INVESTIGATION OF INTERPERSONAL COOPERATION IN CONSTRUCTION
PROJECT TEAMS: AN AGENT-BASED MODELING APPROACH
By
Jing Du

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILSOPHY
Construction Management
2012

ABSTRACT
INVESTIGATION OF INTERPERSONAL COOPERATION IN CONSTRUCTION PROJECT
TEAMS: AN AGENT-BASED MODELING APPROACH
By
Jing Du
Construction project teams have been criticized to be incapable of delivering projects in a
predictable and reliable manner. A major issue is contemporary construction practice doesn’t
encourage a closer cooperation among project participants. Hence as addressed by a variety of
industry wide reports, the real challenge of the construction industry is to move from the current
modus operandi to a more collaborative approach. To achieve this goal, knowledge should be
advanced to fully capture the processes of cooperation in construction settings. This dissertation
is intended to provide an innovative insight into the dynamics of cooperation in construction that
we couldn’t access before.
The basic hypothesis of this dissertation is that cooperation at the individual level among
construction project team members can significantly affect team performance. In particular,
diverse cooperative behaviors of project participants and corresponding micro-level processes
may determine project team performance in certain scenarios. To test this hypothesis, a
conceptual framework was developed to capture relevant components of interpersonal
cooperation; then an ABM (Agent-Based Modeling) simulation platform, named VOICE
(Virtual Organizational Imitation for Construction Enterprises) was developed for exploratory
simulation experiments. VOICE allows the investigators to design an artificial construction
organization just like engineers designing a bridge. It is also able to visualize project team
performance by real time performance dashboards and 3D graphs.

Two case studies were conducted using VOICE to illustrate its capacity in investigating
realistic cooperation related issues in construction. The first case study investigated intra-team
cooperation in an estimating team. Specifically, certain cooperation related management actions
-- job acquiring, coordination, and team communication -- were examined to check their
influences on the estimating team performance in DBB (Design Bid Build) projects. The second
case study extended the investigation to a cross-functional context where conditions become
more complex. Simulation experiments were performed based on a real case between a proposal
team and an engineering team in an EPC (Engineering Procurement and Construction)
construction enterprise. The effects of goal congruence, as well as the implications of time
pressure, task dependence and micro-management were examined based on a series of
uncertainty analyses.
The findings have supported the basic hypothesis of this research that cooperation at the
individual level plays a vital role in affecting construction project team performance. It was also
found the investigation of cooperation in construction settings to be nontrivial because of the
complex interactions of cooperative behaviors and processes. This research is anticipated to
constitute a stepping stone for further investigations of cooperation in construction: for
academics, it is expected that the simulation approach and findings of this research would attract
more scholars’ attention to the behavioral and attitudinal aspects of cooperation in construction
settings; for industry, the simulation platform VOICE could be used as a decision support tool
for decision makers of project management to tackle with cooperation related issues.

Copyright By
JING DU
2012

To my father

v

ACKNOWLEDGMENTS

First and Foremost, I would like to sincerely thank my major professor, Dr. Mohamed El-Gafy,
for his guidance, support, patience, and friendship during my graduate studies. He has guided
me throughout my research with his patience and knowledge over the past several years whilst
giving me the freedom to work in my own way. I appreciate all his contributions of time, ideas,
and funding to make my PhD experience productive and inspiring. I will take him as my lifelong friend and collaborator.
I am also deeply thankful to the rest of my PhD committee: Dr. Arika Ligmann-Zielinska,
Dr. Matt Syal and Dr. Tariq Abdelhamid. I have benefitted tremendously from their insightful
comments and constructive advice at different stages of my research. Their unconditional
support was fundamental to the writing of this dissertation and my professional and academic
experience at Michigan State University.
I am also indebted to a number of people in the School of Planning, Design and
Construction: Dr. Scott Witter, Dr. Mark Wilson, Professor Tim Mrozowski, Dr. Sinem
Korkmaz, Dr. Armagan Korkmaz, Valerie Geyer, Judy Pardee, Cathy Morrison, Lauri Stephens,
William Balluff, Pat Daughenbaugh, Mary Anne Merrill and many others. Without their
unreserved support and patience, I couldn’t make it.
Last, I am enormously grateful to my beloved wife, for being there and loving me. I am
so fortunate to have her.

vi

TABLE OF CONTENTS
LIST OF TABLES ……………………………………………………………………...x
LIST OF FIGURES………………………………………………………………….…xii
LIST OF ABBREVIATIONS………………………………………………………...xv
CHAPTER 1: INTRODUCTION .................................................................................... 1
1.1. Background ........................................................................................................... 1
1.2. Problem statement ................................................................................................. 3
1.3. Research goal and objectives ................................................................................ 7
1.4. Method .................................................................................................................. 9
1.5. Dissertation scope ............................................................................................... 14
1.6. Deliverables of this research ............................................................................... 14
1.7. Outline of this dissertation .................................................................................. 15
CHAPTER 2: LITERATURE REVIEW ...................................................................... 18
2.1. Cooperation theory .............................................................................................. 18
2.1.1. Definition of Cooperation in construction ................................................... 18
2.1.2. Categorization of cooperation ...................................................................... 22
2.1.3. Implications of cooperation ......................................................................... 23
2.1.4. Previous cooperation studies in construction area ....................................... 25
2.2. Construction project teams .................................................................................. 32
2.2.1. Definition of construction project team ....................................................... 32
2.2.2. The characteristics of construction project teams ........................................ 34
2.2.3. Cooperation research suitable for the features of construction project team: from
“top-down” to “bottom-up” ........................................................................................ 38
2.2.4. Performance of construction project teams.................................................. 42
2.3. Agent Based Modeling: a bottom-up approach................................................... 45
2.3.1. Concepts ....................................................................................................... 45
2.3.2. Applications of ABM ................................................................................... 48
2.3.3. The rationale for using ABM in this research .............................................. 51
2.4. Summary ............................................................................................................. 54
CHAPTER 3: A CONCEPTUAL FRAMEWORK OF INTERPERSONAL
COOPERATION IN CONSTRUCTION PROJECT TEAMS ................................... 55
3.1. Cooperation: a system perspective ...................................................................... 55
3.2. Antecedents of interpersonal cooperation in construction project teams ............ 58
3.2.1. Task characteristics ...................................................................................... 61
3.2.2. Individual characteristics ............................................................................. 68
3.2.3. Relational behaviors..................................................................................... 74
3.2.4. Organizational context ................................................................................. 86
3.3. Performance indicators of construction project teams ........................................ 91
3.3.1. Outcome performance .................................................................................. 91
3.3.2. Process performance .................................................................................... 93
vii

3.4.
3.5.

The conceptual framework .................................................................................. 95
Summary ............................................................................................................. 99

CHAPTER 4: A BOTTOM-UP SIMULATION MODEL OF INTERPERSONAL
COOPERATION IN CONSTRUCTION PROJECT TEAMS ................................. 100
4.1. Model architecture............................................................................................. 100
4.2. Input modeling .................................................................................................. 101
4.3. Process modeling............................................................................................... 106
4.3.1. Modeling individuals ................................................................................. 106
4.3.2. Modeling relational behaviors ................................................................... 110
4.3.3. Modeling Organizational context .............................................................. 122
4.4. Output measurement ......................................................................................... 129
4.4.1. Time ........................................................................................................... 129
4.4.2. Quality........................................................................................................ 130
4.4.3. Effectiveness .............................................................................................. 130
4.4.4. Efficiency ................................................................................................... 131
4.4.5. Work related pressure ................................................................................ 131
4.5. Basic assumptions of VOICE ............................................................................ 131
4.6. Developing an integral platform ....................................................................... 134
4.7. Summary ........................................................................................................... 135
CHAPTER 5: MODEL VERIFICATION AND VALIDATION ............................. 136
5.1. Verification........................................................................................................ 136
5.1.1. Code debugging ......................................................................................... 136
5.1.2. Logic examination ..................................................................................... 136
5.1.3. Unit test ...................................................................................................... 138
5.1.4. Hypothetical case test ................................................................................ 140
5.2. Validation .......................................................................................................... 143
5.2.1. Input validation .......................................................................................... 144
5.2.2. Output validation ....................................................................................... 151
5.3. Summary ........................................................................................................... 157
CHAPTER 6: CASE STUDIES ................................................................................... 159
6.1. Case Study 1: intra-team cooperation in estimating activities .......................... 159
6.1.1. Case overview ............................................................................................ 159
6.1.2. Model assumptions .................................................................................... 161
6.1.3. Model development ................................................................................... 162
6.1.4. Simulation results....................................................................................... 165
6.1.5. Interpretation of simulation results ............................................................ 172
6.2. Case study 2: Inter-team cooperation in project proposal development ........... 174
6.2.1. Case overview ............................................................................................ 175
6.2.2. Model assumptions .................................................................................... 177
6.2.3. Model development ................................................................................... 181
6.2.4. Simulation results....................................................................................... 184
6.2.5. Interpretation of simulation results ............................................................ 201
6.3. Discussion ......................................................................................................... 204
viii

6.3.1. General findings of simulation studies ...................................................... 204
6.3.2. Operational recommendations ................................................................... 206
6.4. Summary ........................................................................................................... 208
CHAPTER 7: CONCLUSIONS .................................................................................. 209
7.1. Review of the research ...................................................................................... 209
7.2. Contributions of this research ........................................................................... 212
7.3. Potential limitations of this research ................................................................. 216
7.4. Future agenda .................................................................................................... 218
APPENDICES …………………………………………………………………………..221
APPENDIX 1: INFORMATION ABOUT CASE STUDY 1 ...................................... 222
APPENDIX 2: INFORMATION ABOUT CASE STUDY 2 ...................................... 232
APPENDIX 3: SAMPLE CODES .............................................................................. 254
REFERENCES ............................................................................................................... 282

ix

LIST OF TABLES

Table 2.1 Definitions of Cooperation…………………………………………………………... 21
Table 2.2 Quadrants of cooperation enhancement practices and studies …………………….....32
Table 2.3 Five types of teams; modified from (Katzenbach and Smith 1993) ………………....36
Table 2.4 The dimensions of team performance………………………………………………...43
Table 2.5 The applications of ABM………………………………….………………………....49
Table 3.1 Antecedents and determinant factors of interpersonal cooperation…………………..58
Table 4.1 Attributes of projects…………………………………………………………….....103
Table 4.2 Attributes of tasks………………………………………………………………….103
Table 4.3 A sample table head of the online questionnaire…………………………………..106
Table 4.4 Attributes of actors………………………………………………………………...108
Table 4.5 Relationship between tasks and relational behaviors: an example………………….110
Table 4.6 Set of fundamental relational behaviors and specific actions……………………….112
Table 4.7 A sample survey showing individual responsibilities……………………………….125
Table 4.8 Task-Actor Relation Table (TART)………………………………………………...126
Table 4.9 Basic assumptions of VOICE…………………………………………...…………...132
Table 5.1 Test markers and check points of functional units of VOICE……………………...138
Table 5.2 Values of attributes used in the hypothetical cases…………………………………141
Table 5.3 An example of time sheet…………………………………………………………...146
Table 6.1 Fitted PDFs of key variables of individuals and proposal development task……….181
Table 6.2 ANOVA tests under different levels of goal congruence (efficiency)……………....185
Table 6.3 ANOVA tests under different levels of goal congruence (effectiveness)…………...188

x

Table 6.4 ANOVA tests under different levels of goal congruence (quality)………………….190
Table 6.5 ANOVA tests under different levels of goal congruence.…………………………..191
Table 6.6 p-values of micro-management’s influence under levels of goal congruence………198
Table 6.7 p-values of micro-management’s influence under levels of task dependence………201
Table A.1 Task parameters of Project 1 (size=117)……………………………………………229
Table A.2 Task parameters of Project 2 (size=115)……………………………………………230
Table A.3 Task parameters of Project 3 (size=137)……………………………………………230

xi

LIST OF FIGURES

Figure 1.1 Method of this research…………………………...…………………………………10
Figure 2.1 Categories of literature review………………………………………………………18
Figure 2.2 Two approaches to enhancing cooperation in construction project teams………….41
Figure 2.3 The framework of construction project team performance………………………….45
Figure 2.4 The framework of an agent…………………………………………………………47
Figure 3.1 Model of organizational behavior……………………………………………….…..56
Figure 3.2 Three types of task dependence………………………………………...……………64
Figure 3.3 Conceptual framework of interpersonal cooperation in construction project teams..96
Figure 3.4 An IPO view of the framework……………………………………………………...98
Figure 4.1 Model Architecture of VOICE……………………………………………………..101
Figure 4.2 Flowchart of the agent project……………………………………………………105
Figure 4.3 Using Ishikawa diagram to develop the list of activities…………………………106
Figure 4.4 Flowchart of actors’ behaviors (example)………………………………………….122
Figure 4.5 An example organizational chart of a project proposal team………………………123
Figure 4.6 Task delivery pattern in VOICE……………………………………………………127
Figure 4.7 Information flow pattern in VOICE……………………………………………..…128
Figure 4.8 The runtime interface…………………………………………………………….…135
Figure 5.1 Logic errors detection by flowchart examination: example of prioritizing tasks….137
Figure 5.2 An illustrative project team used in model verification……………………………140
Figure 5.3 Simulation results of the first hypothetical case (examples)……………………….142
Figure 5.4 Simulation results of the second hypothetical case (examples)……………………142

xii

Figure 5.5 Surveys to collect task information………………………………………………...146
Figure 5.6 Distribution of task duration………………………………………………………..147
Figure 5.7 Behavioral reactions to exceptions…………………………………………………151
Figure 5.8 Distribution of work duration of the representative scenario………………………153
Figure 5.9 Distribution of mistake percentage of the representative scenario…………………155
Figure 5.10 Distribution of effectiveness of the representative scenario…………………156
Figure 5.11 Distribution of number of iterations of the representative scenario………………157
Figure 6.1 Modeling case 1 with proposed theoretical framework …………………………...162
Figure 6.2 Snapshot of the simulation (case study 1)………………………………………….164
Figure 6.3 Real time bidding work progress (scenario 1)……………………………………..166
Figure 6.4 Real time bidding work progress (scenario 2)…………………………………..…167
Figure 6.5 Work related stress of helpers (scenario 2)………………………………………..168
Figure 6.6 Real time bidding work progress (scenario 3)…………………………………….170
Figure 6.7 Accumulative effectiveness of bidding work (scenario 3)……………………...…171
Figure 6.8 Modeling case 2 with the proposed theoretical framework……………………..…180
Figure 6.9 Snapshot of the simulation (case study 2)………………………………………….183
Figure 6.10 Influence of time pressure on efficiency of proposal development………………186
Figure 6.11 Influence of time pressure on effectiveness of proposal development……………187
Figure 6.12 Influence of time pressure on quality of proposal development………………….189
Figure 6.13 Influence of time pressure on averaged work pressure………………………...…191
Figure 6.14 Influence of goal congruence and task dependence………………………………193
Figure 6.15 Influence of task dependence on changes to efficiency under different levels of goal
congruence………………………………………………………………………194
Figure 6.16 Influence of goal congruence and micro-management……………………………195
Figure 6.17 Influence of goal congruence and micro-management……………………………199
xiii

Figure A.1 Organizational chart of case study 1……………………………………………….231
Figure A.2 Work sampling result of the face-to-face communication in a month………….…241
Figure A.3 Fitted PDFs for input data………………………………………………………….241
Figure A.4 Organizational Chart of company Z……………………………………………….243
Figure A.5 Influences of task dependence and goal congruence on performance of proposal
development…………………………………………………………………….244
Figure A.6 Additional influences of task dependence on the effects of goal congruence…..…247
Figure A.7 Influences of micro-management preference on team performance under levels of
task dependence……………………………………………………………………248
Figure A.8 Influences of micro-management preference on team performance under levels of
task dependence……………………………………………………………………251

xiv

LIST OF ABBREVIATIONS

ABM

Agent Based Modeling

AIA

American Institute of Architects

AON

Activity on node

ANOVA

ANalysis Of VAriance

BIM

Building Information Modeling

CII

Construction Industry Institute

DBB

Design Bid Build

EPC

Engineering Procurement Construction

IFoA

Integrated Form of Agreement for Lean Project Delivery

IPD

Integrated Project Delivery

IPO

Input Process Output

NASFA

National Association of State Facilities Administrators

OOP

Object-Oriented Programming

PBO

Project Based Organization

PDF

Probability Distribution Function

Repast

REcursive Porous Agent Simulation Toolkit

SD

System Dynamics

TART

Task-Actor Relation Table

VOICE

Virtual Organizational Imitation for Construction Enterprises

WBS

Work Breakdown Structure

xv

CHAPTER 1: INTRODUCTION
1.1. Background
The construction industry has been suffering serious performance problems, including
dropping productivity (Allen 1985; Teicholz et al. 2001; Rojas and Aramvareekul 2003),
time and cost overruns (Egan 1998b; Okuwoga 1998; Flyvbjerg et al. 2003; Bordat et al.
2004), frequent change orders and claims (Ren et al. 2001), and high business failure rate
(Kangari 1988; Knaup and Piazza 2007). Construction project teams are criticized as not
fully capable to deliver projects to the client in a reliable and predictable manner (Egan
2002). A widely accepted explanation of these performance issues is that project
participants are unable to work together effectively (Arditi et al. 2000; Cheung et al. 2003;
Phua and Rowlinson 2004; Anvuur and Kumaraswamy 2007). Contemporary construction
system and project management approaches don’t encourage effective cooperation
between project participants (Cheung et al. 2003; Chen and Tien 2007; Kang et al. 2007;
Son and Rojas 2010). Hence the real challenge of the industry is to move from the current
modus operandi towards collaborative approaches (Latham 1994; Egan 1998; Bourn 2001;
Egan 2002; Kang et al. 2007). Achieving this goal requires deeper insight into the nature
and implications of cooperation in the construction context.

Any construction project is a temporary coalition of a number of groups (Shirazi et al.
1996; Dubois and Gadde 2002; Bouchlaghem et al. 2004). The project team members
hold divergent interests and foci (Winch 2003; Bouchlaghem et al. 2004; Ruikar et al.
2007). Their success is defined by achieving performance metrics for the individual
1

organization instead of for the collective project performance (Cornick and Mather 1999).
This situation creates a “fragmented” atmosphere in construction project teams (Baiden et
al. 2006), which in turn leads to a “blame culture” where team members seek to minimize
their responsibility for poor performance, rather than working together towards
cooperation (Baiden et al. 2006). It is thus difficult for a project team to promote the
progress of the work through concerted actions and implementation of decisions
(Hackman 1990; Devine et al. 1999; Grant 2007). As highlighted by Phua and Rowlinson
(2004), cooperation between employees in construction project teams remains a serious
issue that needs attention.

Building on surveys and interview studies, the benefits of improved cooperation in the
construction context have been well recognized -- Better cooperation can break down
barriers (Albanese 1994), foster innovation (Dulaimi et al. 2002; Dulaimi et al. 2003),
exert positive implications for project performance, creativity and learning (Dubois and
Gadde 2002; Dorée and Holmen 2004; Phua and Rowlinson 2004) and, promote
psychosocial outcomes for employees (Pinto et al. 1993b).

In recognition of the importance of cooperation in construction, a number of “Integration
Toolkits” have been proposed and applied in the industry to enhance the cooperation
among team members, such as concurrent engineering (Love et al. 1998), partnering
(Bresnen and Marshall 2000b), project strategic alliance (Li et al. 2001), integrated project
delivery (Eckblad et al. 2007), and also a great interest in collaborative tools and

2

techniques, such as web based project management (Nitithamyong and Skibniewski 2004)
and building information modeling (Azhar et al. 2008).

Despite the distinctness of these approaches, they are all ultimately intended to remove
the obstacles between project participants, increase the transparency, enhance the level of
trust, and create an integrated and collaborative project team effort for project delivery
(Thomas and Thomas 2005; Gudgel 2008; NASFA et al. 2010).

1.2. Problem statement
Despite the attempts to enhance cooperation in the construction industry, there is still a
lack of precise knowledge of the nature and implications of cooperation in construction
project teams, keeping cooperation research in construction somewhat of a “black box”
(Anvuur 2008). Fundamental questions about cooperation in construction project teams
remain unanswered, such as: what factors facilitate/hinder the cooperation in construction
project teams? To what extent do they do so? Are the effects linear (as described in
certain studies, e.g., Cheung 2003; Phua and Rowlinson 2004; Anvuur 2008), or nonlinear
(as described by Thomsen 2005)? Bresnen and Marshall (2000b) asserted that any
contribution leading to a better understanding of the cooperation concept that
unifies/integrates the multiple theoretical perspectives in construction project teams would
be a significant contribution to knowledge. It should be noted that developing a precise
knowledge of the cooperation in construction is a difficult (if not impossible) task, which
requires long-term research efforts by the academia. This dissertation only moves one step

3

closer to provide a unique insight into this theme. Specifically, it addresses four major
theoretical gaps in current cooperation studies in construction.

First, there is a lack of efforts by construction academics and industry to investigate
cooperation at the individual level in construction project teams, despite its evident
importance for successful project delivery (Fisher and Green 2001). A survey reported by
Phua and Rowlinson (2004) revealed that “cooperation between the colleagues in a
construction firm” exerts a much stronger influence (coefficient=0.22) on the construction
project

performance

than

does

“cooperation

between

construction

firms”

(coefficient=0.02). In other words, interpersonal cooperation is the fundamental element
of cooperation in construction project teams. However, the construction literature has paid
too much attention to the cooperation issues between project organizations, and tended to
ignore the need to differentiate cooperation at an individual level from that at an
organizational level (Eckblad et al. 2007). Most construction literature simply refers to
“cooperation of construction project teams” as “inter-organizational integration” (Cheng
et al. 2001; Lam et al. 2001; Welling and Kamann 2001; WL 2004), or as part of
“partnering between construction firms” (Bresnen and Marshall 2000b; Wood et al. 2002;
Naoum 2003). Only a few research efforts in construction have highlighted the need to
investigate individual or interpersonal cooperation and their arguments haven been
conflicting (Phua and Rowlinson 2004; Anvuur and Kumaraswamy 2007; Anvuur 2008;
Son and Rojas 2010). Thus investigating individual-level cooperation in construction

4

project teams is needed to augment the body of knowledge regarding cooperation in the
construction industry.

