa Catarina, Brazil,
heineck@deps.ufsc.br
422 Fernanda A. Saffaro, Marcel G. Trescastro, Alana A. Rodrigues, Fábio K. Schramm,
Carlos T. Formoso and Luiz F. Heineck

INTRODUCTION
Experimentation techniques have been adopted by other industrial sectors with the aim of
supporting product development in a wide range of situations, such as product validation,
designing production operations, and performance evaluation. Those techniques play an
important role when there is much uncertainty in production process.
The uncertain nature of construction requires the use of tools that generate useful information
and support decision making in each phase of the product development process (PDP). In fact,
although the Last Planner‰ system seeks to improve the production flow reliability, assuring
the beginning of tasks only under favourable conditions, Howell and Ballard (1999) admit
that the lack of a better knowledge of work methods is an important source of uncertainty in
the production process. Aiming at increasing plan reliability, those authors suggest the use of
experimentation techniques such as First Run Studies (FRS).
A ‘First Run Study’ (FRS) consists of a detailed plan elaborated by a multifunctional team
composed of representatives of those who will perform the activity. This plan is put into
practice in the first run and followed by methodological study, redesign of the operation and
retrial until a standard is met or beaten (Ballard, 2000). FRS is a way of experimenting7 with
new work methods, similar to that used on prototyping in industrial sectors.
However, FRS and other prototyping techniques have been poorly used in construction for
designing operations and defining work standards. For instance, according to Kondo (1991),
work standards usually include three elements: objective of the work, constraints on carrying
out the work, and methods (means) to be employed in performing the work. It is not clearly
discussed in the literature on construction management how prototyping techniques contribute
to the knowledge of these elements and what is the detail level of the work standard established.
The aim of this paper is to discuss the role of prototyping techniques and how to implement
them for operations design in building projects. This research work was based on three empirical
studies, carried out in two different construction companies from the South of Brazil. The first
study explored the contributions and requirements of on-line learning in construction sites, while
the other two studies investigated the implementation of FRS in two different project environments.

FRS AND PROTOTYPING

Chew et al. (1991) suggest four methods to deal with uncertainty due to new technology
development and its implementation: (a) learning from others experience; (b) simulation; (c)
prototyping; and (d) on-line learning. Prototyping implies actually building and operating the
new technology on a small scale in a controlled environment. On line-learning consists of
examining the actual, full-scale technology while it is operating.
Clark et al. (1992) recognize that, even in the automobile manufacturing industry, uncertainty
persists at the production phase, as a result not only of the lack of knowledge on work methods,
but also of the large number of elements that interact unexpectedly. Those authors suggest that
experimentation techniques should be extended to the production phase. For instance, a pilot
run is a full-scale rehearsal of the commercial production system that is conducted first on
separate pilot lines and later on production assembly lines. A successful pilot run is followed
by ramp-up, the start-up of commercial production, which begins slowly and gradually
accelerates to full production. A pilot run is equivalent to what Chew et al. (1991) calls
prototyping and also to FRS proposed by Howell & Ballard (1999). On-line learning is
equivalent to ramp-up.

Proceedings IGLC-14, July 2006, Santiago, Chile

 Prototyping Contributions for Production Management in Construction 423

In this paper, prototyping is used in a broader sense, i.e. as a general methodology to

approximate the product under investigation, reducing uncertainty irrespective of its
development phase.
Other authors (Rosenthal & Tatikonda, 1992; Brown & Eisenhardt, 1995; Reinertsen, 1997;
Patterson, 1999 and Ulrich & Eppinger, 2000) suggest an early combined use of prototyping
resources. Virtual and rapid prototyping (RP) have been largely adopted to reduce cost and
duration of experimentation in the early phase of product design. Grimm (2004) describes RP
as a collection of technologies driven by computer-aided design (CAD) data that produce
physical models and parts through an additive process.
In construction it is not common to build prototypes, but, rather, the actual building is a
kind of prototype (Koskela, 2000). This means that the production stage is often used for
eliminating errors generated in the design or production planning processes. This peculiarity
of construction increases the importance of researching how prototyping techniques can
anticipate such errors and problems and what are the requirements for its successful adoption.