Second, there is an urgent need to investigate the behavioral and attitudinal aspects of
cooperation in construction project teams, considering the insufficiency of industrial
emphasis on searching for contractual, strategic and technical remedies (Son and Rojas
2010). Attempts to foster cooperation in construction project teams have mainly focused
on improving product delivery processes and strategic alliance to ensure the cooperation
among different disciplines in construction projects (Cheng et al. 2001; Lam et al. 2001;
Welling and Kamann 2001; WL 2004; Baiden et al. 2006). These attempts have not
completely reached the anticipated success. A likely explanation is “they are frequently
superimposed onto environments where adversarial cultures and attitudes still exist”
(Moore and Dainty 2001). Project team members are mostly affected by behavioral, social
and cultural factors rather than simply the institutional and contractual factors (Son and
Rojas 2010). This highlights the importance of looking into the psychosocial perspectives
of cooperation in construction project teams in order to address the cooperation issues in
the construction industry (Anvuur 2008).

Third, the evidence of relationship between cooperation and construction project team
performance has remained anecdotal, subjective and piecemeal, which undermines the
industrial support on enhancing cooperation for better performance (Anvuur 2008).
Existing literature findings (Cheng et al. 2001; Lam et al. 2001; Welling and Kamann

5

2001) are usually based on regional surveys which occasionally contradict with each other
(Phua and Rowlinson 2004). Although research on organizations in other industrial
settings demonstrates a strong positive link between individuals’ cooperative behaviors
and organizational/team effectiveness (Ruekert and Walker Jr 1987a; Souder 1988; Hise
et al. 1990; Motowidlo and Van Scotter 1994; Pfeffer 1994; Podsakoff and MacKenzie
1997; Hoegl et al. 2004; Podsakoff et al. 2009), evidence in the construction industry is
still lacking (Phua and Rowlinson 2004). This situation undermines the industrial
understanding of and support for enhancing cooperation for dealing with performance
issues in the construction industry. It’s worth noting that, solid evidence of the
relationship between cooperation and construction project teams requires long term
investigation with a comprehensive body of samples. This dissertation only aims to
provide a systematic way (and probably an easier way) for such investigations.

Last, existing research efforts aim to provide a remedy for cooperation issues in the
construction industry through “top-down” changes such as partnering (Malone and
Crowston 1990; Bresnen and Marshall 2000b; Wood et al. 2002; Naoum 2003). Such a
“top-down” approach has proven insufficient to capture the real needs of enhancing
cooperation in construction settings (Moore and Dainty 2001). Based on an analysis of
the natural features of construction project teams, this study proposes that a “bottom-up”
approach, i.e., emphasizing the transition from a local diversity of cooperation processes
and behaviors to global performance of project teams, is a good fit for cooperation
research in the construction context. In fact, certain cooperation enhancement approaches
6

in construction, such as Target Value Design (TVD), are effective because they allow
bottom-up changes. Seemingly a top-down approach, TVD actually works by encouraging
the cooperation among project participants in a bottom-up manner (Macomber et al. 2007).
Therefore, it is expected that through enhancing interpersonal cooperation, especially by
behavioral changes of team members, insight into cooperation issues in construction
project teams will be advanced.

In a short, the limitations addressed in this dissertation can be grouped into four clusters:
(1) A lack of effort to investigate cooperation at the individual level in construction; (2) A
lack of attention to the behavioral and attitudinal aspects of cooperation in construction; (3)
A lack of a systematic approach to investigate the relationship between cooperation and
construction project team performance; and (4) A lack of bottom-up research methods to
capture the complexity of the studied theme. Based on these problems, the goal and
objectives of this research are stated.

1.3. Research goal and objectives
The main goal of this research is to investigate interpersonal cooperation in construction
project teams, with a focus on the transition from local dynamics of individual behaviors
and micro cooperation processes to global performance of project teams. Specifically, the
main aim is to develop a conceptual framework and simulation model that allows a
rigorous theory-based investigation of the micro-level behavioral and attitudinal factors
that shape local interpersonal cooperation processes in construction project teams, and

7

their impacts on team performance. This will be achieved through the following specific
objectives:

A. To create a conceptual framework for investigating interpersonal
cooperation in construction project teams
Questions: What is the definition of interpersonal cooperation? What are the
constructs of interpersonal cooperation? What are the implications of interpersonal
cooperation? What are the measurements of construction project team performance?
What are the lessons learned from previous studies?

Significance: Establish a rational and solid conceptual foundation for the current
study.

B. To develop a bottom-up simulation model for investigating interpersonal
cooperation in construction project teams
Questions: What aspects and factors should be included in the proposed simulation
model? How should interpersonal cooperation be represented in the conceptual
model’s architecture? What is the level of detail for such representation? How can
team performance indicators be formulated? How will the model’s development be
realized? How will the model be verified and validated?

Significance: A comprehensive, concise, accurate, and verified simulation model
forms the basis for further analysis.

8

C. To investigate interpersonal cooperation issues in real world construction
project teams
Questions: What are the representative cooperation related issues in real world
construction project teams? How can the conceptual framework and simulation model
be applied to investigate these issues? How can the simulation model be tailored for
specific scenarios? Can recommendations be given to improve construction project
team performance through exploratory simulations based on the simulation model?

Significance: Experiments based on exploratory simulations constitute the basis for
creating transformative knowledge of interpersonal cooperation in construction
settings.

1.4. Method
Figure 1.1 demonstrates the specific tasks undertaken in this research to achieve its three
objectives. This section briefly describes these tasks as the method of this research.

9

Figure 1.1 Method of this research

Task 1: Reviewing literature and proposing a conceptual framework

This step identified the gaps in current studies, framed foci in this research, and built a
solid conceptual framework for interpersonal cooperation in construction. Three subtasks
were conducted to form the point of departure for the entire research: (1) To identify the
gaps of current literature, and conceptualize the research problems about interpersonal
10

cooperation in construction project teams that will be investigated in this research; (2) To
survey a comprehensive body of literature related to the object of study and theoretical
base, including the definition and natural features of construction project teams,
cooperation theory (definition, constructs and influences), team performance (definition
and measurements), and ABM with its concepts, features, applications and rationale in the
proposed study; and (3) To propose a conceptual framework based on the comprehensive
literature review for investigating interpersonal cooperation and team performance in
construction project teams. The proposed framework has demonstrated the conceptual
connection between the components of interpersonal cooperation and the measurements
of team performance and thus constituted the foundation of the simulation model.

Task 2: Conceptualizing model architecture and formulating performance indicators
of construction project teams

Conceptual model architecture served as a transition from a conceptual framework to an
executable simulation model in this research. This step developed a model architecture to
represent interpersonal cooperation in construction project teams. In particular,
interpersonal cooperation was represented as a set of attributes and model rules by the
following steps: (1) model variables of interpersonal cooperation: variables were defined
as the attributes, status and specifications of every interpersonal cooperation construct
(e.g., competence of individuals or task dependence which can affect interpersonal
cooperation) with clear definitions, units and value ranges; (2) model behavioral rules of

11

interpersonal cooperation: the model rules were represented as cooperative behaviors and
logic relationships among different cooperation constructs which were established based
on the literature and case studies; and (3) build an overall model architecture: the model
architecture was finally established to capture attributes and methods derived from
interpersonal cooperation representations.

Notably, the model outputs also needed to be represented as part of the model architecture,
i.e., key indicators of construction project team performance. This study first examined
the major construction literature to identify the team performance measurements applied
in the industry. Then these findings were summarized to provide a framework for
measuring construction project team performance. Finally specific measures of individual
indicators were defined and formulated according to the requirements of model
development. The representation of interpersonal cooperation and project team
performance established the foundation of the model development.

Task 3: Developing an Agent Based Model for interpersonal cooperation in
construction project teams

An Agent Based Modeling (ABM) simulation platform (Virtual Organizational Imitation
for Construction Enterprises, VOICE) was developed based on the conceptual model’s
architecture. The following process was utilized to realize the simulation platform: (1)
schematic design: design and define the basic agents, structures and behaviors, and select
the basic algorithms; (2) prototyping: create an outline model that includes the most basic
12

agents and related behaviors to test the feasibility of the proposed architecture; (3)
designing agents: specialize and revise agents’ properties and behaviors, together with
related algorithms used in the prototype, to make agents close to reality; and (4) designing
the agent environment: specify the agent’s world, which is the relationship between agents
in terms of organizational structures.

Task 4: Model verification and validation

A hybrid method was utilized to verify the developed model, with subtasks including
automated error detection, logic examination, unit testing and hypothetical cases. After
that, the developed model was validated with a twofold procedure following Sargent
(1999), including input validation which focuses on the key variables and micro-level
behavioral rules used in the simulation model and output validation which provided
certain level of justifications of the reasonableness of the outcomes generated by the
model.

Task 5: Simulation experiments and operational recommendations to specific
construction related cooperation issues

To illustrate the usefulness of the proposed model in investigating interpersonal
cooperation in construction project teams, a series of exploratory simulations were
performed using VOICE to examine team performance under different setups and initial
conditions of interpersonal cooperation. In each simulation experiment, organizational
context, relational behaviors of participants, work and other components of the
13

interpersonal cooperation are set to different values, and the results are recorded and
analyzed using uncertainty analysis. Although uncertainty analysis is always used
interchangeably with sensitivity analysis, it is mainly interested in the viability in the
simulation outputs while sensitivity analysis also investigates the quantitative correlation
between inputs and outputs (Smith and Smith 2007). Given the scope of case studies in
this dissertation, uncertainty analysis was used. Finally, operation recommendations were
developed for improving construction project team performance in studied cases through
certain strategies for interpersonal cooperation.

1.5. Dissertation scope
The main focus of this dissertation is the cooperative behaviors and processes pertaining
to the construction management activities (e.g., proposal development) at the home office.
Cooperation related to construction filed operations (e.g., steel erection) has not been
addressed due to the scope of this research. In addition, only project task related
cooperative behaviors and processes have been investigated in this research. Other
influential factors such as work-life balance or learning are not within the scope of this
research since they are the main foci of psychosocial studies (Frosh and Baraitser 2008).

1.6. Deliverables of this research
There are two main deliverables of this research. First, this research developed a
conceptual framework for investigating interpersonal cooperation in construction project
teams. Particularly, the proposed framework captures the key points of interpersonal

14

cooperation in the construction context, inducing the technical, institutional and
behavioral aspects that are believed critical for the performance of construction projects
and project teamwork. The framework highlights elements or antecedents of interpersonal
cooperation, and captures the transition from interpersonal cooperation elements to
ultimate construction project team performance.

Second, another outcome of this research is VOICE, a simulation model to investigate
real-world cooperation issues in construction project teams. VOICE was built on the basis
of the proposed conceptual framework following the modeling principals of ABM.
VOICE is able to reproduce how construction projects are divided into tasks and
processed by roles of a project team such as president, project managers and staff
members. The analytical function of VOICE (e.g., uncertainty analysis) allows
exploratory experiments for examining and quantifying the influences of different
interpersonal cooperation elements on construction project team performance.

1.7. Outline of this dissertation
The remainder of this dissertation is organized as follows:

Chapter 2 discusses the theoretical foundation of the research work, including a brief
discussion of cooperation theory, an overview of previous cooperation studies in
construction academics, the nature and evaluation of construction project teams, and the
rationale of applying ABM in this research.

15

Chapter 3 extends the review of literature on cooperation theory and construction teams to
build a conceptual framework specific for interpersonal cooperation in construction
project teams. Major constructs of interpersonal cooperation are proposed; each construct
is discussed in detail by summarizing theories pertaining to the corresponding theme.

Chapter 4 builds a simulation platform (Virtual Organizational Imitation for Construction
Enterprises, VOICE) on the basis of the proposed conceptual framework, for investigating
interpersonal cooperation and its implications in construction project teams. This chapter
provides the computing details and research strategies of each critical model component.

Chapter 5 introduces the verification and validation (V & V) of VOICE. A hybrid
methodology was employed to determine the validity of VOICE, including surveys,
document studies, structured interviews, and hypothetical case simulation experiments. A
testimonial attests that the V & V approaches used are good enough to ensure the
confidence of the following analyses based on VOICE.

Chapter 6 introduces two case studies to demonstrate the usefulness of VOICE. The first
case study focuses on how cooperative behaviors and institutional arrangement between
members of an estimating team impact management actions and team performance in
typical Design Bid Build (DBB) projects. The second case study investigates the role of
goal congruence in the cooperation between members of different teams during project
proposal development of EPC projects (Engineering, Procurement and Construction).
Causes, behaviors and consequences of goal incongruence are examined using VOICE
16

based on an in-depth case study to an American mega-million-dollar construction
enterprise. Based on an interpretation of the simulation results of the two case studies,
several operational recommendations are given to improve project team performance
through cooperation enhancement.

Chapter 7 reviews the whole work and provides a constructive evaluation of this research.
It identifies common and distinctive features of interpersonal cooperation in different
construction settings. The concluding chapter also presents on the limitations and the
ways to address them in the future.

17

CHAPTER 2: LITERATURE REVIEW
This chapter reviews the literature in construction studies and other relevant areas to
identify the gaps of current studies and conceptualize the focus problems in this research.
Three categories of literature are covered including cooperation theory, construction
project teams and Agent Based Modeling (Figure 2.1).

Literature

Cooperation theory

Construction project
teams

Definition and category

Definition

Implications of
cooperation

Nature of construction
project teams

Previous cooperation
studies in construction

Performance
measurements

Agent Based
Modeling
Concepts
Applications in
construction
Rationales in this
research

Figure 2.1 Categories of literature review

2.1. Cooperation theory

2.1.1. Definition of Cooperation in construction
Cooperation has no explicit meaning in the construction literature (Anvuur 2008). Rather,
it is always used under notions of coordination or partnering (Anvuur 2008), and is often
confused with collaboration (Pocock and Kim 1997; De Saram and Ahmed 2001a;
Cheung et al. 2003; Naoum 2003; Phua and Rowlinson 2004; Jha and Iyer 2006). The
majority of construction literature seems to use cooperation and other relevant terms
18

interchangeably (Bresnen and Marshall 2000a; Vaaland 2004; Eriksson and Laan 2007;
Eriksson and Pesämaa 2007). This variability in terminology clouds the understanding of
the basic concept and makes the integration of the contributions of different researches
impossible. To clarify the distinction, relevant concepts are compared below.

Coordination is the management of the dependencies between activities (Bresnen and
Marshall 2000a; Becerik 2004; Cicmil and Marshall 2005), the “extra activities required
to maintain consistency within a work product or to manage dependencies within the
workflow” (Malone and Crowston 1994). It also refers to additional information
processing when multiple actors work toward a larger goal (Curtis 1989).

Partnering is a generic management approach to align project goals or strategies (Malone
and Crowston 1990). It is a way to achieve an optimum relationship between project
participants (Love et al. 2002; Bayliss et al. 2004), either in particular projects or as a long
term “strategic alliances” (Cowan 1992).

Collaboration is the long-term alignment and integration between organizations (Cheung
et al. 2003). It signifies a higher degree of integration in the partnering relationship
(Hamel et al. 1989; Kanter 1994; Thompson and Sanders 1998; Love et al. 2002), and is
always conceptualized as opposite to competition in inter-organizational relationships
(Thompson and Sanders 1998).

Cooperation is mostly defined as individual behaviors to achieve mutual goals (Hamel et
al. 1989; Thompson and Sanders 1998); or the interpersonal interactions when different
19

functional units work together for overall organizational tasks (Johnson 1975). It reflects
the willful contribution of team members to the achievement of shared goals (Tjosvold
1988; Pinto and Pinto 1990).

From the above discussion, it is clear that cooperation is very different from other
concepts. First, cooperation is mainly a concept about interpersonal interactions to
achieve a mutual goal, while collaboration or partnering is a term used in illustrating the
integration or strategic alliance between organizations. Second, the scope of cooperation
could be either inter- or intra-organization, while collaboration or partnering tends to be
across, rather than within, organizational boundaries. Third, cooperation emphasizes the
pattern of interactions and social relations, while other concepts (e.g., coordination) are
mainly about the management activities. Table 2.1 summarizes definitions given by the
literature with the same inference. Accordingly, in this research cooperation in the
construction context is defined as: A behavior toward the accomplishment of the shared
construction project goals of all project parties, and also a process that underlines the
pattern of interactions among project participants.

The functionality of both the behavioral and process perspectives of cooperation are
supported by the literature (Wagner III 1995; Dukerich et al. 2002; Phua 2004), and
therefore both perspectives are central foci of this research. In construction, cooperative
behavior often involves “problem-solving, creating new value together or striving for winwin scenarios” (Chen et al. 1998), while the process perspective implies the dynamic

20

developmental approach to cooperation (Anvuur 2008). The basic premise of this study is
that understanding both perspectives can provide better insight into the formation and
consequences of cooperation. As the readers will find in chapter 4, the simulation model
developed in this research captures cooperative behaviors as the actions of team members,
and cooperative processes as how project tasks are assigned, processed, coordinated,
delivered and submitted in a project team.
Table 2.1 Definitions of Cooperation
Literature
(Schermerhorn Jr 1975)

Terminology
Cooperation

(Johnson 1975)

Cross-functional
cooperation

(Pinto and Pinto 1990)

Cross-functional
cooperation

(Pinto et al. 1993b)

Cooperative interorganizational
relationships

(Ring and Van de Ven
1994)

Cooperation

(Wagner III 1995)

Cooperation

(Smith et al. 1995)

Inter-organizational
cooperation

21

Definition
“The coordination of behaviors among
individuals to achieve mutual goals”
“The quality of task and interpersonal
relations when different functional
areas work to accomplish
organizational tasks together”
“Joint behavior toward some goal of
common interest. In context of
construction projects, it is
conceptualized as the degree, extent
and nature of interpersonal
relationships among project team
members from multiple functional
areas”
“Socially contrived mechanisms for
collective action, which are continually
shaped and restructured by actions and
symbolic interpretations of the parties
involved”
“The willful contribution of personal
effort to the completion of
interdependent jobs”
“Individuals, groups and organizations
come together to interact and form
psychological relationships for mutual
gain or benefit”
“The presence of deliberate relations
between otherwise autonomous
organizations for the joint
accomplishment of individual

Table 2.1 Cont’d
Definitions of Cooperation
Literature

Terminology

(Smith et al. 1995)

Cooperation

(Tyler and Blader 2000)

Cooperation

(Bennett 2000)

Cooperation

(Mead 2003)

Definition
operating goals”
“Refers to whether or not people act to
promote the goals of the group”
“Cooperation means people working
together on the basis of common
interests. It is the most effective
behavior to adopt in all human
relationships”
“The act of working together to one
end”
Refers to “where individuals acting
together benefit more than they would
alone”

Cooperation

2.1.2. Categorization of cooperation
Cooperation research, as Smith et al. (1995) highlighted in an extensive literature review,
is “rich in theory and diverse in its academic roots”. A rich body of literature discusses the
contents of cooperation (Terhune 1970; Schermerhorn Jr 1975; Axelrod and Hamilton
1981; Schmitt 1984; Pinto et al. 1993b; Chatman and Barsade 1995; Smith et al. 1995;
Wagner III 1995; Chen et al. 1998; Tyler and Blader 2000; Mead 2003; Roberts 2005;
Bewsey and McCord 2006; Anvuur 2008). However, because of the lack of a generic
framework, the literature always gives different or even contradicting conclusions
(Yilmaz and Hunt 2001). In order to propose a generic framework that incorporates all
theoretical perspectives, the first step is to revisit the theoretical traditions of cooperation
research.

The literature has recognized cooperation as a multifold concept (Rabbie 1991; Smith et al.
1995; Tyler and Blader 2003; Anvuur 2008). Oriented from Behavioral Interaction Model,
22

Rabbie (1991) found that two kinds of cooperation can be distinguished, i.e., instrumental
cooperation and social cooperation. In instrumental cooperation, people cooperate to
attain economic or other tangible outcomes; whereas in social cooperation (or relational
cooperation), the aim is to achieve a mutually satisfying relationship with others so that
one can be recognized as a positive social identity. As a result, cooperation can be
grouped into two categories: formal cooperation that stems from the formal hierarchy, and
informal cooperation rooted in a belief that others will faithfully contribute to the group
(Smith et al. 1995). This idea has recently been revisited by Anvuur (2008) who described
two kinds of cooperation: obligatory cooperation and voluntary cooperation. Obligatory
cooperation is triggered by “extrinsic factors” such as contractual obligations and formal
structures of job design. Voluntary cooperation is motivated by “internal factors”, i.e., a
positive sense of “self-identity” (Tajfel 1982). This research mainly focuses on obligatory
cooperation instead of voluntary cooperation.

2.1.3. Implications of cooperation
The benefits gained from better cooperation have been documented by the literature
(Albanese 1994; Dulaimi et al. 2002; Kadefors 2004; Phua and Rowlinson 2004). One
most apparent effect of cooperation is it can break down barriers and facilitate the
building of more stable and long-lasting relationship among different teams (Albanese
1994; Chan et al. 2003; Chan et al. 2009). As a result, enhanced cooperation improves the
level of trust and commitment between project team members (Morgan and Hunt 1994;
Kadefors 2004; Shek-Pui Wong and Cheung 2004; van Marrewijk 2005). Recent findings
23

also indicate that cooperation may foster innovation in construction organizations (Dubois
and Gadde 2002; Dulaimi et al. 2002; Dulaimi et al. 2003; Miozzo and Dewick 2004).

This research paid special attention to the importance of cooperation in enhancing project
performance and success (Latham 1994; Bennett and Jayes 1995; Barlow et al. 1997;
Bresnen and Marshall 2000b; Phua and Rowlinson 2004). Phua and Rowlinson (2004)
conducted extensive interviews and a survey of construction firms in Hong Kong to
identify the impacts of different levels of cooperation on construction success. Results
showed that cooperation between firms as well as between employees in one company can
significantly affect the success of both consulting and contracting firms. A critical
literature review regarding the UK construction industry reached a similar conclusion
(Fisher and Green 2001). Furthermore, these findings are consistent with a wider range of
literature (Latham 1994; Bennett and Jayes 1995; Barlow et al. 1997; Bresnen and
Marshall 2000b). Meissner et al. (2005) found closer cooperation between engineers in
project teams is the basis of high quality projects, short development duration and a
minimum of investment costs. Chen et al. (2004) found the establishment of a conflict
resolution strategy to be a critical success factor for partnering projects. However, as Phua
and Rowlinson (2004) and Anvuur (2008) have commented, the relationship between
cooperation and project performance has rarely been empirically tested; the conclusions
are mostly based on piecemeal and anecdotal evidence (Barlow et al. 1997; Bresnen and
Marshall 2000b; Fisher and Green 2001).

24

2.1.4. Previous cooperation studies in the construction literature
Cooperation research in the construction domain could be grouped into two categories:
“instrumental research”, i.e., those focusing on procedural, contractual, strategic and
technical remedies towards better cooperation, and “behavioral research”, i.e., studies on
the behavioral and attitudinal perspectives of cooperation.

2.1.4.1 Instrumental remedies towards better cooperation

Contemporary industrial remedies to cooperation in construction project teams show a
strong emphasis on procedural, contractual, strategic arrangements and collaborative
techniques and tools. Among the emerging integration approaches, one of the earliest
topics that has been investigated is partnering, a general management approach to line up
project goals and strategies (Bresnen and Marshall 2000; Wood et al. 2002; Naoum 2003).
It could be applied to enhance relationships between contracting parties, either in
particular project partnerships or as “long term strategic alliances” (Cowan 1992). In
partnering projects, all project parties are required to transform their businesses in terms
of relationships, processes, communications and leadership (CII, 1989, 1991). Despite the
difference between approaches, they still rely on a pragmatic and instrumentalist
perspective which states that in order to achieve a partnering arrangement, patterning tools,
techniques, and procedures for organizations to follow should be established (Bresnen and
Marshall 2000).

25

Another stream of thought to align the efforts of project parties is by adopting a Linguistic
Action Perspective (LAP) in construction managament (Koskela and Howell 2002). The
traditional dispatching model of construction management assumes that project tasks,
once authorized, can be automatically understood, started and finshed as planned (Koskela
and Howell 2008). Thus construction management simply refers to a strong casual
connection between the actions and the outcomes. However, as Flores found (1981),
management is indeed “a process of openness, listening, and eliciting commitments,
which includes concern for the articulation and activation of the network of commitments,
primarily produced through promises and requests”. As a result, LAP has been proposed
to highlight the need that project tasks should be “coordinated through making and
keeping commitments, rather than by central control” (Koskela and Howell 2008). At
present, LAP essentially serves as the theoretical foundation of a variety of successful
Lean Construction practices such as Last Planner System (LPS) (Macomber and Howell
2003). For example, a core of the successful lookahead schduleing in LPS is a
commitment cycle beginning with “a request, followed by a promise, performance and
declaration of completion” (Macomber and Howell 2003). In this sense, LPS eocnureages
cooperation because it articulates and activates a network of reliable commitments so
people can organize and assemble the efforts to deliver the promise of the project while
acting in their own interests (Macomber and Howell 2003).

Recently, Integrated Project Delivery (IPD) has also received increasing attention as a
philosophy and approach to encourage cooperation among project participants (Bishop
26

2008). IPD builds on a variety of contractual and economic principles, including early
involvement of project participants, collective decision making, shared project criteria,
shared financial risk, liability waivers, fiscal transparency between project partners, and
intensified design (NASFA et al. 2010). From a wider context of economic models of
contracting, IPD actually reflects the principal of relational contracting (Matthews and
Howell 2005), which refers to “a transaction mechanism that seeks to give explicit
recognition to the commercial relationship between project parties” (Colledge 2005).
Relational contracting views contracts as relations rather than as discrete transactions
(Macneil 1973). It captures a wider social and economic context (Cullen et al. 2005), and
focuses on building long term trust and relationship among project parties (Matthews and
Howell 2005; Sakal 2005). In contrast, traditional construction contracting method
encourages local optimization behaviors which lead to adversarial and competitive
relationship (Colledge 2005; Matthews and Howell 2005).

In fact, construction activities are highly specialized, involving multiple parties with
durations for different commencement and completion, which necessitates relational
approaches even on the simplest of building project (Colledge 2005). Therefore it is not
surprising that relational contracting approaches have been successfully implemented in
many construction projects (Cullen et al. 2005; Gerrard 2005; Lichtig 2005). At present, a
number of national standard construction contract formats have been developed based on
relational contracting, including “Standard Form Multi-Party Agreement for Integrated
Project Delivery” (AIA C191–2009), “General Conditions of the Contract for Integrated
27

Project Delivery” (AIA A295–2008), and “Integrated Form of Agreement for Lean
Project Delivery” (IFoA). The relational contracting formats reflect a tendency of the
industry to remove the obstacles of cooperation and innovation by aligning the interests of
all project participants (Matthews and Howell 2005).

The literature also has shown a growing interest in collaborative techniques and tools,
especially Building Information Modeling (BIM), a technology that applies computer
generated 3D/4D models to simulate and communicate the life cycle of a facility (Azhar
et al. 2008). Project participants -- designers, engineers and contractors -- use BIM to
visualize the project in a simulated environment in order to identify potential problems in
advance, even before the project begins (Eastman et al. 2008). Because of the apparent
benefits gained from BIM, it has now become wide spread in the entire industry (Azhar et
al. 2008). As suggested by McGraw-Hill’s Smart Market Report on Interoperability, 2008
was the tipping point for BIM in the sense that, after 2008 it has become an inevitable
technology in the industry (Gudgel 2008).

2.1.4.2 Studies on behavioral and attitudinal perspectives of individual cooperation

Besides the focus on instrumental aspects of cooperation, a small but fast growing group
of scholars have noticed the importance of looking into the social, behavioral and cultural
aspects of cooperation (Cheung et al. 2003; Shek-Pui Wong and Cheung 2004). Barlow
and Cohen (1996) highlighted the importance of mutual goals, interpersonal trust, and an
appreciation of each other’s commitments in the formulations of most present partnering.
28

Green and McDermott (1996) found an embedded culture in construction settings where
the attitudes and behavior of project team members are deeply ingrained, so it is difficult
for project teams to rapidly change the organizational environment (e.g., moving to
partnering). Besides, cumulative evidence highlights the importance of behavioral and
attitudinal factors in forming a successful collaborative project team, including
commitment, trust, and mutual respect (Li et al. 2000; Cheung et al. 2003). Efforts have
been made to quantitatively link behavioral and attitudinal factors to the ultimate
performance of team integration effort (Cheung et al. 2003; Shek-Pui Wong and Cheung
2004). For example, Cheung et al. (2003) conducted an extensive survey to examine the
factors affecting the quality of partnering. The findings indicate that attitude oriented
factors explain the most variance (0.818). However, although these attitudinal and
behavioral factors are regarded as central to establishing a cooperative culture in project
teams, the fashion in which they are formulated is less clear (Bresnen and Marshall
2000b). Limited efforts have been made to investigate the mechanism and process of how
these behavioral and attitudinal factors shape or affect cooperation in construction project
teams. As described by Kadefors (2004), pivotal behavioral and attitudinal factors of team
cooperation or integration remains a “black box”.

It is worth noting, however, that some scholars have recently started looking into the
formation mechanism of cooperation or integration of construction project teams, instead
of simply providing statistical evidence (Chinowsky et al. 2009; Son and Rojas 2010;
Unsal and Taylor 2010). Assuming construction project teams are temporary
29

organizations, Son and Rojas (2010) developed an Agent Based Model to investigate the
evolution of cooperation within inter-organizational networks of construction project
teams from the game theory and social network perspectives. Their findings revealed a set
of rules regarding the evolution of cooperation among participants from different
organizations. For example, they confirmed that the fewer individuals are familiar with
others in the inter-organizational network, the longer time it takes for networks to reach
stable states. Additionally, the tendency of cohesion increases as the effort to form
relations with outside partners rises. Furthermore, the efficiency of inter-organizational
networks is negatively correlated with the efforts needed to form relations with those from
other organizations (Son and Rojas 2010). This work highlights the role of interpersonal
cooperation in forming a robust inter-organizational project network.

Unsal and Taylor (2010) integrated an Agent Based model with game theory to examine
contractors’ subcontractor selection as a repeated game. Their investigation centered
around the contractors’ investments in learning by introducing an innovation-oriented
organizational change across the project network. The findings indicate that a long term
relation-based selection strategy might increase the adaptation rate to an innovationoriented organizational change. This work, in another sense, proves that cooperation
between a contractor and subcontractors builds on willingness to establish a long term
relationship. Chinowsky et al.

(2009) introduced an innovative modeling approach,

Project Network Interdependency Alignment, to identify potentially excessive or
insufficient communication and knowledge exchanges in a project network which can
30

make projects ineffective. They found that project effectiveness is highly dependent on
the alignment of actual knowledge exchanges with knowledge exchange requirements
across task-organization network dyads. Their work reproduces the process of how the
interdependency of project participants, in terms of knowledge sharing, shapes the
effectiveness of a project. The difference between these newly emerged studies and
previous survey based studies is that they place more attention on the formation process of
cooperation between individuals from different organizations, while previous efforts were
more interested in revealing the ultimate quantitative relationships between driving factors
(such as trust, commitment) and the level of cooperation.

2.1.4.3 Four quadrants of cooperation studies in the construction area

Following the above discussion about cooperation studies and enhancement practices, it
was found that current studies and practices concerning better understanding and
enhancement of cooperation in construction project teams can be framed into four
quadrants as illustrated in Table 2.2. They are organizational level with a focus on
instrumental aspects of cooperation, individual level with a focus on instrumental aspects
of cooperation, organizational level with a focus on behavioral aspects of cooperation, and
individual level with a focus on behavioral aspects cooperation.

In particular, this study has put special emphasis on the behavioral and attitudinal
perspectives of cooperation at the individual level, with a consideration of the

31

organizational and procedural arrangement. This is because of the nature of construction
project teams.

Table 2.2 Quadrants of cooperation enhancement practices and studies

Organizational
level

Individual
level

Instrumental perspective
Partnering
Strategic alliance
IPD
BIM
Inter-organizational
collaboration
Organizational hierarchy
Job responsibility
Coordination

Behavioral perspective
Relational management
Network between organizations
Inter-organizational trust
Inter-organizational relation
Inter-organizational dependence
Social network between individuals
Interpersonal trust
Interpersonal commitment
Interpersonal relationships
Interpersonal cooperation

2.2. Construction project teams

2.2.1. Definition of construction project team
Despite numerous definitions in the literature (Verville and Halingten 2003), a
representative definition of team has been given by Kozlowski and Bell (2003):

“A team is a collective who exists to perform organizationally relevant tasks, share one or
more common goals, interact socially, exhibit task interdependencies, maintain and
manage boundaries, and are embedded in an organizational context that sets boundaries,
constrains the team, and influences exchanges with other units in the broader entity.”

This definition captures certain key elements of construction project teams. Particularly, it
reveals the fact that a team is a collective with many levels of interdependence and it
operates in an environment that determines the quality of functioning. However,
32

construction project teams have certain unique features that are not reflected in this
definition. On a construction project, a project team is comprised of colleagues from
different functions (for example, estimation, scheduling, and accounting departments are
brought together to perform project planning) and disciplines (architects, engineers, and
contractors form a team to perform project activities) who interact regularly to execute
project related tasks and activities, but there is no permanent set of boundaries for
construction project teams. Plus, this approach means a construction project team is
probably embedded in multiple organizational contexts (since its members may come
from different organizations instead of just one). Moreover, construction project teams are
project-based organizations, which means that completing project goals instead of
organizational goals is a better description of the construction project team’s essential
function (Thiry and Deguire 2007). Therefore, based on Kozlowski and Bell’s definition
with a consideration of the uniqueness of construction project teams, this study offers the
following definition for construction project teams:

A construction project team is a temporary collective with members from different
functions and disciplines. It performs construction project related tasks, shares mutually
agreed project goals, interacts regularly, demonstrates task interdependencies, and is
embedded in particular organizational contexts which determine the quality of its
functioning.

33

2.2.2. The characteristics of construction project teams
Several unique features distinguish construction project teams from other kinds of groups.
First, construction project teams are usually regarded as temporary organizations (Son and
Rojas 2010). The construction industry is represented by a huge number of small sized
and distributed players (Bureau of Census 2011). On the one hand, single construction
organization hardly have the necessary ability to perform all the activities of a project
(Son and Rojas 2010). Even within a construction organization, given the increasing
complexity of construction project tasks, there is a need to coordinate between various
functional departments within the organization (Ahuja et al. 1994). Thus construction
projects, especially those that are large and complex, are ultimately built on a
collaborative effort across the boundaries of a number of construction organizations or
functionalities (Son and Rojas 2010). On the other hand, construction projects are also
usually “one-of-a-kind” efforts, i.e., no project before or after will be exactly the same. As
a result, when there is a project, different disciplines or functions are brought together to
form a construction project team; but once the project is finished, the team will be
dismissed immediately (Alshawi and Faraj 2002). This nature of construction projects
creates a need for “temporary modes of operation and thereby tends to promote highly
dispersed management practices that do not dovetail very well with other organizational
processes” (Bresnen et al. 2005). There is seldom a common organizational structure
existing in construction project teams, and the hierarchy and power alignment varies
depending on the individuals involved (Newcombe 1996).
34

Second, the increasing use of outsourcing has led researchers to introduce the term
“virtual teams” to describe construction project teams (Nayak and Taylor 2009).
Chinowsky and Rojas (2004) defined virtual teams in engineering, procurement, and
construction activities as “a group of people with complementary competencies executing
simultaneous, collaborative work processes through electronic media without regard to
geographic location.” The increasing application of “virtual” in the construction industry
comes from the advances in technology and market globalization in recent years (Nayak
and Taylor 2009). On the basis of emerging information communication technologies, a
growing number of construction companies are outsourcing portion of their work to
vendors, such as those in IT systems (Barthorpe et al. 2004), and complex project design
(Joseph 2005). A recent survey reveals that 44% of engineering companies in the United
States are using outsourcing to release cost burden and improve efficiency (Bryant 2006).
Due to the significant geographic distances that separate clients from vendors, much
teamwork is executed in a fashion that crosses time, space, organizations and cultures
(Maznevski and Chudoba 2000). Besides the potential benefits gained from virtual teams,
the construction literature has already noted the challenges faced in virtual project teams,
such as issues concerning communication, trust development and quality control (Nayak
and Taylor 2009).

Third, construction project teams can also be considered as a “potential team” following
the work of Katzenbach and Smith (1993). Katzenbach and Smith (1993) proposed the
“team performance curve” theory to differentiate team into five types according to their
35

level of effectiveness and impacts on performance: “working group”, “pseudo team”,
“potential team”, “real team” and “high-performance team”. The definition and major
features of each team are given in Table 2.3. According to their definition, most
construction project teams can be categorized as “potential teams”: construction project
team members are actively engaged in project activities, but hesitate to fully coordinate
and contribute their efforts because each member represents the interests and foci of
his/her own organization or department. As a result, team conflicts, disagreements and
misunderstandings are commonly seen in construction project teams. Katzenbach and
Smith (1993) found that although the members were targeting the same goal, conflict and
an unpleasant atmosphere were almost inevitable. The “potential” therefore refers to the
chance to become a real or high-performance team through team development including
effective training, cooperation experience, realigned business goals or even, simply, time
(Katzenbach and Smith 1993).
Table 2.3 Five types of teams; modified from (Katzenbach and Smith 1993)
Types

Working
group

Definition

Relationship
between
members
A group with no clearly defined Team members
performance goal or need to become a are loosely
team. The members might share grouped
information with each other primarily
in order to help within individual job
responsibility.
Example: A group of administrative
staff whose tasks include answering
phones and scheduling appointments.
They are all working as part of the
same group, but have no common and
measurable goal to attain (Feinberg
36

Team’s impacts
on performance
There is no
“common goal”
and team’s
impact on the
performance is
very low

Table 2.3 Five types of teams; modified from (Katzenbach and Smith 1993)
Cont’d
Types

Definition

Relationship
between
members

2009).
Pseudo
A group
which
might
have
team
performance goals but doesn’t focus
on collective effort to achieve the
desired results.
Example: A group of people who
claim they are a team, but has no
interest to develop a common goal
(Feinberg 2009).
Potential
A group might have an agreed
team
performance goal and members might
try to work towards the goal; but the
goal needs clarification.
Example: A project team with
members from different disciplines
working towards the completion of the
project; however, conflicts always
exist because they represent the
interest of different organizations or
departments (Katzenbach and Smith
1993).
Real team
A small number of members who are
committed equally to agreed mutual
goals working closely together to
prove their accountability as a member
in the team. Normally, team members
have complementary skills.
Example: A surgical team performing
a heart transplant. This team consists
of medical specialists ranging from
nurses to various doctors. Team
members’ skills are highly matched
and team goal is clear and determinant
(Feinberg 2009).
High
A group that builds on a real team,
performance with
all
members
completely
team
committed to each other with respect
to personal success and growth.
Example:
Manhattan
Project’s
members built their personal growth
on the success of the project (Bodwell
2006).
37

Team’s impacts
on performance

Nominal working
relations between
team members
have been
developed

Performance goal
are not addressed
at all or are
superficially
defined

Team members
are actively
involved in team
tasks, but have
not fully
contributed their
efforts.

There is a
common
performance
goal, but needs
clarity about
purposes, goals,
and requires the
development of a
common work
approach

The members
have
complementary
skills, and
cooperate (rather
than compete)
with each other.
Team structure is
clear and
effective.

Team members
have a clear
performance goal
and team’s
impacts on
performance is
very high

The members are
highly
collaborative and
committed to
each other

A very high
impact on
performance

2.2.3. Cooperation research suitable for the features of construction project team:
from “top-down” to “bottom-up”
Given the features of construction project teams (i.e., temporary, virtual and potential), the
position and scope of the proposed cooperation research must consider the uniqueness and
critical needs of construction project teams. This study’s positions and scope are based on
the following arguments:

First, the contention that construction project teams are temporary teams highlights the
need to examine the relational factors (e.g., interpersonal cooperation, trust and
commitment) between team members instead of legitimation building. Lundin and
Soderholm (1995) found that, in temporary organizations, the relationships between
individuals, such as commitment building, are of central interest, compared to
legitimation building. Individuals hold their own defined expectations and carry different
levels of experience into the team. These experiences and expectations are possibly
different or even conflicting (Lundin and Söderholm 1995). Moreover, every member in a
temporary organization knows exactly that there is an specified end to the collective effort,
and individuals could also come and go at different times in the team (Miles and
Laboratories 1964), as a result the “rules of the game” may frequently change since new
members may appear at any time (Lundin and Söderholm 1995). Therefore, common
organizational structure or hierarchy does not count; rather, interpersonal relational factors,
such as interpersonal commitment (Lundin and Söderholm 1995), trust (Kanawattanachai
and Yoo 2002), and interpersonal cooperation (Bijlsma and Costa 2003) constitute the
38

basis for forming a temporary organization. This argument has been supported by the
construction literature: Chinowsky et al. (2010) proposed a social network model to
depict the relations between team members in a construction project. The findings indicate
that a project network has no stabilized structure among different projects. As such, it was
found that professional trust, effective interpersonal communication and knowledge
sharing form the foundation of delivering a successful construction project (Chinowsky et
al. 2010).

All the evidence above implies that cooperation research suited to the

construction context should pay special attention to the relational factors among project
team members.

Second, the view of construction project teams as virtual teams highlights the importance
of social aspects of cooperation between team members in terms of fostering the
cooperative efforts of the team. Due to geographical distribution, face-to-face
communication rarely occurs in many construction project teams. This results in weaker
social relations between team members and makes the team become more “task-focused”
rather than “socially focused” (Powell et al. 2004): cohesion and trust among team
members becomes more difficult when people rarely meet face-to-face (McDonough III et
al. 2001); Effective communication and coordination among team members are
considered as serious challenges (Nayak and Taylor 2009); and interpersonal relationships
between team members are very hard to build (Robey et al. 2000). Therefore, Powell et al.
(2004) highlighted the significance of social-emotional problems involved in bringing
team members closer and forming interpersonal cooperation in virtual teams. Although
39

social-bonding can be accomplished partially via electronic communication tools (Joseph
2005), it was argued that the focus of virtual project teams remains on relationship
building rather than on actual business (Robey et al. 2000). In short, studying cooperation
issues in construction project teams must emphasize the social-emotional process between
team members.

Last, the potential team nature of construction project teams indicates that the future trend
in team performance improvement should focus on aligning the efforts of individuals and
building long term cooperative relationships between team members. A variety of
construction literature has already highlighted the need to move from traditional project
teams to high-performance project teams in the construction industry (Jenner 1997; Boetti
and Leandro 2008; Novelo and Gabriel 2010). This will require additional research efforts
to investigate the characteristics of high performance project teams (Novelo and Gabriel
2010). According to Katzenbach and Smith’s performance curve theory (Table 2.3), the
difference between a potential team and a high-performance team is lack of interpersonal
alignment and long term commitment and trust between team members (Katzenbach et al.
1993). Therefore, the fundamental step for building future high performance construction
project teams is figuring out how to coordinate the efforts of project team members, as
well as exploring why people form long term interpersonal relationships in a construction
project team. To achieve this target, the first step is to better understand interpersonal
cooperation mechanism in construction project teams is necessary; i.e., why and how
people cooperate in a construction project team.
40

The natural features of construction project teams offer a clear clue to what the proper
research positioning for cooperation research should be in the construction context. Most
existing efforts have put too much emphasis on organizational collaboration and strategic
alignment as discussed in previous sections. Such a method aims to provide a remedy of
cooperation issues in the construction industry through “top-down” contractual,
institutional and strategic improvements. Although this has worked to some extent, the
“top-down” approach has proven insufficient to capture the real needs of enhancing
cooperation in construction project teams (Moore and Dainty 2001).

Based on the analysis of the natural features of construction project teams, this study
proposed that a “bottom-up” approach, i.e., investigating the behavioral and attitudinal
aspects of cooperation at the individual level, is a good fit for cooperation research in the
construction context (Figure 2.2). It is expected that, by enhancing interpersonal
cooperation, especially by behavioral changes of team members, new insight into
cooperation issues in construction project teams will be advanced.
Top-down approach

Contractual/Procedural/Strat
egic/Technical remedies

Inter-organizational

Cooperation issues in
construction project teams

Bottom-up approach

Enhancing interpersonal
cooperation by behavioral

Inter-personal level

Figure 2.2 Two approaches to enhancing cooperation in construction project teams
41

2.2.4. Performance of construction project teams
Team performance, in general, is a multifold concept. The literature has suggested that
team performance could be defined at different levels and for different perspectives
(Klammer 2002). First, team performance is defined by the literature at different levels. In
2008, Mathieu et al. conducted an extensive review on the team performance literature
from 1997 to 2007. They found that the literature uses team performance on three levels:
organizational-level performance, which considers team and organizational outcomes as
identical; team behavior and outcome performance, which defines team performance as
the process toward and results of the expected performance goal; and role-based
performance, which considers team performance as team members’ individual
performance (Mathieu et al. 2008). Second, the literature also tends to define team
performance from different perspectives. Based on the work of Cohen and Bailey (1997),
Klammer (2002) summarized the use of team performance in the literature, and
highlighted three dimensions of team performance that are commonly used in research
pertaining to teams: behavioral dimension, attitudinal dimension and outcome measures
(i.e., the quantity and quality of outputs). As a result, Salas et al. (2007) found the use of
team performance is quite mixed in current literature, with more than 130 different
frameworks for team performance. Following the above notion, Table 2.4 demonstrates
the dimensions of team performance on the basis of the literature (Salas et al. 2007).