RESEARCH METHOD
Figure 1 summarizes the research design adopted in this research work, including research
questions, sources of evidence, and variables.
Three empirical studies were developed, over a period of six months. The first two were
carried out in Company A, involved in highly complex construction projects – typically
industrial, commercial and hospital projects in which the client facilities must operate normally.
The third study was carried out in Company B, in a much less complex environment compared
to the previous studies. That company was involved in the development and construction of
repetitive residential high-rise buildings. In the first two studies, action-research empirical
studies were carried out in the same building site, for introducing prototyping techniques in
Company A. In Company B, a descriptive case study was undertaken in one of its projects,
since prototyping techniques have been in use there for several years.
In the first study, the use of on-line learning was investigated in the assembly of a reinforced
concrete precast stair-case. The high daily cost of the crane hired to assemble the prefabricated
structure did not allow stoppages and attempts to study different work methods. Chew et al.
(1991) argues that the worse scenario happens when it is not possible to explore the other
three learning methods previously or simultaneously.
In the second and third empirical studies the possible contributions and the necessary
requirements for carrying out FRS were investigated. In both situations there was enough
time for building a prototype for a well defined repetitive unit.
The variables used in the analysis are related to the components of a standard (objectives,
constraints and work method). In this research work, method means a specific way of doing a
task, which comprises the work content (what operations are necessary), the method design,
the resources employed, the duration of the operations (production capacity) and the work in
progress (WIP). This definition is based on the necessary information for establishing work
specifications suggested by Currie (1977), and on the work-standard elements proposed by
Imai (1997), Monden (1998) and Liker (2004).

Production planning and control

424 Fernanda A. Saffaro, Marcel G. Trescastro, Alana A. Rodrigues, Fábio K. Schramm,
Carlos T. Formoso and Luiz F. Heineck

Figure 1 – Research design

RESULTS AND DISCUSSION

FIRST STUDY
Prior to the start of the activity, a fairly simple 3D model of the prefabricated staircase (Figure
2) was developed by one of the researchers. This model was used as a resource for discussing
the production process with the foreman who was in charge of the assembly process. At each
cycle of the on-line learning some additional data were collected and represented graphically
using operations design tools, such Standard Operation Sheets8 and Resource Balanced Charts.
The aim of the on-line learning experiment was to reduce the cycle time, due to the high
crane daily costs, and also to improve the safety conditions. Moreover, the need for reducing
work-in-progress was analyzed.
Despite the information available, discussing the production process was not easy, mostly
due to the fact that the foreman found it difficult to make his tacit knowledge explicit.
The exercise of virtually trying different method designs9 was fruitful because discussions
and externalisation of foreman knowledge were limited by real world problems. This detailed
plan was helpful to promote discussions regarding work methods and to improve knowledge
on how to do the job, even though it has not been followed in practice due to real world
constraints. It is important to highlight that these real world constraints were not concerned
with the work method, but to broader project management issues, such as contracts,
prefabricated plant operations, etc.

Proceedings IGLC-14, July 2006, Santiago, Chile

 Prototyping Contributions for Production Management in Construction 425

Figure 2 – 3D model of the precast staircase

Apart from these problems, on-line learning made it possible to understand the work content,
to identify work method constraints and to define basic directives to be followed. For example,
managers had not realized that the column-beam joint involved grouting after welding. The
foreman knew this requirement but was not aware that it was necessary to cure the grout for
12 hours before assembling stair flights. This information resulted in a basic directive: grouting
of SB7 and SB8 (only these two beams received the weight of the flights- see Figure 2) should
be done by the end of the day in order to allow curing overnight.
Due to adjustments that had to be made in SB8 at the prefabrication plant, there were
sometimes delays in the delivery of that beam, causing changes in the assembly sequence on
site. Although the standard was not followed, some improvement opportunities were identified,
such as changes in set up operations and in the design of methods (e.g. some operations could
be carried out simultaneously). It was also possible to reduce the cycle time (2 days for each
floor instead of 3 days – the initial prediction of the foreman). The foreman accepted this new
target (Figure 3), because there was an expectation that problems were solved. However, there
was a lack of involvement of production in the full implementation of the proposed
improvements. For that reason, some of them were not fully applied.
The positive results of this intervention stimulated the researchers to investigate the use of
FRS in the same building site, but in a different type of process. There idea was to develop an
earlier understanding of the work method and the sequence of operations by using FRS. It was
expected that FRS was a more adequate experimentation environment, by decreasing production
pressures and getting production managers’ commitment.

Production planning and control

426 Fernanda A. Saffaro, Marcel G. Trescastro, Alana A. Rodrigues, Fábio K. Schramm,
Carlos T. Formoso and Luiz F. Heineck