42

Table 2.4 The dimensions of team performance
Organizational
level

Measurements

Measurements

Measurements

Team

Measurements

Measurements

Measurements

Measurements

Measurements

Measurements

Outcome/process
measures

Behavioral
measures

Attitudinal
measures

level
Individual
level

Because of the mixed meaning of team performance, Brannick et al. (1997) suggested
that team performance should be defined to fulfill a variety of perspectives including
“purpose of measurement,” “measurement attributes,” and “measurement process.” The
definition and measure of team performance, therefore depend on the needs. For example,
if the purpose is to evaluate the results of teamwork training, the performance of
individuals within a team is the best measure. But if the purpose is to reward a team, it is
better to evaluate the performance of the entire team. The same way, emphasizing
behavioral and attitudinal performance indicators, or outcome performance indicators
should be considered within the context (Brannick and Prince 1997). In construction
context, Rojas-Villafane (2010) argued that team performance should be defined as how
well the construction project team has worked to achieve its predefined goals, and it must
highlight aspects under the control of the team. According to this claim, Rojas-Villafane
(2010) grouped construction project team performance into process performance (or team
work performance), and outcome performance (or task performance). Team outcome

43

performance refers to how well the team did its job; and process performance measures
how the team performance is developed and improved over time (Rojas-Villafane 2010).

Rojas-Villafane’s argument is confirmed by the construction literature. The use of team
performance in construction publications can be grouped into two trends. One measures
team performance as the assessment of team outcomes, such as time, cost, quality,
productivity, risk and others (Barrick et al. 1998; Tesluk and Mathieu 1999; Langfred
2000; Kim and Burton 2002; Horii et al. 2005a; Kang et al. 2007; Leicht et al. 2010), and
the other trend measures team performance based on the quality of the team’s actions and
processes toward the accomplishment of outcome goals, such as efficiency and
effectiveness of communication among team members, wasted time, coordination time,
rework time, accumulative unfinished work, effective working time, and manager rated
work efficiency (Ancona and Caldwell 1992; Tommelein et al. 1999; Wong and Burton
2000; Chinowsky et al. 2010). Following the above notions, this research defines
construction project team performance as the final result of team work, i.e., team outcome
performance, and the quality of the actions and process toward the outcome goals in
particular construction projects, i.e., team process performance (Figure 2.3).

Previous literature review has served as the theoretical foundation to build a conceptual
framework for interpersonal cooperation in construction project teams. In order to convert
the conceptual framework to an executable simulation model, simulation methods were

44

reviewed. It was found Agent Based Modeling to be a perfect fit for this research. In the
following section, ABM is introduced.

Construction Project Team

Team outcome performance
• Final results of team work

Team process performance
• Quality of actions and processes
toward outcome goals

Figure 2.3 The framework of construction project team performance

2.3. Agent Based Modeling: a bottom-up approach

2.3.1. Concepts
As a computational modeling approach, Agent Based Modeling (ABM) is a suitable tool
for use in social research to study human and organizational issues in a diversity of areas
(North and Macal 2007). It is a computational method that builds a common environment
for heterogeneous and autonomous agents to share, and allows the agents to
simultaneously interact with each other for self-interest (Ligmann-Zielinska and
Jankowski 2007). Unlike top-down modeling approaches (e.g., System Dynamics,
Discrete Event Simulation etc.), in ABM the collective behavior of the simulated system
is not predefined, but emerges from individual agents who act based on what they
perceive to be their own interests. Thus, ABM is capable of reproducing the emergent
properties of the studied systems (Macal and North 2007).
45

2.3.1.1. Agent

The fundamental concept of ABM is “agent,” i.e., the basic unit that makes independent
decisions in the model (Macal and North 2007). An agent is an artificial entity (a
computational unit) that is able to perceive and act upon a common environment and
make decisions independently for itself to satisfy its own objectives or interests (Macal
and North 2007). Numerous arguments remain over the term “agent” because this idea has
been applied in many different domains (Schieritz and Milling 2003). However, certain
common characteristics of agent have been identified: Macal and North’s (2005)
representative description of agent is:

(1) Identifiable and discrete;
(2) Self-contained with a clear boundary;
(3) Situated in an environment where it interacts with other agents;
(4) Goal-directed: has own set of goals that it actively acts to fulfill;
(5) Autonomous: has a pool of attributes and rules determining its decision-making
processes and corresponding behaviors;
(6) Learning ability: able to adapt behaviors according to experience.

Additionally, Macal and North claimed that agent behaviors may vary in their complexity,
availability of information, the internal models of agent’s toward the external world, and
the extent to which an agent’s past memory can be retained in decisions (Macal and North
2005). Based on proposed features, they provided an agent framework as following figure:
46

Environment
Agent
•
•
•
•
•
•

Attributes
Behavioral rules
Memory
Resources
Decision making sophistication
Rules to modify behavioral rules

Figure 2.4 The framework of an agent; modified from (Macal and North 2005)

In organizational simulation, an agent may be seen as a person and/or a group of persons
who share a congruent target and the same function (e.g. a department). Also, an agent
can be seen as an organization in the macro environment. The definition of an agent in a
practical model depends on the granularity of the model. In this research, an agent is
regarded as an individual team member.

2.3.1.2. A bottom-up approach

ABM is a bottom-up approach (Bonabeau 2002). First, the fundamental elements of a
system, i.e., agents, are defined in great detail. Then agents are linked together to build a
series of subsystems, which in turn are connected again multiple times until a complete
system appears (Epstein and Axtell 1996). For example, the Santa Fe Institute has begun
the development of an artificial stock market since the 1990’s. These “bottom-up” models
always start from specified details of agent behaviors. On the basis of sophisticated
47

computational tools, these models can be used to describe macro characteristics of the real
world stock market by “growing” from a group of interacting individuals (LeBaron 2002).
In contrast, in top-down simulation, “aggregated quantities,” instead of the behaviors of
individual agents, play the sole role of determining the system’s outcome (LigmannZielinska and Jankowski 2007). The bottom-up approach of ABM makes it a perfect
candidate for tackling the emergency properties of complex systems, which refers to the
phenomenon that “the whole is more than the sum of its parts because of the interactions
between the parts” (Bonabeau 2002). Because ABM “models and simulates the behavior
of the system’s constituent units and their interactions, capturing emergence from the
bottom up” (Bonabeau 2002), it has been regarded as a canonical approach to model
emergent behaviors of construction systems (Prietula et al. 1998). Plus, because of the
bottom-up feature, ABM has also greatly promoted the development of Generative Social
Science which explores complex social processes as an emergent phenomenon growing
from dynamic interactions between fundamental entities with simple rules and parameters
(Epstein 1999).

2.3.2. Applications of ABM
Fields in which ABM can be utilized range from biology to engineering (LeBaron 2002;
Cederman 2003; Christiansen and Altaweel 2004; Gratch and Marsella 2004; Folcik and
Orosz 2006). Macal and North (2007) summarized the main applications of ABM as in
the following table:

48

Table 2.5 The applications of ABM (Macal and North 2007)
Society and Culture
Business and Organizations
• Ancient civilizations
• Manufacturing
Operations
• Civil disobedience
• Supply chains
• Social determinates of
terrorism
• Consumer markets
• Organizational
• Insurance industry
networks
Economics
• Artificial
financial Military
markets
• Command & control
• Force-on-force
• Trade networks
Biology
Infrastructure
• Population dynamics
• Electric
power
markets
• Ecological networks
• Transportation
• Animal group behavior
• Hydrogen
• Cell behavior and sub
infrastructure
cellular processes
Crowds
• Pedestrian movement
• Evacuation modeling
ABM has been utilized by a small but growing community of scholars to tackle a range of
difficult problems in the construction area, including engineering design (Soibelman and
Pena-Mora 2000), project organizations and network (Jin and Levitt 1996; Horii et al.
2005a; Taylor and Levitt 2007; Du and El-Gafy 2011), construction operations (Mohamed
and AbouRizk 2005; Watkins et al. 2009; Kim 2010), project management (Christodoulou
2010), supply chain (Xue et al. 2005), construction policy (Ducrot et al. 2004),
construction safety (Walsh and Sawhney 2004), and evacuation in different building
environments (Zheng et al. 2009). Although the topics are diverse, the foci of interest can
be grouped into three categories:

•

Project organizations: Increasingly, scholars are proposing the use of ABM to
study the complex behaviors of project organizations and networks, e.g.,
49

organizational performance, work related behaviors, project team cooperation, and
cross-cultural issues (Jin and Levitt 1996; Horii et al. 2005a; Taylor and Levitt
2007; Du and El-Gafy 2011). Representative work in this area is Virtual Design
Team, or VDT (Jin and Levitt 1996). As a multi-agent based simulation platform,
VDT aims to investigate issues in project based organizations (Jin and Levitt
1996). By merging the organizational structure with a project’s task precedence
diagram, VDT analyzes how interdependencies between concurrent activities
introduce additional coordination needs, and how organizational structure and
communication tools change the “coordination capacity” of project teams (Jin and
Levitt 1996). As a well-documented and well-validated organizational simulation
tool, it has been widely applied to studying project based organizations to estimate
project durations, costs, and quality. Many models have been extended from VDT
to investigate organizational learning, project network dynamics, integrated
technology impacts, and organizational alignment (Ortiz de Orue et al. 2009;
Unsal and Taylor 2009; Wong et al. 2009).
•

Construction site: These applications focus on the issues on a construction site,
such as onsite operations, equipment logistics, safety, and evacuation (Walsh and
Sawhney 2004; Mohamed and AbouRizk 2005; Watkins et al. 2009; Kim 2010).
Watkins et al. ’s work (2009) is representative. Using ABM, they reproduced a
simple masonry work site to explore the potential effects of individual and crew
interactions. Especially, they were interested in investigating the impacts of
50

construction site congestion on productivity. Their work revealed that there is a
nonlinear relationship between the number of laborers and overall productivity.
They proposed that ABM can be “used to efficiently utilize construction space,
and develop plans and schedules that account for congestion arising from crew
interactions in space” (Watkins et al. 2009).
•

Project management: These applications utilize ABM to enhance traditional
project management, such as project scheduling, resource optimization, and supply
chain management (Xue et al. 2005; Christodoulou 2010). For example,
Christodoulou (2010) created an ant colony optimization (ACO) model using
ABM to address the project shcdule optimization problem. He found that it is very
difficult to schedule projects with strict resource constraints using traditional
approaches, such as Critical Path Method (CPM); instead, ABM performed better
on such problems. The findings indicate a high accuracy (97%) of using ABM as
an enhanced scheduling approach in resource-constrained projects (Christodoulou
2010).

2.3.3. The rationale for using ABM in this research
There are several reasons driving the use of ABM in the proposed research. First, ABM is
a natural fit for the proposed bottom-up approach of investigating interpersonal
cooperation in construction project teams. As discussed in the previous section (section
2.2.3), this research proposes studying the micro-level behaviors of individuals to gain
deeper insight into the cooperation issues in the construction industry, because of the
51

natural features of construction project teams. This requires a comparable research
approach which can reproduce the transition from individuals to collective behaviors.
Epstein’s description of using ABM to accomplish generative research is a perfect
example: ABM generates a population of autonomous and heterogeneous agents (e.g.,
project team members), defines decentralized behaviors, and allows them to interact with
each other by simple rules, and therefore “grows” the macro results from the bottom up
(Epstein 1999). A variety of construction studies have used ABM to support a bottom-up
investigation (Walsh and Sawhney 2004; Watkins et al. 2009). It is expected that the
relationship between individual behavior and overall cooperation in construction project
teams is complicated; therefore ABM is a canonical approach to use in this research.

Second, ABM is able to capture the complexity embedded in cooperation issues in
construction project teams. Construction project, by its nature, is a complex system
(Bertelsen 2003a). Even in average projects, there are numerous topologies of the
interactions among project team members, heterogeneous participants’ behaviors, and
inevitable uncertainties (Pich et al. 2002a). Schalcher (2009) further concluded that the
highly interacting construction project teams and the entire construction process as well as
various interrelated activities, contribute to the complexity of construction projects. These
interactions, mostly between project team members, are nonlinear and dynamic; thus they
are difficult to study using pure mathematical and statistical analyses. In this sense, ABM
is a good candidate for mapping the complexity of the cooperative process in construction

52

project teams since it begins with agents’ preferences and behaviors, as well as
interactions among them (Bonabeau 2002).

Third, ABM provides a natural description of the studied system, i.e., cooperative process
in this research, which enables a better understanding of research problems, and enhances
the communication between the modeler and the participants (Bonabeau 2002). The
proposed research, building on real construction project data, involves an intensive
communication between the researcher and the domain experts (i.e., practitioners under
study). Top-down modeling, such as System Dynamics (SD), requires domain experts to
have a global understanding of the studied objects. In the construction domain, however,
practitioners typically have no insight into the overall processes, only their own
responsibilities and rules. This makes the communication of modeling results a problem in
top-down modeling. Instead, ABM works on micro-level processes and describes the
problem by reproducing each “behavioral” entity, i.e., the actions of a real team member.
This strategy provides a realistic picture for domain experts and thus enhances the
understanding of the model.

Given the above reasons, ABM is a good fit for studying the interpersonal cooperation
issues in construction project teams.

2.4. Summary
This chapter has reviewed an extensive body of literature pertaining to cooperation
theories, previous cooperation studies in the construction literature, theory about
53

construction project teams, and simulation technologies. The literature review identified
gaps of current studies and practices of cooperation in construction project teams,
introduced basic concepts and characteristics of cooperation and project team, and
validated the rationale of using ABM as the simulation method. This chapter builds a solid
point of departure for this research. In the next chapter, literature review is extended to a
deeper level to develop a conceptual framework for interpersonal cooperation in
construction project teams.

54

CHAPTER 3: A CONCEPTUAL FRAMEWORK OF INTERPERSONAL
COOPERATION IN CONSTRUCTION PROJECT TEAMS
This chapter extends the literature review to build a conceptual framework specified for
studying interpersonal cooperation and its implications in construction project teams. In
particular, generic driving factors of interpersonal cooperation are discussed with relevant
theories and studies; factors that are critical in the construction context are highlighted; a
list of performance metrics is identified as the index for quantifying the effects of
interpersonal cooperation on the performance of construction project teams. Finally, a
conceptual framework is developed based on an IPO (Input-Process-Output) view of
organizations.

3.1. Cooperation: a system perspective
Following recent findings of organizational science (Robbins 2005), individuals, teams
and groups act interdependently in organizations to achieve an ultimate goal. On the one
hand, psychosocial conditions of individuals, as the fundamentals of organizational
processes, significantly affect the technical quality of any formal or informal group
activities. On the other hand, the team runs as a group of distinct individuals and exerts
noticeable impacts on behaviors (Robbins 2005). People-organization relationships play
an irreplaceable role in forming organizational behavior and defining the outcomes.
Building on Robbins and Langton’s (1998) model of organizational behavior (Figure 3.1),
work performance is human output dependent on individual behaviors, group behaviors
and the organizational system, and affected by environmental change and stress.

55

OUTPUTS
Policies
and
practices

Culture

Organizational
structure

Work and
technology
ORGANIZATIONAL
LEVEL

Group
decision
making

Communication

Group
structure

Work
teams

Other
groups

CHANGE
&
STRESS

Leadership

Conflict

Power and
politics
GROUP LEVEL

Personality

Value and
attitude

INPUTS

Perception

Motivation

Competence

Learning

Individual
decision
making

INDIVIDUAL
LEVEL

Figure 3.1 Model of organizational behavior; modified from (Robbins and Langton 1998)

Robbins and Langton’s work (1998) constitutes the point of departure of this research to
explore the constituents of cooperation. It suggests a system approach to be adopted in the
56

investigation of interpersonal cooperation, considering interactions of a variety of process
levels including individual level, group level and organizational level. As addressed by
Bresnen and Marshall (2000b), any contribution leading to a better understanding of the
cooperation concept that integrates the multiple theoretical perspectives in construction
project teams would be a significant contribution to construction knowledge.

In addition, it is worth noting that these processes have different weights in the
construction context. For example, in design bid build projects, conflict management as a
part of team process is typically important because inter-organizational cooperation
between project parties (e.g., owner, general contractor, and sub-contractors) determines
the quality of teamwork and ultimate project performance (Vaaland 2004). Therefore,
constructs of interpersonal cooperation should be differentiated according to their
importance to construction projects and project teamwork.

An extensive literature review was therefore conducted, and the result supports the
assumption. An overall analysis on the cooperation literature highlighted the fact that
cooperation literature themes can be naturally grouped into four categories: individual
factors, relational factors, organizational factors, and task-related factors. These four
categories of factors define interpersonal cooperation in construction settings. The
remainder of this chapter summarizes the literature findings in detail.

57

3.2. Antecedents of interpersonal cooperation in construction project teams
The literature review highlights the important roles of four categories of factors in the
formation of interpersonal cooperation in construction project teams, including individual
factors, relational factors, organizational factors, and task-related factors. Therefore, this
research proposes that the antecedent variables of interpersonal cooperation in
construction organizations can be grouped into four distinct categories: (1) task
characteristics; (2) individual characteristics; (3) relational behaviors; and (4)
organizational context. Table 3.1 lists the categorized factors that shape interpersonal
cooperation, and identifies generic factors and those that are specific for construction
project teams. In the following sections, more detailed findings from the literature review
are discussed.
Table 3.1 Antecedents and determinant factors of interpersonal cooperation

Task characteristics

Factors
Task
dependence
(Construction
specific)

Explanation
The extent to which team
members are dependent on each
other to perform individual tasks.
Representation: work flow;
information need; a probability
that a task should be halted when
others are undergoing.

Task
uncertainty
(Construction
specific)

The variability and difficulty to
performance a particular task.
Representation: level of
complexity and repetitiveness

Task visibility
(Construction
specific)

The extent to which the
environment allows the
evaluation and monitoring of
58

Reference and Researches
(Deutsch 1949; Van de Ven et
al. 1976; Hechter 1988; Winch
1989; George 1992; Pinto et
al. 1993a; Baccarini 1996; Jin
and Levitt 1996; Loch and
Terwiesch 1998; Kazanjian et
al. 2000; Thompson 2003;
Levitt 2007; Chinowsky et al.
2009; Taylor et al. 2009;
Chinowsky et al. 2010; Unsal
and Taylor 2010)
(Van de Ven et al. 1976; John
and Martin 1984; Winch 1989;
Moenaert and Souder 1990;
Pich et al. 2002b; Whitley
2006)
(Souder 1981; Allen 1984;
Davis 1985; Moorman et al.
1993; Itoh 1994; Wagner III

Table 3.1 Cont’d
Antecedents and determinant factors of interpersonal cooperation
Factors

Roles
(Construction
specific)

Demographic
characteristics
(Generic)

Factors
identified by
CSF studies
(Construction
specific)
Communicatio
n
(Construction

Relati
onal

Individual characteristics

Personality
(Generic)

Explanation
individual performance in a
group.
Representation: the relevance of
a task
Personality traits that are
important determinants of
personal cooperativeness.
Representation: especially
agreeableness

Position of an actor engaged in
group activities.
Representation: individual
hierarchical level and position in
the work flow
Demographic factors that are
important determinants of
personal cooperativeness.
Representation: e.g., age,
experience, organizational tenure

Reference and Researches
1995; Ballard and Howell
1998; Soltani et al. 2002;
Bowden et al. 2006; Fan et al.
2007; Xue et al. 2007)
(Taylor 1911; Fayol et al.
1929; Galbraith 1973; Allen
1984; Tellegen 1991; Kaman
and McCambridge 1992;
Kuprenas and Nasr 2000;
Jesus and Vorster 2002;
Varvel et al. 2004; Cheung et
al. 2006; Pinto et al. 2009; Liu
and Zhai 2010; Nakata and Im
2010; Nguyen et al. 2010; Yiu
and Lee 2010) (Johnson and
Singh 1998; Miller et al. 2000;
Lawless 2001; Dikmen and
Birgönül 2003; Fleetham and
Griesmer 2006; Giritli and
Civan 2008; Singh and Eng
2009)
(Pennings 1975; Chatman and
Barsade 1995; Levitt 2007)

(Gouldner 1954; March and
Simon 1958; McCann et al.
1981; Ruekert and Walker Jr
1987b; Lu and Argyle 1991;
Wagner III 1995; McCabe et
al. 2005; Dabke et al. 2008;
Davis and Songer 2009; Pinto
et al. 2009; Shan 2010; Singh
and Jampel 2010)

Other individual factors that are
critical to work performance
Representation: e.g., competence,
supportiveness, organizing skills

(Bentley 1981; Graham 1989;
Jaselskis and Ashley 1991;
Belassi and Tukel 1996; Chua
et al. 1999)

Certain tasks need extra
information from other actors
before processing. As such, team

(Laughlin 1978; Tjosvold
1988; Sally 1995; Jin and
Levitt 1996; Clemmer et al.

59

Table 3.1 Cont’d
Antecedents and determinant factors of interpersonal cooperation
Factors
specific)

Explanation
members need to exchange
information and ideas for
processing tasks.

Coordination
(Construction
specific)

Finishing a particular task relies
on different divisions of labor;
therefore, the manager who is
responsible for a task needs to
coordinate between his/her
subordinates; also, requesting or
giving approval on particular
tasks belongs to coordinating
effort.
Trust-related behaviors are
demonstrated as the attitude and
preferences towards others’ work.
Literature indicates individual
assessment on task quality is
rather subjective. It reflects trust
level between people. An actor
compares the quality of
delivered/submitted work with
own quality threshold, and if
quality is not satisfied, work will
be reworked, returned, reassigned
or reported.

Trust-related
(Generic)

Reciprocal
(Generic)

Discussion/me

Reference and Researches
1998; Malcurat et al. 2000;
Shohet and Frydman 2003b;
Hoegl et al. 2004; Wong et al.
2005; Chen and Tien 2007;
Chinowsky et al. 2009; Balliet
2010; Migliaccio and Martinez
2010)
(Laughlin 1978; Trevino et al.
1987; Tjosvold 1988;
Clemmer et al. 1998; De
Saram and Ahmed 2001b;
Hoegl et al. 2004; Consoli et
al. 2007; Levitt 2007; Shen
and Chang 2010)

(Bonoma 1976; Laughlin
1978; Tjosvold 1988; Bennett
and Jayes 1995; McAllister
1995; Munns 1995; Clemmer
et al. 1998; Korczynski 2000;
Zineldin and Jonsson 2000;
Wood et al. 2002; Bijlsma and
Costa 2003; Hoegl et al. 2004;
Kadefors 2004; Shek-Pui
Wong and Cheung 2004;
Diallo and Thuillier 2005;
Wong et al. 2005; Bromiley
and Harris 2006; Khalfan et al.
2007; Pinto et al. 2009;
Girmscheid and Brockmann
2010)
(Deutsch 1949; Thompson
1967; Laughlin 1978; Tjosvold
1988; Pinto et al. 1993a; Jin
and Levitt 1996; Clemmer et
al. 1998; Ahmad 1999;
Thompson 2003; Hoegl et al.
2004; Fong and Lung 2007;
Levitt 2007; Taylor et al.
2009; Unsal and Taylor 2010)

Reciprocal activities come from
the need to balance member
contributions, i.e., all members of
the project team should be
equally involved to achieve
shared goals and the members
complement one another as best.
Relevant behaviors include
dealing with overloading, giving
assistance to other team members,
etc.
Cooperation also builds on
(Antony 1976; Laughlin 1978;
60

Table 3.1 Cont’d
Antecedents and determinant factors of interpersonal cooperation

Organizational context

Factors
eting
(Construction
specific)

Explanation
discussing problems and conflicts
constructively, sharing a common
purpose in discussion, and
encouraging diverse viewpoints.

Reference and Researches
Raboud 1988; Tjosvold 1988;
Schwartzman 1989; Boden
1995; Jin and Levitt 1996;
Clemmer et al. 1998; De
Saram and Ahmed 2001b;
McCowan et al. 2003; Adams
2004; Hoegl et al. 2004;
Tepper 2004; Levitt 2007;
Ioannidou and Fruchter 2009)
Organizational The contractual/formal alignment (Ouchi 1980; Souder 1981;
structure
of power, responsibility, and
Keller and Holland 1983;
(Generic)
arbitrary job delegation pattern
Allen 1984; Davis 1985;
Representation: hierarchy,
Peters 1986; Hechter 1988;
proximity, accessibility
George 1992; Pinto et al.
1993a; Itoh 1994; Wagner III
1995; Anvuur 2008)
Organizational How activities or tasks on the
(Pinto et al. 1993a) (Taylor
rules
and project team are mandated or
1911) (Fayol et al. 1929)
procedures
controlled
(Gouldner 1954; March and
(Generic)
Representation: work procedures Simon 1958; McCann et al.
1981; Galbraith and Nathanson
1982) (Ruekert and Walker Jr
1987b) (John and Martin 1984;
Moenaert and Souder 1990;
Moorman et al. 1993; Pinto et
al. 1993a; Pinto et al. 2009;
Nakata and Im 2010).
Organization
Shared values that facilitate
(Deshpande and Webster Jr
cultural norms individuals to understand
1989; Chatman and Barsade
(Generic)
organizational functions and in
1995; Wagner III 1995; Chen
return serves as common behavior et al. 1998; Yilmaz and Hunt
norms in the organization
2001; Nguyen et al. 2010)
Representation: individualism,
collectivism

3.2.1. Task characteristics
Task-related factors have been considered the most important component of cooperation
(Yilmaz and Hunt 2001). The interest in task characteristics among cooperation scholars
dates back to Deutsch (1949), who claimed that social interaction between people mainly
61

stems from the perceived dependence of individual goals towards tasks. As such,
individuals tend to cooperate if they view the goals of tasks to be positively dependent and
task characteristics require such cooperation to achieve dependent goals (Deutsch 1949;
Deutsch 1973; Deutsch 1980). Extended by Deutsch and Krauss (1960) and Thompson
(1967), Deutsch’s theory has resulted in the growing interest to extend the pool of task
characteristics as determinants of cooperation, such as task dependence, task complexity
and task visibility (Van de Ven et al. 1976; Itoh 1994). In the following section, these
task-related variables will be discussed in detail.

3.2.1.1. Task dependence: The extent to which project team members are dependent on
each other to perform individual tasks

Task dependence refers to the extent to which team members are dependent on each other
to perform individual tasks (Van de Ven et al. 1976). Among all the task-related factors,
task dependence is considered to be most correlated to the level of cooperation in
construction settings (Pinto et al. 1993; Thompson 2003). Because of the interdisciplinary
of construction projects and iterative nature of construction activities, information of a
construction project has to be synchronized and activities have to be coordinated well (Jin
and Levitt 1996; Maheswari et al. 2006). Otherwise, it might produce mistakes
necessitating rework and creating crises (Loch and Terwiesch 1998; Kazanjian et al. 2000;
Thompson 2003). Therefore, in typical construction projects, task dependence of

62

operational and managerial activities is extremely important (Winch 1989; Baccarini 1996;
Chinowsky et al. 2008).

The construction literature has recognized the importance of task dependence to be a
moderator for cooperation enhancement intra- and inter- construction project teams (Jin
and Levitt 1996; Levitt 2007; Chinowsky et al. 2009; Taylor et al. 2009; Chinowsky et al.
2010; Unsal and Taylor 2010). For instance, Unsal and Taylor (2010) found that task
dependence can improve innovation adoption in a project network, and encourage the
long term relationship between general contractor and subcontractors. Jin and Levitt
(1996) suggested that managers’ coordination efforts are highly dependent on the types
and extent of task dependence. Thompson (1967) grouped task dependence into three
types -- pooled, sequential and reciprocal -- with reciprocal dependence at their highest
intensity of interaction (Figure 3.2). Regarding construction as a complex system
(Bertelsen 2003b), reciprocal task dependence is probably the most common dependence
in construction project teams (Thompson 2003). Because reciprocal task dependence
means the highest level of interaction intensity (Thompson 1967), intense coordination
work is required to adjust the efforts of different actors (Levitt 2007). In his seminal work,
Thompson (1967) theorized that the correct way to get people working together
effectively is to structure respective work tasks by intensity of dependence, and then
manage each of those dependencies with different coordination methods. For example, a
pooled dependency requires standardization in rules and operating procedures, while the
coordination methods for the other two dependencies are slightly more flexible.
63

Especially for reciprocal dependence, “mutual adjustments” between each other and
constant “information sharing” are needed.
A

B

C

A

B

C

Team outcome

Team outcome

Pooled task
dependence
• Each organizational
department or
business unit
performs completely
separate functions
and then pools the
results together;
• Intensity of
interaction: loosest

Sequential task
dependence
• One unit produces an
output necessary for the
performance by the next
unit;
• Intensity of interaction:
moderate

Team outcome
A
B

C

Reciprocal task
dependence
• One unit’s output is
another’s input, with
tasks occurring in a
cycle
• Intensity of interaction:
intense

Figure 3.2 Three types of task dependence; self-made based on (Thompson 1967)
As to the measure of task dependence, Pennings (1975) pointed out that the measure of
task dependence should cover four basic examples of interconnectedness between people:
task (the flow of work between actors), role (the position of actors engaged in concerted
action), social (mutual needs or goals of actors) and knowledge (the differentiated
expertise of actors). Thompson (1967) defined dependence in terms of work flow and
suggested that it be measured by focusing upon the flow of work, materials and objects
between unit personnel. Building upon Thompson, it was induced that the hierarchy of
increasing levels of task dependence between unit personnel can be determined by
observing whether the work flow is (1) independent, (2) sequential, (3) reciprocal, or (4)
64

in a team arrangement (Van de Ven et al. 1976). Yilmaz and Hunt (2001) proposed
measuring task dependence by the information need of tasks, i.e., whether additional
information is needed to perform a particular task.

Following the previous works, task dependence is a factor that is particularly important to
the cooperation in construction. This research describes task dependence in construction
project teams as the workflow relationship between team members. Particularly, such
workflow is mostly described by network techniques, such as activity on node (AON).

3.2.1.2. Task uncertainty: The variability and difficulty to perform a particular task

Task uncertainty is another critical determinant of construction project teamwork. Winch
(1989) investigated the roles of task uncertainty in construction organizations’ intention
to contract with other organizations from the perspective of transaction cost. He suggested
that there are four types of uncertainties happening in construction organizations: task
uncertainty, natural uncertainty, organizational uncertainty, and contracting uncertainty.
These uncertainties, commented by Winch, force construction organizations to contract
construction services to other construction organizations to reduce potential costs, instead
of employing the capacity to provide these services themselves (Winch 1989). Pich et al.
(2002b) described task uncertainty as the insufficiency of project information, i.e., the
probability of contingencies happening in a project activity network. They found that
managers must evaluate the availability of project information, and to configure a
supporting project infrastructure, i.e., planning, coordination and monitoring systems.
65

Whitley (2006) studied task uncertainty and complexity in project-based firms (PBFs),
such as construction organizations, and found when task uncertainty increases, detailed
coordination of tasks by a formal control system is needed, since the teams are hardly able
to handle the increasing uncertainty by themselves. Also, teams need to commit enough
resources when tasks have considerable degrees of uncertainty.

Task uncertainty refers to the variability and difficulty into performing a particular task
(Van de Ven et al. 1976). Following the literature (Van de Ven et al. 1976; John and
Martin 1984; Moenaert and Souder 1990), it has been mostly measured by the level of
repetitiveness and complexity. Van de Ven et al. (1976) found that the level of task
uncertainty can affect the use of cooperation modes within an organization: if a task is
analyzable and non-variable, most task activities can be performed using standardized and
programmed impersonal means; in contrast, an increase in the degree of task uncertainty
for an organizational unit is associated with a greater use of the personal or group
coordination.

Following the above notions, task uncertainty can be measured as the level of difficulty
and the information need to process a task. If a task is more difficult to process, or
requires more information, it is considered to be more uncertain.

66

3.2.1.3. Task visibility: The extent to which the environment allows for evaluation and
monitoring of individual performance in a team

Task visibility is another topic relevant to the study of people in cooperative relationships.
For any task, the level of visibility refers to the extent to which the environment allows for
the evaluation and monitoring of individual performance in a team (Davis 1985). It can be
measured by the extent to which a task’s outcome affects others’ work, i.e., the relevance
of a task (Moorman et al. 1993). George (1994) studied the effects of task visibility on
people’s cooperative behaviors, and found if individual contribution to a task is hardly
identifiable, people tend to decrease their efforts toward working in a group. Such “social
loafing” or “free riding” behaviors have been well supported and summarized by a variety
of researchers (Souder 1981; Allen 1984; Davis 1985; Moorman et al. 1993; Wagner III
1995).

Fan et al. (2007) studied the free riding phenomenon in project teams; their findings
indicate that if team members found their efforts to be indeed dispensable, they tend to not
work at full capacity and make only limited contributions by simply relying on other team
members to finish the tasks. As a result, many scholars suggested using information
system to improve task visibility in construction projects (Ballard and Howell 1998;
Soltani et al. 2002; Bowden et al. 2006; Xue et al. 2007). In addition, the quality control
procedures in most construction project teams can also affect task visibility. Thomsen’s
work (2005) suggested that construction project teamwork is mainly quality driven, in a

67

sense that quality of project tasks are strictly monitored in most cases. For example,
manager of the estimating team may compare the estimates submitted by estimators
against historical records. Such quality control procedure can significantly affect the
cooperative processes among project team members. Therefore, task visibility is another
important factor particular for construction project teams.

Following the above notions, task visibility is highly associated with the evaluation and
monitoring system in a project team. This research considers task visibility as the
accumulation pattern of work mistakes, i.e.., the extent to which mistakes are monitored
and detected by the team. It depends on the quality monitoring procedure of a construction
project team. If a team has a stricter quality monitoring procedure, tasks have higher level
of visibility and vice versa. It is worth noting that the evaluation of mistakes is on the
basis of managers’ judgment, considering that managers are the evaluators of construction
team members’ work outcomes.

3.2.2. Individual characteristics
Since Lewis (1983) proposed the theory of personality, most scholars in organizational
science and psychology have recognized people’s behavior to be a function of both the
environment and individual characteristics. The literature has accepted the view that
personal characteristics are important determinants for predicting behaviors (Peters 1986).
In terms of cooperative behavior, Argyle (2009) found that people possess different
dispositions that affect cooperation. Some would like to place higher priority on working

68

with others for mutual interest and common goals, while others prefer to maximize their
own interest regardless of others' welfare. A closer examination reveals that such
difference stems from personality traits, roles at work, and demographic characteristics
(Yilmaz and Hunt 2001).

3.2.2.1. Personality: Emotional, attitudinal, and behavioral response patterns

Personality is the behavioral patterns of an individual to the environment which is
particularly affected by emotional and attitudinal factors (Tellegen 1991). It has long been
used as a measure of personal cooperativeness. For example, Greenberg and Baron (2010)
used three personality types to distinguish personal cooperativeness, including
"cooperators," "competitors," and "individualists." The big five personality is the most
widely used framework to predict personal cooperativeness (Taylor 1911). It suggests that
five fundamental personality traits (Fayol et al. 1929), i.e., agreeableness, extraversion,
emotional stability, conscientiousness and openness, can affect personal cooperativeness
to different extents. For example, Ross et al. (1911) found agreeableness and extraversion
to be significant related to individual cooperativeness, while openness and
conscientiousness are least helpful in distinguishing among people's differences. Despite
the disagreements among researchers, the positive relationship between agreeableness and
cooperativeness has been supported by most literature (Taylor 1911; Allen 1984; Pinto et
al. 2009; Nakata and Im 2010). Evidence indicates that people featured in agreeableness

69

are more cooperative, trusting, and tolerant, and are more supportive to other team
members (Nakata and Im 2010).

In construction literature, the influence of personality on cooperation is evident: using Big
Five personality traits theory, Liu and Zhai (2010) investigated the relationship between
project team members’ personality and their ability to handle conflict. The findings
indicate that constructive solutions to technical conflict incidents are always built on
cooperative actions and compromising styles, which mainly come from the personality
trait of “extraversion”. Since conflict is almost inevitable in project based teams, it is
suggested that “extraversion” is an important personality trait in selecting project team
members to encourage fast and effective solutions for any conflict outcomes. Also
employing the Big Five personality framework, Yiu and Lee (2010) found that particular
personality traits, such as extraversion, openness, and conscientiousness can significantly
moderate the relationships of negotiating behaviors and negotiation outcomes when
construction disputes happen. A similar conclusion has been obtained by Cheung et al.
(2006), and Kaman and McCambridge (1992). Other evidence also indicates that there is a
direct link between personality of team members and the performance of construction
project teams through the improvement of communication, trust, and interdependence
(Kuprenas and Nasr 2000; Jesus and Vorster 2002; Varvel et al. 2004). Therefore, a
growing body of construction literature has begun to investigate how to leverage the
personality of team members to enhance cooperation and improve the performance of
construction project teams (Johnson and Singh 1998; Miller et al. 2000; Lawless 2001;
70

Dikmen and Birgönül 2003; Fleetham and Griesmer 2006; Giritli and Civan 2008; Singh
and Eng 2009).

Following the above literature discussions, this research models personality traits of team
members as an important component of cooperation in construction project teams. These
personality traits, will be represented as the emotional, attitudinal and behavioral response
patterns of team members.

3.2.2.2. Roles: Hierarchy position in an organization

Situational factors are another important determinant of cooperation (Chatman and
Barsade 1995). Pennings’s work (1975) suggested that the position of actors engaged in
activities should be considered when analyzing the effects of personality. Chatman and
Barsade (1995) found that the match between personality and organizational context (e.g.,
culture) explains the origin of cooperation.

Project Management Institute (2008) has suggested that there are three major roles in most
construction project teams. Furthermore, the difference between different roles’
responsibilities has been highlighted (PMI 2008). For example, a project manager is
responsible for the success of a construction project, but his/her staff members may be
only interested in particular crafts. As a result, there is a divergence between different
roles in terms of their responsibilities and targets. Such divergence leads to the need of
cooperation. In other words, roles of team members in a construction project team are
directly related to the interpersonal cooperation. In response to these findings, VDT’s
71

modeling work distinguishes the actors’ role in the project team by demonstrating an
organizational hierarchy (Levitt 2007). This research provides a way to represent the
divergence of team members’ roles by representing their position in an organizational
hierarchy.

3.2.2.3. Demographic factors

Compared to the conclusive findings pertaining to the impacts of personality on individual
cooperative behaviors, functions of demographic factors are still arguable (Yilmaz and
Hunt 2001). Although some demographic differences are significant predictors of several
cooperative behaviors (Gouldner 1954), scholars have failed to agree on conclusions. For
example, Lu and Argyle (1991) found a negative influence of age on cooperative
behaviors, while Wagner (1995) reports a positive correlation between the two. This
situation, on the other hand, validates the need for a more careful examination of the
relationships between demographic factors and individual cooperativeness. Even though
disagreement still exists, the literature has already identified the important demographic
predictors of cooperativeness, such as experience, education, and organizational tenure
(March and Simon 1958; McCann et al. 1981; Ruekert and Walker Jr 1987b).

The evidence of demographic factors’ influence on construction project teams is
accumulating, according to recent researchers: Shan (2010) and Dabke et al. (2008) found
that demographic factors can significantly affect the work-life satisfaction of construction
project team members; McCabe et al.

(2005) revealed that individual demographic
72

factors exerts different influence on project team members’ attitudes towards the
construction risk. Singh and Jampel (2010) conducted a survey of engineers in a Public
Works Department (PWD), in order to examine difference in their leadership capacity. It
was found that the decision making capabilities of these engineers were divergent across
different demographic variables, such as age group, years of experience, family
encouragement, etc. Davis and Songer (2009) developed an index named RTCI to show
the magnitude of resistance to change. Their findings indicate that RTCI is significantly
affected by the demographics of team members in AEC organizations. In particular,
different demographic groups (differentiated by profession, experience or even gender)
perform quite differently in accepting changes occurring in their organization.

According to the above evidence, this research considers individual demographic
variables of team members (e.g., years of experience) as important independent variables
for studying cooperation in construction project teams.

3.2.2.4. Other individual factors: findings of CSF studies

Moreover, this research also examined the findings of CSF studies to gain deeper insight
into the importance of individual characteristics. The reason for doing this is that the aim
of this research is to study project team performance; as a result, factors that are not only
considered critical for cooperative behaviors but also important for team effectiveness
should be regarded as within the scope of this research. Therefore, factors that are highly
related to work performance, i.e., critical successful factors, are considered (Pinto and
73

Slevin 1988; Belassi and Tukel 1996; Chua et al. 1999; Westerveld 2003; Fortune and
White 2006). It has been found that project participants’ individual attributes such as
competence, skills, knowledge, experience, and references, may significantly affect the
performance of construction management and operation work (Bentley 1981; Graham
1989; Jaselskis and Ashley 1991; Belassi and Tukel 1996; Chua et al. 1999).

3.2.3. Relational behaviors
Relational behaviors are those that cause team members to develop mutually beneficial
and trustworthy relationships (Yilmaz and Hunt 2001). The literature has identified
certain common behavioral patterns between individuals that are believed to enhance
interpersonal cooperation (Laughlin 1978; Tjosvold 1988; Clemmer et al. 1998; Hoegl et
al. 2004). In particular, Clemmer et al. (1998) have revealed five behaviors between
individuals that may contribute to the formation of cooperation: (1) sharing purpose; (2)
negotiating agreement; (3) encouraging discussion and diverse viewpoints; (4) developing
safe and open environment; and (5) maintaining fairness and equity. In general, the
evidence provided by the literature is extremely consistent in terms of the relational
behaviors that contribute to interpersonal cooperation. On the basis of an extensive
literature review from both the general and construction perspectives, this research
identifies several fundamental relational behaviors that are commonplace in construction
project teams. They are communication, coordination, trust-related activities, reciprocal
activities and meetings. In the rest of this section, these behaviors will be discussed in
detail, with descriptions about how to model them in the proposed model.
74

3.2.3.1. Communication: Exchanging project task-related information

Communication is the activity of exchanging information (Balliet 2010). Communication
is particularly important to construction project teams. A project is a temporary
organization involving the efforts of different disciplines. To achieve project goals,
intensive information should be exchanged among different project parties. For example,
cost estimating team needs project specification from the owner to develop primary and
engineering estimates. Project control team provides feedbacks of project progress to the
schedulers so schedules can be updated to meet anticipated deadline. As a result,
following Levitt’s work (2007), project team work is fundamentally information exchange.
The positive relationship between communication and cooperation in construction
projects has long been proven by the literature. As early evidence, Sally’s (1995) metaanalysis concluded that communication increases cooperation by 40 percent. After that, a
number of construction studies has identified the positive influence of communication
exerted on cooperation in construction project teams. Balliet (2010) performed a metaanalysis on the relationship between communication and cooperation, and found
communication can significantly improve cooperation in construction project teams.
Migliaccio and Martinez (2010) suggested that the cooperation, communication, and
coordination between tribe and the transportation agency are interdependent. Chen and
Tien (2007) designed a peer-to-peer communication network for enhancing the design of
projects. It was found that such communication can enhance data exchange and
integration. The concurrent cooperation can speed up if communication between designers
75

is well handled. Through a survey, Wong et al. (2005) found that open and frequent
communication may maintain cooperation between construction partners in a partnering
relationship. Malcurat et al. (2000) found that synchronous communication is necessary
for fostering cooperation in small scale projects.

To model communication activities, it is necessary to review how the literature describes
communication. Shohet and Fryman (2003b) studied the communication patterns between
construction managers, and found that the content of the communications are construction
instructions, materials and equipment related issues, quality management, allocation of
manpower and cost control. The first two topics represent technological issues, accounting
for 41% of total communication, followed by allocation of manpower (30%). This
indicates that the communication happening in construction project teams are mostly
related to direct project tasks, i.e., technological and human resource issues.

Another significant work is that done by Chinowsky et al. (2009). When modeling
communication, they used the data of information and knowledge exchange frequency or
density to measure the amount of communication. The information and knowledge in their
work are both project task-related, since any project activity requires a transfer of
information (Chinowsky et al. 2009). Moreover, they described the information exchange
as a two-directional process: one direction is from receiver to sender, i.e., a member seeks
a set of key individuals from whom information will be obtained for the assigned tasks;
the other direction is from sender to receiver, i.e., a member provides information (either

76

sufficient or insufficient) to the requester to assist him/her in finishing his/her tasks
(Chinowsky et al. 2008).

Jin and Levitt (1996) modeled communication as a work item representing three functions:
information exchange, exception, and decision. The information processors (i.e., team
members associated with the generation and processing of these communication work
items) exchange messages along formal channels of communication. In case the actor
decides to reply to the message, he/she must process the information using a certain
amount of time. But if the actor decides not to respond, the exchanged message will be
removed from the system and the model will add 1 to the number of non-attended
communications.