Figure 3 – Two days cycle time

SECOND STUDY
The FRS was carried out in the construction of a repetitive unit, which consisted of two hospital
apartments. There were 62 of these units in the building. FRS activities were scheduled well
in advance of the remaining sixty apartments. Each of those units involved the construction of
brickwalls and dry-walls, installation of building services, and finishings.
Although only two units were built, it was possible to: (a) identify activities that had not
been previously planned; (b) establish a new sequence of operations from the discussion of
alternative work methods; and (c) establish the production capacity for activities that were
well defined10 . FRS clearly contributed for developing a more in-depth knowledge of work
methods for these activities (work content, method design and production capacity). As the
schedule of the remaining sixty units was very tight, the prototyping exercise proposed the
elimination of some operations and the reduction of durations for others. Most proposed
improvements were successfully applied, but the standardized sequence defined in the FRS
was not fully adopted due to the lack of production constraint removal.
Two main factors increased the duration of the FRS, causing an overlap with the installation
process: (a) there were many design errors and problems that had not properly dealt with (e.g.
lack of space to fit the pipes in), and (b) delays in the delivery of materials. Such an overlap
caused much rework, since some design problems were only detected at the production of the
first apartment units.
Moreover, the development of the FRS caused a work overload for production managers
and its benefits could not be fully perceived by them. As a result, they could not get fully
involved in the decision-making process that should happen in a prototyping exercise. This
problem was partly caused by the fact that FRS activities were not included in the look-ahead
planning, nor in the short-term trouble-shooting list of activities, which could have helped
managers to perform tasks associated to FRS.

Proceedings IGLC-14, July 2006, Santiago, Chile

 Prototyping Contributions for Production Management in Construction 427

THIRD STUDY
The third study was carried out in a company that had been using successfully prototyping
techniques for improving production processes in a much less complex environment. This
company used FRS for producing one apartment unit well in advance in relation to the
production process. This prototype typically consisted of the construction of dry-wall, and
installation of building services.
Four main factors contributed to the success of FRS. First, the company had systematically
used prototyping in all projects. Second, the company kept a long term relationship with most
subcontractors, which made it easier to standardize the same managerial procedures involved
in FRS. Furthermore, partnerships were established with material suppliers who provided the
necessary resources for carrying out the FRS. Third, most design and production problems
had been solved either in the design process or in previous projects. Lastly, the quality control
at the initial construction phases (structure and external walls) was very effective, limiting to
a great extent possible deviations in the main dimensions of the apartments. For that reason,
prototyping was focused on the validation and fine adjustment of the product, using a detail
design and existing work standards as references. For instance, one of the FRS main objectives
was to define solutions for solving tolerance problems in wet areas (i.e. bathroom, kitchen,
laundry), making sure that the connections between finishing elements (for example, wall
tiling, sink and WC basin installation) are properly done.
Therefore, compared to previous studies, prototyping in this environment was a much
simpler and more focussed activity. This type of prototyping fits Ulrich & Eppinger’s (2000)
definition: “any entity exhibiting at least one aspect of the product that is of interest of the
development team can be viewed as a prototype”. In fact, Patterson (1999) stresses that prototype
construction should focus on areas where outcomes are uncertain. In this study, this means
that if the dimensions of wet areas were correct, most probably all succeeding finishing activities
would be straightforward as they would exactly follow what was designed.
The presence of the production manager was demanded only when a major problem was
faced. Besides that, there was a self-organizing cellular like team which avoided the inclusion
of FRS activities in the lookahead planning. FRS activities were conducted autonomously,
apart from the hierarchical site chain of command.
This third study highlighted a successful application of FRS which yielded the necessary
information in connection to product design validation. It might be argued that the FRS
objectives were restricted and hence its success, but it should be remembered that this particular
study was made possible due to a prototyping culture well established in the company.

CONCLUSIONS
The main contribution of this research work was related to the role of experimenting techniques
(FRS and on-line learning) and the requirements for their implementation. Table 1 summarizes
the main findings of each case study.
The most important contribution of the two prototyping techniques investigated was to
understand work methods and to establish standard operation elements. The lack of produbúion
constraint removal did not allow the standard work to be fully adopted, and interrupted the
continuity of the cycle observation–reflection–action that could create in-depth knowledge on
work methods. It is necessary to establish an appropriate environment for the full application
of work-standards in the real world. In that case, it is possible that a new set of constraints

Production planning and control

428 Fernanda A. Saffaro, Marcel G. Trescastro, Alana A. Rodrigues, Fábio K. Schramm,
Carlos T. Formoso and Luiz F. Heineck

concerning work methods might arise and consequently new directives would be defined. In
this context a hypothesis must be investigated, concerning the capacity to deal with prototyping
issues in a dynamic environment without the need to establishing detailed standards, but just
basic directives.
st st st
Research 1 Study 2 Study 3 Study
Findings Prefabricated stair Hospital Apartments Residential Apartments
Contributions - Work content - Work content - Validation of the product
- Sequence of the - Sequence of the built, assuring design
operations operations dimensions
- Capacity production - Capacity production
- Set up operations - Method designs
- Method designs*
- WIP*

Requirements - Design problems solved - Reduce manager work - Quality control limits the
- Managers committed overload – include FRS dimensions variations–
and in grounded on site on lookahead planning replication of solutions set
- Simultaneous use of up by FRS
prototyping techniques - FRS simplification –focus
on uncertainties only
- Partnership to manage
resources supply