The previous research efforts provide a rationale for the modeling work in this research.
Following these works, this research models the communication as activities of
exchanging information. Commutation is needed when a particular project related task
cannot be processed without additional information. Finally, this research also models
communication as a combination of two directional activities (sender-to-receiver and
receiver-to-sender).

3.2.3.2. Coordination: Activities to maintain the consistency of work flow

The meanings of coordination and cooperation are different. Coordination is the
management of the interdependencies of tasks between individuals to maintain the
consistency of workflow; while cooperation is the process and behaviors of an individual
77

to voluntarily contribute to the common goal (Consoli et al. 2007). The literature has
suggested that coordination may improve cooperation between individuals and
organizations, since it promotes information exchange, enhances communication, fosters
trust, resolves disputes, and maintains relationships (Trevino et al. 1987; Consoli et al.
2007; Shen and Chang 2010).

Coordination is another important factor for construction projects and project teams.
Levitt (2007) found that coordination work in a project team arises from two natural needs
of construction projects: the need to resolve project task interdependencies and the need to
handle exceptions. First, dependencies between the tasks are actually created by the
division of effort which generates a need for coordination, i.e., additional communication
and corresponding decision making for the completion of project tasks. Second, it was
found that when a team member is confronted with a task for which she/he does not
possess the requisite skills or experience an exception happens. In that situation,
supervisors need handling processes to deal with any exceptions that have been generated.
This process takes time and results in so-called coordination costs. Levitt (2007) also
provided examples of the consequences of coordination work: team members may be
required to partially or completely rework activities that generate exceptions, may need to
respond to communications from other actors and may need to attend scheduled meetings.
These activities increase the amount of total work that must be done to complete a project.

78

A number of other construction studies have also identified the elements of coordination
work in construction projects. Trevion et al. (1987) pointed out that the coordination
work of a manager includes interpreting the environment, resolving disputes, setting
objectives, making decisions, distributing rules, and instructing. De Saram and Ahmed
(2001b) determined a list of coordination-related activities (64 coordination activities) in
construction projects. The coordination work, according to their findings, can be grouped
into

the

following

categories:

providing

leadership,

facilitating,

controlling,

communicating, and recording; each of these works can be further divided into
sequencing work, deploying work, remedying work, etc.

A more detailed description of coordination work in construction project teams was given
by Shen and Change (2010), who suggested that goals of coordination are: (1) Instruction:
The manager gives orders or provides rules such as procedures or communication
channels that the contractor is expected to follow. For example, the owner asks the
contractor to submit an acceleration plan at the progress meeting. So the meeting’s goal is
classified as instruction; (2) Clarification: Exchanging ideas and clarifying issues such as
arguments with residents or falling behind schedule. For example, contractors hold a
coordination meeting with residents to discuss compensation for a building collapse. The
meeting’s goal is clarification; (3) Facilitating: Doing things that are helpful to executing
a project, such as gathering information about the contract requirements, applying good
technical practices, rescheduling the sequence of site work, expediting the purchase and
delivery of materials. For example, a contractor arranges activities in the construction plan.
79

The construction plan’s goal is facilitating; (4) Controlling: Carrying out activities to
ensure that the schedule, safety or level of quality meets requirements. For example, an
engineer inspects the steel size in the site visits so the goal is controlling; (5) Sharing
information: Distributing information such as monthly reports or meeting minutes to other
parties. The goal is to share information; and (6) Maintaining relationships: Contacting
others for emotional connection or enhancing understanding. For example, engineers meet
and chat casually through informal discussion so the goal is maintaining relationships.

Following the above notions, this research describes coordination as the activities required
to maintain the consistency of work flow. That is, in construction project teams, this
includes to instruct and assign project related tasks, to clarify and approve the purpose or
content of the tasks, and to solve conflicts to make progress. Considering that the
information sharing function of coordination has been covered by the communication
function as discussed above, this research will not consider it as part of coordination.
Similarly, the exception handling function of coordination (as suggested by Levitt 2007)
is covered by another relational behavior (reciprocal activities to be discussed later) and
therefore, it will not be considered here.

3.2.3.3. Trust: Believing others will perform their duties without supervision

A widely accepted definition for trust is given by Mayer et al (1995), which reads: trust is
“the willingness of a party to be vulnerable to the actions of another party based on the
expectation that the other will perform a particular action important to the trustor,
80

irrespective of the ability to monitor or control that other party.” In a construction
environment, trust is simply defined as the confidence and belief in the capability and
willingness of others to perform their duties without supervision (Girmscheid and
Brockmann 2010).

The role of trust in enhancing cooperation between individuals has long been identified by
the literature (Bonoma 1976; Bennett and Jayes 1995; McAllister 1995; Kadefors 2004;
Pinto et al. 2009; Girmscheid and Brockmann 2010). Trust can improve cooperation by
encouraging more coordination of efforts, more contributions per member, and more
achievement pressure (McAllister 1995); by creating norms of cooperation (Korczynski
2000) and minimizing transaction costs under conditions of uncertainty and high asset
specificity (Bromiley and Harris 2006); and also by building the team with reduced
opportunism (Girmscheid and Brockmann 2010). The growing literature on trust research
in construction generally supports these findings (Bennett and Jayes 1995; Munns 1995;
Zineldin and Jonsson 2000; Wood et al. 2002; Bijlsma and Costa 2003; Kadefors 2004;
Shek-Pui Wong and Cheung 2004; Diallo and Thuillier 2005; Wong et al. 2005; Khalfan
et al. 2007; Pinto et al. 2009; Girmscheid and Brockmann 2010).

Trust is also a critical factor to construction project teams because of the interdisciplinary
nature of construction projects. Although a number of studies have provided the evidence
of the antecedents of trust or how to build trust in a project team (Moorman et al. 1993;
Morgan and Hunt 1994; Bennett and Jayes 1995; McAllister 1995; Zineldin and Jonsson

81

2000; Kadefors 2004; Khalfan et al. 2007), the aim of this research is not investigating
these factors, but modeling trust in construction project teams and its consequences
regarding cooperation between team members. Therefore a look into the objects and
elements of trust is needed. Mayer et al. (1995) found the common elements of trust
include:

•

“willingness of one party to be vulnerable to the actions of another party”;

•

“reasonable confidence of the trustor that the trustee will behave in a way
beneficial to the trustor”;

•

“risk of harm to the trustor if the trustee will not behave accordingly”; and

•

“the absence of trustor's enforcement or control over actions performed by the
trustee”.

These elements indicate a belief between each other that work will be done in a mutually
beneficial way without additional enforcement or control. This finding has been supported
by the construction literature: On the role of trust in construction project teams,
Girmscheid and Brockmann (2010) found its most significant contribution to be
“minimizing the energy and time for controls”. They observed that trust does not
eliminate conflicts between team members directly, because conflicts arise more from
divergent interests and foci of different groups or individuals than from mistrust; rather,
trust functions as a way to reduce additional energy and controls and therefore brings
substantial net time savings. Consequently, the team is built up because team members

82

forgo opportunistic behavior in a trustworthy environment. This process could happen
with individuals, teams or even events (Girmscheid and Brockmann 2010).

The above findings provide a clear clue for modeling trust related behaviors in
construction project teams for this research. In conclusion, trust related behaviors are
those demonstrating the belief that team members will perform their duties without
additional supervision or control. In a sequential task environment, that implies an
intention of believing the work delivered by predecessors, while setting fewer safeguards
(Sako et al. 2006).

3.2.3.4. Reciprocal activities: Extra activities generated by reciprocal task dependence

A reciprocal activity demonstrates the situation in which the output of one unit (e.g., a
department or an individual) becomes the input of another, with the addition of being
cyclical (Thompson 2003). In construction project teams, the concept of “reciprocal” is
closely related to task dependence (Levitt 2007).

As discussed in the previous section, the dependence among tasks can significantly affect
the behaviors of team members (Deutsch 1949; Pinto et al. 1993a; Thompson 2003).
Amongst three types of task dependence, reciprocal task dependence requires high
intensity of interactions (Thompson 1967). Jin and Levitt (1996) described these
interactions (related to reciprocal task dependence) as a set of reciprocal activities
between project team members. Using VDT model, they investigated two kinds of
reciprocal activities oriented from reciprocal task dependence: one is information related
83

reciprocal relations, which represent reciprocal activities pertaining to the shared
information requirement. For example, in proposal development activities of a power
plant, the proposal team members may need specific structural information from the
engineers such as specification and diameters of pipes, or volume of concrete. Meanwhile
the engineers might need product information (e.g., size and type of the generator) to
conduct structural design. The other reciprocal activity is work related reciprocal relation,
which describes the situation that an exception generated by one actor brings physical and
actual consequences to the work of another. For the above example, if the structural
design is changed by the engineers, the proposal team must modify or even redo the
estimation; on the other hand, if the proposal team decides to take a different generator,
the engineers might need to redo the structural design. Therefore different levels of
coordination are needed depending on these reciprocal relations (Jin and Levitt 1996).

Following the above notions, and considering that information related reciprocal relations
are covered by the communication activities, this research describes the reciprocal
activities as those generated by the different reciprocal task dependence, especially those
actions that demonstrate the back-and-forth cycle feature, such as handling the additional
tasks returned by colleagues because of perceived quality issues.

84

3.2.3.5. Meetings: an event that clears work-exceptions by enhancing information
exchange and solving problems

Construction meetings are very common in construction project teams. Typical
construction meetings include pre-construction meetings, hustle meeting, safety meetings,
and sub-contractor meetings (Kerzner 2007). A construction meetings is a medium of
communication and coordination (Jin and Levitt 1996; De Saram and Ahmed 2001a;
Levitt 2007). But different from other communication and coordination efforts, meetings
are mostly structured and formal (Jin and Levitt 1996). The functions of meetings in terms
of building a cooperative and effective construction project team have been documented
by the construction literature: Ioannidou and Fruchter (2009) found the proper frequency
of meetings can facilitate the building of an effective global AEC team by allowing the
reuse of knowledge. De Saram and Ahmed (2001b) found conducting regular meetings to
be one of the most common coordination activities, as well as one of the most time
consuming activities, based on a survey of construction managers. Raboud (1988) found
that meetings can improve safety on large scale construction projects.

In the VDT model, Jin and Levitt (1996) modeled meetings as communication and
coordination enhancement. That is, through formal and regular meetings, the actors may
exchange information and reduce informal interpersonal communications to achieve the
efficient finish of work; meanwhile, attending meetings will also waste additional time of

85

the actors. Therefore, the choice of attending a meeting or not is based on the decisional
preference of actors and the organizational context.

Since meetings are regarded as serving similar (but more formal) functions as
coordination efforts, this research sees the consequences and relevant elements of
meetings to be the same as coordination, as discussed above. The literature reveals that a
meeting can clear work-exceptions by enhancing information exchange and solving
problems (Antony 1976; Boden 1995; McCowan et al. 2003; Adams 2004; Tepper 2004).
It is worth noting that, a meeting serves as an “ongoing flow of organizational activities”
which is likely to lead to more subsequent meetings, depending on the coordinating
demands (Schwartzman 1989). Therefore, this research measures the meeting duration as
a dynamic variable that depends on the needs (relative number of management
exceptions). Therefore, the description of a meeting in this research is: an event that clears
work-exceptions by enhancing information exchange and solving problems, with the
duration depending on the relative number of management exceptions (e.g., design change,
error detected, following Jin and Levitt 1996).

3.2.4. Organizational context
Although interpersonal cooperation research has a long history of being aligned with
discipline of psychology (e.g., interpersonal attraction or psychological attachment), the
literature finds it can also be characterized and fostered by institutional factors including
hierarchy, rules and regulations (Hechter 1988; Smith et al. 1995). This raises continuous

86

interest in investigating the effects of the organizational context in forming cooperation.
Organizational context in this research refers to the structures of policies and rules, within
which “an organization arranges its lines of authority, work and communications, and
allocates rights and duties” (BusinessDictionary 2012). It presents “the manner and extent
to which roles, power, and responsibilities are delegated, controlled, and coordinated, and
how information flows between levels of management” (BusinessDictionary 2012).

3.2.4.1. Organizational structure: Report direction and proximity

Among the variety of topics, the role of organizational structure in cooperation receives
the earliest attention. Organizational structure refers to the formal organizational structure
that determines the arbitrary delegation pattern within an organization and reflects the
report directions (Du and El-Gafy 2011). Since Mintzberg (1992) proposed his theory
about organizational structure, numerous efforts have highlighted the impacts of
organizational structure on interpersonal cooperation: Ouchi (1988) and Smith et al (1995)
suggested that a formal organizational hierarchy, such as contractual obligations,
functions as an alternative to socialized control in cooperation between individuals. That
is, by delicate job design and delegation, individuals are forced to work together, whereas
organizational structures detail how departments and groups must function (Itoh 1994).
Anvuur (2008) holds a similar argument that better designed intergroup structures may
help develop and maintain the attitudes and culture for better cooperation on any
construction projects. Relevantly, Hechter (1988) and Wagner (1995) found that group

87

size is an important predictor of within group cooperation; i.e., in smaller groups,
cooperation is more likely to happen. George (1992) explained this phenomenon from the
sense of social loafing: he finds as the group size becomes smaller, it becomes easier to
evaluate individual contributions because they are more “visible and identifiable,” and as
a result, people work hard to avoid free riding and social loafing, and finally, they adopt
more cooperative behaviors.

Some scholars explain the influence of organizational structure from the perspective of
proximity and accessibility. It has been well documented that physical structure of
organizational settings can influence the type of interactions and communications that
occur within and among groups in an organization (Souder 1981; Davis 1985). For
example, evidence indicates that propinquity can lead to enhanced communication among
project team members (Keller and Holland 1983; Allen 1984). Peters (1986) also found
that individuals tend to interact and communicate with others when the physical structure
is encouraging. Instead of solely focusing on physical proximity, individual perceived
accessibility has also been recognized as a critical facilitator of cooperation. Pinto et al.
(1993a) found that if a person perceives a stronger ability to approach or communicate
with another organizational member, she/her tends to interact and cooperate more
frequently. Such a perception of accessibility, as suggested by Pregent (1988), is primarily
affected by the individual’s position in the organization and commitment to others etc.
The above evidence highlights the importance of organizational structure in investigating
interpersonal cooperation in construction project teams.
88

3.2.4.2. Workflow and information flow: organizational procedures

Another important factor of the organizational context pertaining to construction projects
is organizational rules and procedures: Following the definition given by Pinto et al.
(1993a), organizational rules and procedures refer to how project activities or tasks on the
project team are mandated or controlled. The significance of organizational rules and
procedures has been recognized since Taylor (1911) and Fayol et al.

(1929) who

proposed that coordination in an organization is accomplished by establishing rules and
procedures throughout the management hierarchy. Their assertion has been extended by
the efforts of subsequent scholars who posit that rules and procedures form as a
mechanism for integrating or coordinating activities (Gouldner 1954; March and Simon
1958; McCann et al. 1981; Galbraith and Nathanson 1982). For example, McCann et al.
(1981) described rules and procedures as the most common approach for coordinating
activities, controlling behaviors, minimizing cross-functional conflicts, and maintaining
organizational structure. Reukert and Walker (1987b) found that rules and procedures are
useful for establishing cooperation among individuals or departments. Similar findings
have been obtained by more studies that indicates that increased formalization in an
organization creates a more harmonious climate, and exerts a direct influence on the
development of cooperation (John and Martin 1984; Moenaert and Souder 1990;
Moorman et al. 1993; Pinto et al. 1993a; Pinto et al. 2009; Nakata and Im 2010).
Following these pieces of evidence, work flow and information flow are considered as
two important channels for affecting cooperative behaviors in construction project teams.
89

3.2.4.3. Cultural norms: the pattern of shared values and beliefs

Besides the aforementioned two factors (i.e., organizational structure and organizational
rules and procedures), many scholars also recognize organizational cultural norms as an
important contextual variable. Organizational cultural norm, following Deshpande and
Webster (1989), refers to “the pattern of shared values and beliefs that help individuals
understand organizational functioning and thus provide them norms for behavior in the
organization”. Most literature differentiates organizational cultural norms into two types:
collectivism and individualism (Wagner III 1995; Chen et al. 1998; Yilmaz and Hunt
2001). Collectivistic culture values collective goals and joint contributions, whereas
individualistic culture emphasizes the maximization of personal achievements and
interests (Chatman and Barsade 1995). Evidence has highlighted the importance of
collectivism in fostering interpersonal cooperation. For example, findings reported by
Chatman and Barsade (1995) indicated that a collectivist cultural norm will lead to higher
cooperativeness, no matter what personalities the individuals possess. Yilmaz and Hunt’s
(2001) research revealed collectivist organizational norms to be influential on individual
cooperation, which highlights the importance of developing and enforcing collectivist
organizational norms in order for individuals to establish a cooperative environment. A
recent experimental psychology study (Nguyen et al. 2010) also found that collectivism
can facilitate individual cooperativeness in different social cultural context. These
findings indicate a rationale for investigating the differences between cooperative
behaviors of construction project team members under different cultural norms.
90

3.3. Performance indicators of construction project teams
As addressed in previous chapters, although the impacts of interpersonal cooperation on
the performance of construction project teams have been well recognized, evidence
remains anecdotal, subjective and piecemeal (Anvuur 2008). A major work of this
research is to qualify the relationship between interpersonal cooperation and project team
performance using a simulation model. As a result a list of performance metrics must be
proposed as the index of interpersonal cooperation’s effects. Two groups of performance
indicators are suggested below based on a twofold measurement framework in chapter 2,
i.e., outcome performance and process performance.

3.3.1. Outcome performance
A stream of construction scholars has been using construction project team performance
as outcome performance. Ancona and Caldwell (1992) used team performance as the
managers rated quality, cost/schedule performance, conflict resolution ability, and overall
performance. Barrick et al. (1998) argued that team’s current performance is typically
based on the supervisor’s perceived team productivity (subjective ranking) or quantity of
productivity (objective indicators). Tesluk and Mathieu (1999) used supervisor rated
outcome performance for construction and maintenance road crews as an assessment of
team performance.

As another example, Langfred (2000) described supervisors’ evaluation of the accuracy
and quality of the tasks as team performance for military project teams. Horii et al.

91

(2005b) studied the cultural impacts on construction project team performance, and in
their study team performance is referred to as duration, cost, and risks. Kang et al. (2007)
studied the performance of a construction defection detecting team, and used total number
of detected defections, speed and accuracy to measure team performance. Oshinubi (2008)
studied the impacts of manager’s leadership style on a construction project team, and used
team performance for measuring whether the work was finished on time, whether the
work is finished under budget, and whether the work is finished within the profit margin.

Recently, Leicht et al. (2010) proposed an observational research method to study team
performance in construction management. In their study, team performance followed the
definition given by Mathieu et al (2000), which refers to team performance only as the
objective “outcome of team” both in quality and in quantity. Several studies used a
simulation approach to investigate impacts of team and task characteristics on project
team performance which also provided specific team outcome indicators. Wong and
Burton (2000) measured team performance as total task completion time in their research
on virtual teams by a simulation model. Similarly, Kim and Burton (2002) measured team
performance by duration, cost, and quality of projects in a simulation study pertaining to
project teams. Among all the measures, the list of Tommelein et al. (1999) is most
representative: they measure construction trade outcome performance using the following
indicators: capacity -- units of work per unit of time that a crew can complete given
sufficient resource for their work; duration -- how long it takes to finish an entire project;
and productivity -- number of completed work units per unit of time.
92

Following the previous literature, this research measures the construction project team
outcome performance as:

•

Time: how long it takes to finish an entire project;

•

Quality: the number of mistake per unit of work amount that a crew commits after
the entire team-specific work has been done.

•

Team cost (when applicable): the direct and indirect expense in dollar amounts a
crew spent to finish a given amount of work provided their work is unconstrained;

3.3.2. Process performance
Besides outcome performance, the construction literature has also developed a set of
process performance indicators for measuring construction project team performance.
Ancona and Caldwell (1992) used team performance as the manager’s rated efficiency
and the team’s ability to resolve conflicts. Tommelein et al. (1999) applied two indicators
to measure team process performance of the construction trade. One is “buffer,” i.e., the
“work units accumulated ahead of a crew, from which they can draw at will to perform
work”; the other one is “wasted time,” i.e., “time during which a crew is not able to
realize its production capacity due to constraints that hampers their work, which results
in lost productivity”. These two indicators reflect the inefficiency of team work due to
unsmooth work flow. Wong and Burton (2000) paid special attention to team coordination
time, and re-work time as complementary indicators to team total work time, when
studying the performance of virtual teams.

93

The construction literature has also long emphasized the communication, coordination, or
cooperation efficiency and effectiveness between team members as important team
performance indicators (Jin and Levitt 1996; Chinowsky et al. 2010). A recent work,
reported by Chinowsky et al. (2010), regards the performance of construction project
team being built on strong communications between the team members. They used the
efficiency and effectiveness of information and knowledge transfer between team
members as the measurements of project team performance, including weekly and
monthly communication frequency and knowledge exchange density.

One of the representative works in this area is VDT. In 1996, Jin and Levitt proposed a
multi-agent simulation framework for studying the influence of coordination behaviors
and organizational design on the performance of engineering project teams. In order to
evaluate project team performance, they developed a set of indicators for describing the
quality of coordination and work process, which are: relative coordination load, i.e., the
ratio of re-work amount and corresponding coordination work amount to the total work
amount; coordination verification quality, i.e., the extent of total failed work identified as
re-work amount; and process efficiency, i.e., the ratio of actual duration and cost for
finishing a project to the planned time and cost. It is believed these indicators are able to
capture the project team effectiveness and resulting performance (Jin and Levitt 1996).
Following the above notions, this research measures the construction project team process
performance as:

94

•

Team effectiveness: the ratio of time a crew spent on all non-production work to
total work time;

•

Team efficiency: number of completed work units per unit of time; it equals total
work units of a project divided by duration;

•

Work-related pressure (work backlog): measured by the estimated cumulative
unfinished work amount.

•

Exception handling (when applicable): measured by the number of problems being
solved per unit of communication time;

3.4. The conceptual framework
Based on the discussions above, a conceptual framework for integrating interpersonal
cooperation and team performance in construction project teams is proposed (Figure 3.3).
The proposed framework demonstrates the relationship between interpersonal cooperation
and the measurements of team performance. It constitutes the foundation of the simulation
model. According to this framework, there are four major categories of factors
representing or determining the processes

of interpersonal cooperation: task

characteristics, individual characteristics, relational behaviors and organizational context.
Each category contains a variety of factors that have been proven important to
interpersonal cooperation. Interpersonal cooperation in turn determines team performance,
i.e., team outcome performance and team process performance.

95

Team process performance

Team outcome performance

- Effectiveness
- Efficiency
- Work related pressure

- Time
- Cost
- Quality

Team
Performance

Interpersonal
Cooperation

Task
characteristics
- Task dependence
- Task uncertainty
- Rask visibility

Individual
characteristics

Relational
behaviors

- Personality
- Roles
- Demographic
factors

- Communication
- Coordination
- Trust-related
- Reciprocal
activities

Organizational
context
- Org structure
- Work
procedure
- Info flows

Figure 3.3 Conceptual framework of interpersonal cooperation in construction project
teams

However, certain questions regarding this framework remain unanswered. Typically, how
are the pieces of this framework linked together? Robbins and Langton’s (1998) model of
organizational behavior provides a clue to linking the pieces together. Following their
model, an organization is a multilevel system consisting of individuals and affected by the
environment. From a technical perspective, the function of any organization is processing

96

inputs such as information, materials, tasks or other entities related to organizational
functions. During operation, inputs are fed into the organization where relevant
individuals (actors) are involved in the processing of inputs. The efficiency and quality of
input processing are thus affected by the characteristics of the individuals, such as
competence and attitudes (Figure 3.1). In cases when the inputs are highly complex,
requiring diversity of skills or commitments from the actors, a group of actors need to
work together to approach the targets. As a result, team process occurs where the team
function becomes smoother and collective decision making substitutes for individual
decision making (Tuckman 1965). In this case, group structure, communication, conflict
management, and leadership start to emerge and play a critical role (Robbins 2005). The
processes at the group level may further promote the emergence of or be affected by
processes at the organizational level, including policies, organizational structure and
culture (Robbins 2005). Finally, inputs pass through three levels of processes – individual
level, group level and organizational level – and are finally demonstrated as the results
(Figure 3.1).

Robbins and Langton’s model actually reflects an Input- Processing-Output (IPO) point of
view (McGrath 1984) of an organization. It views an organization as a medium where
inputs are processed inside, and performance is presented as the outputs of the processing.
Along with the same perspective, a construction project team can be treated as a
processing system in this dissertation. Figure 3.2 demonstrates the IPO picture of a
construction project team. In particular, its inputs are project related tasks since
97

construction project teams are considered to be project based organizations (PBOs) where
construction projects are the major source of work (Bosch-Sijtsema et al. 2006; Thiry and
Deguire 2007). Total work effort of a project has to be decomposed into smaller chunks of
work, called tasks, in order to be carried out by individual actors or a group of actors
(Project Management Institute 2004). Tasks are processed by the project team and
accordingly, the outputs are the performance of the tasks in general, or team performance
in processing them. In the middle, the constructs of interpersonal cooperation –
individuals, relational behaviors and organizational context – act as the media of
processing. Following Robbins and Langton’s (1998) model, the constructs are
interdependent and interacting, showing a complex internal structure: individuals are the
actors of tasks, taking major responsibility for task processing and directly determining
the quality and efficiency of outcomes; relational behaviors demonstrate specific actions
of individuals on tasks and the connections among individuals; organizational structures
perform as the context of the processes and the flows of tasks and corresponding info.
PROCESS

INPUT

OUTPUT

Relational behaviors
Tasks

Individuals (actors)

Performance

Organizational context
Figure 3.4 An IPO view of the framework
The proposed framework (Figure 3.4) has served as the foundation of an Agent Based
Simulation model that captures interrelationship among components of interpersonal

98

cooperation, and between interpersonal cooperation and the measurements of team
performance. A simulation model was built based on the proposed framework to link the
independent variables (constructs of interpersonal cooperation) and dependent variables
(team performance).

3.5. Summary
This chapter has extended the literature review to investigate elements of interpersonal
cooperation in construction settings. Specifically, the aspects directly related to
construction projects were highlighted including task characteristics and some of
relational behaviors. Performance metrics for construction project teamwork was also
identified. Finally, a conceptual framework for interpersonal cooperation in construction
project teams has been proposed to link elements of interpersonal cooperation to
construction project team performance. This conceptual framework constitutes the
foundation of a simulation model which will be discussed in the next chapter.

99

CHAPTER 4: A BOTTOM-UP SIMULATION MODEL OF INTERPERSONAL
COOPERATION IN CONSTRUCTION PROJECT TEAMS
This chapter presents a simulation platform (Virtual Organizational Imitation for
Construction Enterprises, VOICE) for investigating interpersonal cooperation and its
implications in construction project teams. The method adopted is Agent Based Modeling
to capture the transition from local processes of cooperation to global performance of
project teams in a bottom-up manner. VOICE reflects a multifold level of realization of
the proposed conceptual framework (Chapter 3) with an overall consideration of
construction project tasks, project teams, work processes and individuals.

4.1. Model architecture
Figure 4.1 demonstrates the model architecture of VOICE. The development steps of
VOICE are: (1) Input modeling: identifying and modeling concurrent projects and
corresponding tasks; (2) Process modeling: identifying and modeling the actors, including
individual characteristics and relational behaviors; defining and modeling the
organizational structure, work process and information flow as the context; and (3) Output
measurement: formulating key performance indicators for construction project team
performance. It is worth noting that although existing knowledge has highlighted the
importance of the above model components to be different, this research doesn't assign a
predefined weight to each model component. It is because in a bottom-up simulation the
importance will automatically emerge as a part of the emergent properties (North and
Macal 2007). As a modeler, the only task is to design the bottom rules of the model. The
remainder of chapter discusses the details of each modeling step.
100

Project

Task
INPUT MODELING

President

Manager

Work
process

Org
Structure

Staff

Work
process

Information
flow

Org
Structure

Information
flow

PROCESS MODELING

Performance of project team
OUTPUT MEASURE

Figure 4.1 Model Architecture of VOICE

4.2. Input modeling
Construction project teams are project based organizations, or PBOs (Levitt 2007). In a
construction project team, different functions and personnel are organized around each
particular construction project (Kodama 2007). For example, many construction project
teams have project managers who manage teams of employees. These employees are
101

often from different disciplines and have different job titles, but all are needed to get the
project done. The primary “productive” activity of any construction project team is thus
volume-based or operations-oriented, mainly the processing of project tasks (Sydow
2004). Therefore VOICE models projects and corresponding tasks as the sole input of a
project team.

A project is a multi-task effort that must be completed by a group of actors with each of
them having different characteristics (e.g., size, complexity, mistake percentage, etc.). In
VOICE, a project generates a list of tasks according to the predefined task arrays (a set of
arrays indicating the sequences and attributes of each task). A task is a basic executable
work effort that is assigned to corresponding actors sequentially for processing. For
example, quantity take-off for concrete usage of a foundation wall is a typical task for an
estimator. Even though this task is still dividable, i.e., it can be further divided into subtasks such as “studying the drawings”, “calculating volumes” and “determining concrete
specification”, these sub-tasks are normally executed by a single estimator. Therefore
quantity take-off of a foundation wall can be considered as the most fundamental
executable unit and is modeled as a single task in VOICE. Based on the findings of
Chapter 3, VOICE models project and task attributes that are regarded as crucial for the
performance of a project (Table 4.1 and Table 4.2).

102

Table 4.1 Attributes of projects
Name
ID number
Size
Priority
Complexity
Time limit
Cost limit
Mistake

Value Remark
[1, *] A unique and universal ID number for each project
(0, *] Size of the project, measured as the sum of its tasks’ work
amount
[1, *] An indicator of the project’s priority; the higher the more
urgent
[1, *] An indicator of the work difficulty of a project’s tasks; higher
number means more time for processing its tasks
(0, *) The time requirement of a given project, measured in the ticks
(0, *) The budget of a given project
[0,1]
The final mistake percentage of a project, it’s a function of the
mistake percentage of all its tasks

Table 4.2 Attributes of tasks
Name
Task number
Project label

Value
[1, *]
[1, *]

Work
amount

(0, *]

Priority

[1, *]

Difficulty

[1, *]

Authority
level

1,2,3

Approval

0, 1

Dependence

0, 1

Information
status
Concurrent

0, 1
1..,n

Remark
A universal number (but not unique) for each task
An indicator overridden from project ID number, showing
which project this task belongs to; a particular task cannot be
identified unless both task number and project label are given:
Task.id=task.number & task.projectLabel
The work amount of task, measured as the work hours when
processed by an actor with competence=1. It relates to the size
of the project:
Project.size=∑taski.amount
An indicator of task’s priority, which relates to the priority of
the project; higher priority means the work can jump to the top
of work list. The actual priority relates to the project priority:
task.actualPriority=project.priority*task.priority
An indicator of the work difficulty of a task. The actual
difficulty level of a task relates to the project complexity:
Indicating which position level has the right to process or
approve a give task; for example, task “deciding profit rate” is
fairly high authority work, with authority level of 3
0 means this task has not been approved ;1 means it has been
approved
1 means following tasks are dependent on this task while 0
means not
0 means this task can be processed without more information,
while 1 means more information is needed
An absolute number indicating how many following tasks will
103

Table 4.2 Cont’d
Attributes of tasks
Name
indicator

Value

Mistake

[0,1]

Starting
address

-

Remark
be generated after the completion of this task, the default value
is 1
The final mistake percentage of a task, depending on the
actor’s processing quality
The address of the 1st actor for this task; the task will be
assigned to this actor, and then passed through all relevant
actors.

Figure 4.2 demonstrates the flowchart of agent project. VOICE provides several unique
features that make work modeling closer to reality: (1) VOICE considers a concurrent
working environment, so multiple projects and tasks can enter into the model
simultaneously; (2) VOICE also reflects dependencies among tasks, hence the generation
of every task is strictly dependent on an attribute “task.dependence” (see Table 4.2); (3)
The assigning of tasks builds on real work flow mapping, and the destination of every task
is embedded in the attribute’s “starting address” as well as work process, which is
discussed later (Figure 4.6).

In order to collect the information needed to model tasks, the following steps were
followed. First, relevant project documentation were collected to summarize the
background information, including project specification, job description for each team
member, and other technical documents about the construction projects and project team.
Then, semi-structured interviews and surveys were conducted to determine the proper list
of tasks for a construction projects. Only the daily tasks directly pertaining to the
operations and management of construction projects are emphasized.

104

Start

Task arrays

Read attributes of the ith
task from the task arrays

Any on-going task’s
projectLabel=ID
of this project ?

No

Yes
Project remaining
work >0?
Are on-going tasks required for
successive tasks?

Yes

No

The successive tasks are
concurrent?
No

Yes

Generate a task
Project.size=task.amount

Generate multiple tasks
Project.size=∑task.amount

i=+1

No

i=+n (n is the number of
concurrent tasks)

Remove this
project

END

Figure 4.2 Flowchart of the agent project
105

Yes

Various techniques were employed to enhance the communications between the
researcher and the respondents, such as Work Breakdown Structure (WBS) or Ishikawa
diagram (Figure 4.3). Meanwhile, attributes of these tasks (e.g., duration, difficulty) were
determined by the documents (e.g., time sheets) or surveys (Table 4.3).

Figure 4.3 Using Ishikawa diagram to develop the list of activities

Table 4.3 A sample table head of the online questionnaire

No.

Task

Processing/lead time (hour)
Min
Likely
Max

Priority (0~10)
(10: most urgent)

Complexity
(0~10)

4.3. Process modeling

4.3.1. Modeling individuals
Project teams are modeled as a group of actors (or agents) who are able to execute the
generated construction project tasks. VOICE follows a set of assumptions on the basis of
106

behavioral decision theory (Cyert and March 2005): (1) there are multiple roles inside one
construction organization (e.g., president, manager and staff), and each role can be
represented by an agent; (2) each agent is imbued with bounded rationality; this moves
toward more specific and particular work aims instead of following a common aim for the
whole organization.; (3) each agent occupies a particular position inside the organization
that defines what task(s) the agent does, and with whom the agent interacts. As a result,
the authority position, work arrangement and formal/informal network determine the
agent’s behaviors; and (4) each agent possesses specific knowledge, skills, and
capabilities.

VOICE models three types of actors in a construction project team following the
construction literature (PMI 2008). Project Management Institute (2008) describes most
construction organizations with a traditional hierarchical structure, and identifies three
levels of roles in any construction project teams: The top level is President or Vice
president of a construction project team, whose responsibilities include scope
management of projects, strategic decision making pertaining to the key points of projects
and communication of project performance. They work on the organizational level, i.e.,
are interested in the success of the entire project team. In the middle are project managers
or team coordinators, who are responsible for construction information integration of
different disciplines, coordination of construction project tasks, and conflict management.
They work on the project level, i.e., are responsible for the success of particular
construction projects. The lowest level is staff, which is responsible for the most specific
107

tasks, such as the estimation of a particular craft (e.g., concrete, steel and piping),
structural design, scheduling and operations of construction sections. Staff members work
only on the craft basis.

Following the findings in Chapter 3, attributes of three types of actors are modeled as
Table 4.4. In order to represent these characteristics, the following tasks were performed:
(1) Document research: the company documents, such as job description, were studied to
obtain information about individual position, job responsibility, and demographic
characteristics. Especially, information pertaining to the success of construction projects
is highlighted. Typical questions include: the specific responsibility of a team member in
the project team (e.g., estimating, project controlling), years of experience as a project
coordinator and others. (2) Surveys and semi-structured interviews: a set of surveys and
interviews was conducted with each employee so that information about personality could
be collected. Typical questions include: the responses to unclear project task requirements
(e.g., incomplete drawings and vague project specifications for estimators), the number of
communications between a planning team member and an engineer and others. (3)
Summarizing: finally, all collected data about individual characteristics were summarized
and conceptualized to a representation model. Table 4.4 illustrates a list of modeled
attributes.
Table 4.4 Attributes of actors
Name
Value
Competence (0, *]

Remark
Note
An indicator of actors’ work efficiency, measured Generic
with the processed work amount in unit time.
attribute
108

Table 4.4 Cont’d
Attributes of actors
Name
Exception
handling
time
Capacity

Value
(0, *]

Work
quality
Authority
position

(0, *]

(0, *)

1,2,3

Assigning
preference

1,2,3

Quality
preference

1,2,3

Quality
threshold

(0, 1)

Exception
indicator

(0, *]

Exception
threshold

[1, *]

Salary rate

(0,* ]

Remark
An absolute quantity indicating the time needed to
deal with exceptions (emerged tasks, e.g., staff’s high
work related stress, work quality issues).
The most acceptable total work amount for an actor

Note
Specific for
manager and
president
Specific for
staff
A relative quantity indicating the processing quality, Specific for
measured in 1- percentage of processing mistakes.
staff
An indicator of the actors’ position levels: Generic
president=3, manager=2, staff=1. This attribute is attribute
related to the task authority level: if
task.authority>actor.authority, then the actor will ask
for approval from higher authority position.
When a task is assigned to the staff (only applicable Specific for
for manager):
manager
1 means assigning tasks to the staff who can finish
the task in the shortest time; 2 means assigning tasks
to the staff that can finish the task with the best
quality; and 3 means assigning tasks to the staff with
the smallest work pressure.
When a task delivered/submitted to the manager has Specific for
quality issues:
manager and
1 means returning the task to who processed this staff
task; 2 means reassigning the task to other
subordinates that can finish the task with the best
quality (only applicable for manager); and 3 means
correcting the mistakes by self.
An absolute quantity indicating the quality threshold Specific for
of an actor; if the delivered/submitted task’s mistake manager and
percentage is bigger than the threshold, then the actor staff
will take action based on quality preference.
An indicator monitored by the president showing the Specific for
total management exceptions happening in the president
organization.
An absolute quantity indicating the exception Specific for
threshold of the president; if the exception indicator president
is bigger than the threshold, then the president will
set up a meeting to address these exceptions.
An absolute quantity showing the tick-based (tick is Generic
the basic time unit in VOICE) salary rate for the attribute
actor

109

4.3.2. Modeling relational behaviors
In this research, relational behaviors are represented as the most fundamental and generic
activities of an actor that constitute his/her daily cooperation related actions pertaining to
the operations and management of a construction project. Individual relational behaviors
are different from a list of tasks: relational behaviors are abstract and conceptual activities
that an actor might take to finish one or more tasks. For example, to evaluate
subcontractors' quotes (a task with a specific target), an estimator may need to
“communicate” and “coordinate” (work-related relational behaviors that are needed for
fulfilling the task). Table 4.5 shows an example that finishing one task requires the actor
to assume multiple relational behaviors.

Table 4.5 Relationship between tasks and relational behaviors: an example
Behaviors CommuniTasks
cate
Study potential opportunity
√
Contact subcontractors
√
Price self performed work
√
Evaluate sub quotes
√
Finalize & Submit the bid
Add failed bids to archive

Coordinate

Reciprocal
activities

√
√
√
√

√

Considering the difficulty of summarizing and conceptualizing individual relational
behaviors, the following efforts were made: (1) Online surveys: the developed list of
behaviors was ranked or modified by the stakeholders, so that the size of the list could be
further narrowed down. (2) Semi-structured interviews: interviews were conducted with
key employees so that differences between individuals could be highlighted. Also, the

110

explicit information of individual decision making processes and preferences could be
documented. (3) On-site observations: the activities of each actor were observed and
recorded on a daily basis. These activities, done simultaneously, were categorized by the
author on site into different types of relational behaviors. (4) Summarizing: finally, the
relational behaviors were summarized and categorized to a definite set list. The complex
daily activities of each actor can be regarded as the result of the modification and
combination of the behaviors in this list. Table 4.6 shows identified relational behaviors
and corresponding specific actions modeled as “behavioral modules” in VOICE. To be
noted, these behavioral modules can be grouped into two clusters: generic behaviors,
those are commonly observed in organizations other than construction, and construction
specific behaviors, those are critical to the operations and management of construction
projects. Amongst, routine work represents the generic operational activities in most
organizations and therefore is considered as a generic behavior in VOICE. Trust related
behavior refers to the monitoring of quality which is commonly seen in other types of
organizations and thus is also generic. Reciprocal activity is another generic component in
VOICE because it is related to the reciprocal interdependence of tasks which can be
observed in other types of organizations in manufacturing or consulting industry (Kodama
2007). The coordination and communication behaviors modeled in VOICE are
construction specific, since the actions considered in the simulation are those directly
associated with construction projects.

111

Table 4.6 Set of fundamental relational behaviors and specific actions
Relational
behaviors
Routine
work
(generic)

Definition

Specific actions

Description

The most
generic and
daily
activities
associated
with the
direct
finishing of
tasks

Prioritizing

Comparing the priorities of concurrent
projects as well as relative priorities of tasks
to make a work plan where most emergent
task jumps to the top of work list
Directly working on specific task.
Processing tasks is demonstrated as the
reduction of remaining work amount. Work
time and quality depend on competence and
work quality and task difficulty represent
the reduction of tasks’ work amount
Submitting
the
finished
tasks
to
supervisor/co-workers
The manager/coordinator/president assigns
tasks to different subordinates based on
‘assigning preference’
Certain tasks need to be approved by higher
management lines before processing, e.g.,
“determining profit margin”. This activity is
considered to be a management exception
which affects work effectiveness and
efficiency.
Management approves the tasks per the
request of subordinates. This activity is
considered as a management exception
which affects work effectiveness and
efficiency.
The manager/coordinator works on solving
the exceptions attributed to conflicts. For
example, the proposal team coordinator
works with the engineering team
coordinator to clarify information for
proposal development.
Certain tasks need extra information from
other actors before processing. Hence the
communication is shown as changing of
tasks’ information status in VOICE. This
activity is also considered as a management
exception.
An event that clears work-exceptions by
enhancing information exchange and
solving problems. Meetings aim to
eliminate management exceptions. They can
enhance the communication, information

Processing

Submission
Coordinati
on
(constructi
on
specific)

Activities to
maintain the
consistency
of work flow

Assigning

Requesting
approval

Approving job

Conflict
management

Communic
ation
(constructi
on
specific)

Exchanging
project taskrelated
Information

Information
exchange

Meeting

112

Table 4.6 Set of fundamental relational behaviors and specific actions
Cont’d
Relational
behaviors

Definition

Specific actions

Social contact

Trustrelated
(Construct
ion
specific)

Reciprocal
(generic)

Believing
other team
members will
perform their
duties
without
supervision
Extra
activities
generated by
different
reciprocal
task
interdepende
nce

Monitoring
quality

Reporting
overburden

Correction/rewor
k

Description
sharing and other work related problems.
The duration is determined by the number
of management exceptions.
Informal communication and connections
between team members. It is not modeled in
this dissertation considering the scope of
current study.
Comparing
the
quality
of
delivered/submitted work with own quality
threshold, and if quality is not satisfied,
work will be reworked, returned, reassigned
or reported based on “quality preference”.
Unsatisfactory work is also a management
exception.
If the total work amount of assigned tasks is
over the capacity of actor, he/she may
suspend the work and report work related
overload to upper level management. The
management, based on different preference,
may return the work or reassign the work.
This action is considered as a management
exception.
If a job is returned by other
departments/actors,
the
actor
may
redo/reassign/return/report
the
job
according to different preferences. This
action is considered as a management
exception.

Based on previous work, including an extensive literature review and case studies, a
variety of relational behaviors are modeled to represent the preferences and judgments
directly related to work decisions, as shown in Table 4.6. These relational behaviors are
further divided into a set of specific actions. The definitions of the fundamental relational
behaviors and related actions are shown as the follows:

113

4.3.2.1 Prioritizing

Construction project teams always handle multiple projects and corresponding tasks
simultaneously (Payne 1995). Therefore VOICE considers concurrent task scenarios,
which means for a project team member, there are multiple tasks to perform at the same
time. It is assumed an actor can only process one task at one time; therefore, he/she needs
to prioritize tasks. The actual priority of a task depends on two factors: the priority of the
project among all concurrent projects, and the priority of the task among all concurrent
tasks. Then, actual priority of a task can be obtained by:

task.actualPriority=project.priority*task.priority

(1)

where a higher number means more urgent. An actor will compare the priority of the tasks,
and then select the most urgent task (with the highest priority number) for further actions
in every simulation tick.