Table 1 - Summary of research findings

This appropriate and dynamic environment should result not only from the exploration of
other prototyping techniques but also from the use of practices that support the prototyping
process, as suggested by Ulrich & Eppinger (2000), such as cooperation among the
organizations, enhanced communication, and learning activities. Those practices, if connected
to the prototyping process, should mitigate the work overload caused by the realization of the
prototype and increase the benefits that follow from its undertaking.
Regarding the application of other prototyping techniques, virtual models must be used for
eliminating design uncertainties and errors (e.g. poorly defined column-beam joint, lack of space
for fitting elements), removing product related problems that do not allow the prototyping exercise
to focus on work-methods and standards. Also, RP could have been an effective way to make
explicit the foreman’s knowledge that was essentially tacit. It would allow the exploration of
vicarious learning (learning from others’ experience) which is the most basic learning method
suggested by Chew et al (1991). This study has indicated that even virtual models for trying
different work methods are useful because they provide opportunities for what-if discussions
that make explicit the knowledge related to real world constraints.

REFERENCES
Ballard, G. (2000). The Last Planner System of Production Control, PhD Thesis, School of
Civil engineering, Faculty of Engineering, The University of Birmingham.

Proceedings IGLC-14, July 2006, Santiago, Chile

 Prototyping Contributions for Production Management in Construction 429

Bertelsen, S.; Koskela, L. (2004). “Construction beyond lean: a new understanding of

construction management.” Proceedings of 12th annual conference of the international group
for lean construction, Elsinore, Denmark.
Brown, S.L.; Eisenhardt, K.M. (1995). “Product development: past research, present findings
and future directions.” Academy of Management Review, 20 (2), p.343-378.
Chew, W.B.; Leonard-Barton, D.; Bohn, R.E. (1991). “Beating Murphy’s law”. Sloan
Management Review, 32 (3), p. 5-16.
Clark, K.B., Chew, W.B., Fujimoto, T. (1992). “Manufacturing for design: beyond the
production/R&D dichotomy.” In: Integrating Design and Manufacturing for Competitive
Advantage. Edited by Susman, Gerald I. Oxford University Press, Inc.
Currie, R. M. (1997). “Work study”. Fourth edition. Great Britain: A Pitman International
Text, 326p.
Grimm, T. (2004). User’s guide to rapid prototyping. Society of Manufacturing Engineers,
404p.
Howell, G; Ballard, G. (1997). “Factors affecting project success in the piping function”.
Edited by Alarcon. Lean Construction. Rotterdam: Balkema. p.161-185.
Howell, G. Ballard, G. (1999). “Design of construction operations.” White Paper 04. [S.l.]:
Lean Construction Institute.
Imai, M. (1997). Gemba Kaizen: a commonsense, low-cost approach to management. McGraw-
Hill, 343p.
Johnston, R.B. & Brennan, M. (1996). “Planning or organizing: the implications of theories of
activity for management of operations.” Omega, Int. J. Mgmt. Sc., 24 (4), 367-384.
Kondo, Y. (1991). Human motivation: a key factor for management. 3A Coorporation,186p.
Koskela, L. (2000). An exploration towards a production theory and its application to
construction. PhD Thesis, VTT Publications 408. VTT Technical Research Centre of Finland,
Espoo.
Koskela, L. (2001). “On new footnotes to Shingo.” Proceedings of 9th annual conference of
the international group for lean construction, Singapore.
Koskela, L.; Howell, G. (2002). “The theory of project management: explanation to novel
methods.” Proceedings of 10th annual conference of the international group for lean
construction, Gramado, Brazil.
Liker, J. K. (2004). The Toyota Way: 14 Management Principles from the World’s Greatest
Manufacturer. McGraw-Hill, 352p.
Monden, Y. (1998). Toyota Production System: an integrated approach to just-in-time. 3rd
Edition. Industrial Engineering and Management Press.
Patterson, M.L. (1999). Leading product innovation: accelerating growth in a product-based
business. USA: John Wiley & Sons, Inc., 413p.
Reinertsen, D. (1997). Managing the Design Factory: A Product Developer’s Toolkit. New
York: Free Press, 269p.
Rosenthal, S.R. & Tatikonda, M. V. (1992) “Competitive advantage through design tools and
practices.” In: Integrating Design and Manufacturing for Competitive Advantage. Edited by
Susman, Gerald I. Oxford University Press, Inc.
Shingo, S. (1996). Sistemas de produção com estoque zero: o sistema Shingo para melhoria
contínua. Porto Alegre: Bookman, 380p.
Ulrich, K.T. & Eppinger, S.D. (2000). Product design and development. Second edition, London:
McGraw-Hill, 358p.

Production planning and control