4.3.2.2 Processing

In VOICE, processing a task means the actors (project team members) reduce a certain
amount from the total work amount of the task. An actor’s processing time depends on the
actor’s work competence and the task’s difficulty, and therefore can be given by:

actor.processTime=task.actualDifficulty*task.amount/actor.competence

(2)

where a task’s actual difficulty depends on the relevant project’s complexity and its own
difficulty level among all tasks, i.e.,
114

task.actualDifficulty=project.complexity*task.difficulty

(3)

The time unit used in the model is a simulation tick. As a result, the remaining amount of
a task at every tick is given by:

task(tn+1).amount= task(tn).amount-actor.competence/task.actualDifficulty

(4)

Moreover, actors may make mistakes when processing tasks. The probability of making a
mistake depends on the work quality of the actors; thus, the final mistake percentage of a
task can be given by:

task.mistake=task.mistake+actor.mistake

(5)

4.3.2.3 Submission

Submission occurs when an actor has a supervisor or a successor depending on the
reporting structure and work process. In this case, the actor submits the task to the
supervisor or successor when it has been finished, i.e., work amount becomes zero:

task.amount=0

(6)

Then, the task is passed to the supervisor or successor, and the work amount is adjusted to
its original value.

4.3.2.4 Assigning

In some cases, managers may choose to assign tasks to their subordinates instead of
conducting them themselves (such tasks are indexed with “assign=1” in task arrays). For
115

example, a project manager may assign look ahead scheduling to a project control
specialist. When there are multiple subordinates, the managers need to choose one based
on assigning preference (see Table 4.4). Then the task will be moved from the manager to
the selected staff member, and the assign indicator of the task will be adjusted to 0.

4.3.2.5 Requesting and approving job

In a construction project, decisions are made at different levels. For some decisions, only
project managers or president has the authority; while some decisions can be made by
staff members. For example, in cost estimation, an estimator may have the right to
determine the unit price of concrete framework based on means book or quota from
suppliers. But the president might be the only person who can determine the final
overhead rate. If approval from higher level of management is needed in order to process
a project task, the actor has to submit the request. As a result, VOICE models the
requesting and approving activities as particular behaviors in construction project teams.
Each task has an authority level index indicating which level of actor has the right to
process. If the authority level of a task is higher than the actor’s level, the actor will check
the approval status of the task to see whether it has been approved (approval status=1) or
not (approval status=0). If the task has not been approved, the actor will submit it to the
upper level actors for approval. For instance, the task “determining profit margin” might
be a task beyond the senior estimator’s authority; therefore he/she needs to submit this
task to the president. Then the president will approve the profit margin and return it to the

116

requester. Asking for approval and approving are regarded as “management exceptions”
in this model, which refers to the time spent on indirect work. Other management
exceptions include: seeking additional information, returning unqualified work, reporting
work stress-out, and etc. Such behaviors are not directly related to the work itself, and
they have thus been considered to be an inefficient amount of time spent on a project. One
of the aims of this model is to reduce such inefficient use of time.

4.3.2.6 Conflict management

Construction project involves the efforts of different disciplines. The cooperation between
members from different organizations becomes a critical issue for most construction
projects due to the possible misunderstanding of project goals and divergent interests
(Baiden et al. 2006). When a conflict occurs between two actors, they may seek a solution
directly between each other. For example, in proposal development, proposal team
member requires intensive information from the engineering design team for quantity
take-off.

If a proposal team member cannot get necessary information for proposal

development from the engineering team, he/she may contact the counterpart in the
engineering team for additional information. However, if after several iterations the
problem remains unsolved, the actor may seek help from his/her supervisor, such as a
manager or a coordinator, for a better or faster solution. Then the supervisor may
communicate with responsible actors or their supervisors. For example, a coordinator of
the proposal team contacts a coordinator of the engineering team to expedite the

117

information exchanging between two teams. In certain cases, the supervisor might raise
the issue to upper management such as president. Because of the involvement of higher
authority, the conflict solving process may be expedited, but it adds up to nonproductive
time. Whether to raise issues to higher levels depends on the preference of an actor, and
the leadership style of the team leaders (such as micro-management preference).

4.3.2.7 Information exchange

Following Levitt’s work (2007), construction project management activities can be
regarded as exchanging project task related information. In some cases, there is a need for
additional information for processing a project task. For example, in the proposal
development of a power plant project, a proposal team member may need the job site
configuration information from the engineers to make a preliminary estimation. Such
information exchange channel is always different from the channel for work process. Take
the same proposal development instance: although an information exchange happened
between the proposal team member and an engineer, it is not necessary that there is a
work relation between them. To reproduce such official/unofficial communications
happening among actors, this model indexes the information status of every task. If
task.infomation=1 (instead of 0), it means additional information is needed. Then the actor
will pass this task to the information source through the “information flow” structure. The
information source then will index the task.infomation as 0 and return it to the requester.

118

During this process, a certain amount of time is spent based on the exception handling
time of the actor, and the management exception indicator increases.

4.3.2.8 Meeting

Meeting plays a critical role in construction project management (PMI 2008). VOICE
models meeting as an event to reduce misunderstanding, enhance communication and
solve problems pertaining to the construction project management. The president
examines the management exception indicator at every tick, which shows the frequency of
approval related activities, quality related activities, communication activities, etc.:

President.exception=∑(approval_related_activities+work_stress_reports+communication_activities+quality_related_activities)

(7)

If the management exception indicator is beyond the threshold of the president, the
president will hold a meeting to solve the exceptions. The meeting duration is given by:

Meeting duration=a*(number of exceptions/threshold)

(8)

where a is the preference of the president. During the meeting, all actors (president,
managers and staff) must stop their work. After the meeting, the management exception
indicator will be cleared to 0; plus, all current tasks will be approved (task.approval=1),
the work quality will be improved (mistake percentage is reduced), and information will
be shared thoroughly (task.information=0).

119

4.3.2.9 Monitoring

Some tasks pass through different actors. When an actor receives a task from others,
he/she will check the quality of the work. In case the mistake percentage of the task is
higher than his/her threshold, i.e.,

task.mistake>=actor.mistakeThreshold

(9)

actors may adopt the following actions based on position and preference: correcting the
mistakes themselves, returning the unqualified task to the responsible actors, or assigning
this task to subordinates to correct. All actions require additional time and increase the
management exceptions.

4.3.2.10 Correction/rework

As discussed, if the quality of a task is not satisfactory, a series of reciprocal activities, i.e.,
“back-and-forward” actions will happen among actors. For all actors, if they receive
returned tasks, they may need to make corrections or redo the entire tasks. In VOICE,
returned tasks will be assigned highest priority, and the task mistakes portion will be
treated as new tasks for processing.

4.3.2.11 Overburden

In this model, work related pressure is defined the as the total remaining work amount:

staff.workPressure=∑(task.remainingAmount×task.difficulty/staff.competence) (10)

120

If the estimated work related stress is beyond the capacity of the staff, the staff will refuse
additional tasks and report it to the manager for re-assigning. From the definition of
reciprocal activities we can see that such activities are not relevant to the direct work,
therefore, they are counted as management exceptions.

The above actions constitute the fundamentals of relational behaviors of between team
members. For a particular case (i.e., different team members), the ultimate flowchart of an
actor’s relational behaviors is comprised of the above fundamental actions. For each of
them, the upper half demonstrates the “action process,” which refers to the actual work
conducted by the actors, and the lower half demonstrates the “decision process,” which
means the decisions made on every contingency. In VOICE, although different actors
possess different behaviors, the final behaviors and decision processes of a particular actor
are a combination of these fundamental components. Figure 4.4 demonstrates examples of
actors’ behaviors, comprised of the above fundamental behaviors. For each flowchart, the
upper half demonstrates the “action process” which refers to the actual work conducted by
the actors, and the lower half demonstrates the “decision process” which means the
decisions made at every contingency.

121

President

Manager

Staff

Figure 4.4 Flowchart of actors’ behaviors (example)

4.3.3. Modeling Organizational context
Following Chapter 3, VOICE represents a construction organization with three
components: (1) reporting structure (or administration structure); (2) work process; and (3)
information flow as the channel of information. Each component symbolizes a particular
channel between actors when a related set of behaviors occurs.
122

4.3.3.1 Reporting structure

Reporting structure is the formal organizational structure that determines the arbitrary
delegation pattern within an organization and reflects report directions. It can be
demonstrated by a formal organizational chart (Daft 2009); therefore, official
documentation was used to depict the organizational chart (Figure 4.5).

For interpretation of the references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation.
Figure 4.5 An example organizational chart of a project proposal team

If there is no official organizational chart in a studied company, then a short interview was
conducted with the president or managers to draw the organizational chart.

123

In VOICE, this administration structure is modeled as an association class connecting
president, manager and staff. All authority relevant activities taking place among actors
occur through the administration structure, such as “requesting and approving job” or
“assigning job.” Figure 4.6 represents the administration structure by solid black lines,
and the arrow represents the report direction. The current proposed model limits the
administration structure to a three-level pattern. Additional levels will be presented in
future work.

4.3.3.2 Work process

According to the literature, procedures within an organization are critical for the
formulation of cooperativeness. The procedures are represented as work process in this
research. Work process in the construction domain refers to the “necessary procedure in
construction companies’ execution of their business” (Cheng and Tsai 2003) that typically
includes the planning and control of projects through conceptual planning, design, bidding,
construction, and commissioning (Shohet and Frydman 2003a). This research studies the
work process as the pattern of task sequence and work arrangement, i.e., the channel that
allows a sequence of managerial activities to flow through actors.

In VOICE, work process is modeled as an association class connecting actors and tasks.
Direct task processing is executed via the work process structure. For example, an
estimator (staff) is required to conduct a WBS analysis for a project. In VOICE, this task
(WBS analysis) is passed from project to the estimator via the work process channel.
124

Once finished, he/she may pass the WBS analysis result to another estimator for further
work. This is modeled in VOICE as passing the task from one actor to another via the
work process channel.

The work process is represented by developing a Task-Actor Relation Table (TART),
which reflects the task assignment for every actor. First, the developed list of activities
(Step 2, formulate representation of work) was sequenced based on the interviews.
Process mapping technique was used as a communication enhancement tool. Then the
follow-up semi-structured interviews and online-surveys were conducted to determine the
responsibilities of each employee (Table 4.7). Finally, all information was collected to
develop a TART table (Table 4.8). In this TART table, all the tasks (e.g., sub tasks of
estimating) are arranged according to the sequence of the task queue (input by users),
which is shown as the column heads; all the actors (e.g., senior estimator and his/her
helpers) are arranged in the row head. Then, the cells are marked if an actor is responsible
for a task (e.g., an estimator helper is responsible for quoting subcontractors). The work
process pattern is shown with zigzag lines inside the TART (Table 4.8). Finally, the real
world work process is translated to abstract task delivery pattern in VOICE (Figure 4.6).

Table 4.7 A sample survey showing individual responsibilities
Task Code

Name

O.PLN.001
O.PLN.002
O.PLN.003
O.PLN.004

Study potential opportunities
Contact subcontractors
Price self performed work
Evaluate subcontractors' quotes
125

Responsible
employees
A
S
A
S
A
S
A
S

Signature

Table 4.8 Task-Actor Relation Table (TART)

126

Step 2: Delivering tasks

Step 1: Making TART

Figure 4.6 Task delivery pattern in VOICE

4.3.3.3 Information flow

Information flow is the formal/informal communication connection that determines the
coordination, message and knowledge movement. Because information flow is the most
intangible structure compared to the other two structures (reporting structure and work
process), representing information flow requires intensive and reactive interview efforts
except for document studies. Process mapping techniques were used to enhance the
communication between the researcher and the respondents.

In VOICE, information flow is modeled as a set of channels where work related
information is transferred from one actor to another, by modeling it as a class of
127

associations reflecting the coordination, message and knowledge movement among actors,
while at the same time reducing uncertainty (Nonaka and Konno 1999). The movement
of information is coded in the “ask-and-answer” fashion. For example, in a power plant
project, an engineer needs critical information about the project specification from the
project planning team member, such as the nominal Megawatt reading requirement from
the client; this is modeled as passing the task with a null information status
(task.infoStatus=0) from the engineer to the planning team member; and the planning
team member’s answer can be modeled as changing the information status to applicable
(task.infoStatus=1) and returning the task to the engineer. Meanwhile, the planning team
member may need designing information for cost estimating, such as the job configuration.
He/she then will send a task with a null information status to the engineer, and engineer’s
answer can be modeled as switching the information status to applicable. As a result, the
reciprocal information exchange activities can be demonstrated with two monodirectional
flows (Figure 4.7).

Figure 4.7 Information flow pattern in VOICE

128

4.4. Output measurement
The framework of team performance proposed in Chapter 3 was supported by the domain
experts from the studied cases. Accordingly, specific measures of the indicators, i.e., how
the studied companies measure these indicators, are presented as follows:

4.4.1. Time
The duration of a project (tm) follows functions 11 and 12:

project.size(tn+1)= Project.size(tn)-∑task(tn).amount

(11)

if project.size(tm)=0, then tm is the time for a project

(12)

Moreover, VOICE provides indicators for time used for different activities:

Total time=process time+communication time+idle time

(13)

Work time=process time+communication time

(14)

Effective time=process time

(15)

Idle time is the span of time when the actors “have nothing to do” (normally this is due to
waiting), and process time stands for the time directly used for processing the tasks,
following this equation:

Process time=task.actualDifficulty*task.amount/actor.competence

=( project.complexity*task.difficulty) *task.amount/actor.competence
(16)
129

4.4.2. Quality
Quality of a project team is associated with “detected failed production work volume” (Jin
and Levitt 1996). This research uses a mistake percentage to represent the failed
production work amount of each task, which ranges from 0 to 1 (100%). Provided that
processing tasks are the sole function of a project team in this research, the ultimate
quality of the entire teamwork is thus described as one minus weighted sum score of task
mistakes, as shown in the following function:

project.quality=1-∑(taski.amount*taski.mist)/project.size

(17)

where taski.mist stands for the percentage of mistakes in a particular task and follows this
equation:

task.mist(after processing)= task.mist (before processing) + (1-staff.quality)

(18)

4.4.3. Effectiveness
Following Jin and Levitt’s (1996) work, the effectiveness of a project team is defined as
the percentage of productive work time versus total work time. This research therefore
formulates effectiveness as a ratio of effective work time (direct processing of tasks) to
total work time. It is given by the following function:

project.effectivess= ∑(effective time/Total time)

130

(19)

4.4.4. Efficiency
Levitt (2007) defines two types of efficiency for any project team. Among them, time
efficiency is given by a ratio of estimated work duration to simulation duration. If setting
simulation duration to a standard unit of time such as one day, the estimated work
duration demonstrates the actual work (measured in man hours) that can be finished by
the team within one day. It reflects the average work amount completed per time unit.
This research follows a similar definition and measures efficiency as:

project.efficiency= ∑taski.amount /Total time

(20)

4.4.5. Work related pressure
Work related pressure is a subjective judgment made by team members regarding the
relation between work environment and individual ability (Cox et al. 2010). Although it is
hard to quantify, perceived work related pressure is directly related to the workload (Hall
2004). From a purely technical perspective, this research measures work related pressure
as total work amount (measured in time) of all tasks at hand:

staff.workPressure=∑(task.remainingAmount×task.diff/staff.comp)

task.remaingingAmount=task.amount-finished amount

(21)

(22)

4.5. Basic assumptions of VOICE
The basic assumptions of VOICE have been addressed in previous sections. This section
summarizes them in table 4.9 for the better reference of the readers:

131

Table 4.9 Major assumptions of VOICE
Agents
Project

Input

Task

Process

Individuals

Relational
behaviors

Assumptions
• A project can be divided into a sequence of executable work efforts
called tasks;
• The project is completed once all its corresponding tasks are
finished;
• Multiple projects can be handled by a team simultaneously.
• A task is the most basic executable work effort for a project team
member;
• Task amount is measured by “hours”, i.e., how many hours it takes
to finish a task by a team member with average competence;
• A task is finished when task amount equals zero; Once finished, it
will be removed and its amount is reduced from the project;
• Tasks have different priorities; A task with higher level of priority
is processed first;
• Some tasks need approval from managers or president, or
additional information before processing;
• If a task is dependent on another one which has not been finished, it
cannot be processed;
There are three major roles in a construction project team, including
president, manager and staff member. The major responsibilities of
them are:
President: holding meetings, handling exceptions, approving
jobs and etc;
Manager: Assigning tasks, coordinating conflicts, quality
monitoring, handling exceptions (e.g., team members’
complaints about overburden), approving jobs, attending
meetings and etc;
Staff: Processing tasks, exchange task related information,
delivering or submitting finished tasks, quality monitoring,
reporting exceptions, attending meetings and etc.
Specific assumptions are shown in 4.3.1
VOICE models the following relational behaviors. Each behavior is
modeled as a “behavioral module” in a “Library of Behaviors” (see
appendix 3). In VOICE, an actor will first examine the situation. Then
based on his/her judgment on the situation and his/her preference, a
certain behavioral module will be triggered.
• Prioritizing: An actor can only process one task at a time; therefore
prior to further actions, an actor may order all tasks in hand based
on the readings of their priorities;
• Processing: Processing a task means reducing certain amount from
the task every simulation tick. The amount reduced depends on task
difficulty and competence of the actor. During this process, actors
may commit mistakes shown as a mistake percentage of the task.
The mistake of the entire project is measured by the weighted
132

Table 4.9 Cont’dassumptions of VOICE
Major
Agents
•

•

•

•

•

•

•

•

•

Organizatio
nal context

Assumptions
mistake percentage of all its tasks;
Submission: Once a task is finished and there is no successive
actors (according to work process), the actor will submit the task to
his/her superior;
Assigning: A manager may assign a task to his/her subordinates
based on the assigning preference (e.g., speed driven or quality
driven);
Requesting/approving: some tasks require approval from superiors;
in this case, the actor will render this task to his/her superior, who
will approve the task or render it again to his/her superior based on
the technical information of the task and authority level of the
actor. After a while (depending on the actor’s exception handling
time), the approved task will be returned to the requestor;
Conflict management: If a conflict cannot be solved by staff
members, it will be raised to the manager or coordinator for further
actions;
Information exchange: If the available information for a task is less
than the required information (measured as a percentage), the actor
will send this task to another actor (with needed information). After
a while (depending on the actor’s exception handling time), the task
will be returned to the requestor with necessary information;
Meeting: If the number of all exceptions in a team is bigger than the
threshold of the president, a meeting will be held. The duration of a
meeting depends on the number of exceptions and preference of the
president. After a meeting, all tasks are approved, information is
provided, and exceptions are cleared;
Monitoring: If the mistake percentage of a task is bigger than the
threshold of a staff member or a manager, it will be returned to the
original actor, or will be corrected at a cost of additional time and
etc. The reactions to unqualified tasks depend on the preference of
the actor.
Correction/rework: If an actor receives a returned task marked as
unqualified, he/she will redo it to improve quality. The time spent
on correcting/redoing a task depends on the mistake percentage of
the task and competence of the actor;
Overburden: An actor sums up total amount of tasks (burden) in
hand – if this number is bigger than his/her capacity, he/she will
suspend working, and return new tasks to the manager. The
manager will reassign it to a staff member with smaller level of
burden.

• Reporting structure: It is assumed that construction project team has
a three level hierarchical organizational structure;
• Work process: The procedure of processing a task; it shows the
133

Table 4.9 Major assumptions of VOICE
Cont’d

Output

Agents

Assumptions
sequence of delivering a task among team members. It always starts
from a manager;
• Information flow: The channel connects information requestors and
providers. Information only refers to task related information, i.e.,
that is needed for processing a task. Two-way information flow is
modeled as two one-way information flows.

Outcome
• Time: Hours spent on a project;
performance • Quality: The mistake percentage of a project, which equals
weighted average of the mistake percentages of all its tasks.
Process
• Efficiency: Finished task amount per unit time (hour);
performance • Effectiveness: Ratio of productive time versus total time.
Productive time is defined as time directly spent on processing
tasks;
• Work pressure: Total work amount of tasks in hand for an actor;

4.6. Developing an integral platform
A set of ABM development platforms were compared in terms of their flexibility, richness
of libraries, development user interface, analytical functions, and support from developers.
REcursive Porous Agent Simulation Toolkit Simphony, or Repast S (Collier 2003) was
determined to be a suitable development platform for this research. Repast S is a pure
Java-based implementation, which has been widely accepted recently in academia. It
builds on Object-Oriented Programming (OOP) where each object can be naturally
regarded as an agent in VOICE. Repast S is an open source platform so the functions
provided can be easily tailored to the requirements of this research. In addition,
developers of Repast S have built an interactive supporting forum where development
problems can be solved immediately. A snapshot of Repast is provided below (Figure 4.8).

134

List of agents
and connects

Animation
Document
tree

Tabs for animation and charts

Attributes

Figure 4.8 The runtime interface

4.7. Summary
This chapter developed an Agent Based Model for interpersonal cooperation in the
construction context named VOICE. VOICE builds on the conceptual framework
proposed in chapter 3. In order to convert the conceptual framework to an executable
simulation model, elements of interpersonal cooperation were conceptualized, and
performance metrics were formulated. In the next chapter, a verification and validation
framework for VOICE will be introduced.

135

CHAPTER 5: MODEL VERIFICATION AND VALIDATION
Unlike a purely theoretical study, this research requires a certain degree of confidence in
VOICE’s ability to generate similar results as observed in real world construction teams,
and in return, that decisions and analyses derived from it can be useful and convincing.
Model verification and validation (V&V) are used to examine the extent to which VOICE
acts in its designed purpose and generates results that are similar to observations.

5.1. Verification
A verified model works as designed (North and Macal 2007). The following methods
were used for verification, including code debugging, logic examination, unit test and
hypothetical case test.

5.1.1. Code debugging
Java Eclipse (2012) provides an automated debugging framework and can help increase
the efficiency of model programming by reducing the time spent seeking “failureinducing circumstances.” Early in the model development, all functional components of
VOICE, including the model initializer, agent classes and implementation class were
examined with the debugging function of Java Eclipse.

5.1.2. Logic examination
In order to verify the programming logic to be a correct realization of model design, the
program flowchart of each model functional component was carefully examined.

136

START

Task
urgentTask=new
Task()
List=all tasks at
hand

List.hasNext()
Y
N

Task task=list.next()

task.priority>urgentTask.priority

N

Y
urgentTask=task
END

(a) Original flowchart

(b) Revised flowchart

Figure 5.1 Logic errors detection by flowchart examination: example of prioritizing tasks

Figure 5.1 demonstrates a real example of a logic error detected in the programming of
VOICE. In VOICE, all actors need to prioritize tasks prior to other actions. A dummy task
is created at the very beginning with the least possible priority and named as urgentTask.
Then a real task will be compared with urgentTask on their priority measures. The real
task will be saved as urgentTask, replacing the dummy one if it has a higher priority.
137

Repeating this process, an urgentTask is finally obtained. However, a flowchart
examination found a potential logic error in Figure 5.2 (a): when an actor has no real tasks
at hand, the dummy urgentTask will be retained and processed further as a real one
because actors cannot differentiate between dummy and real tasks. Therefore, a condition
judge should be given to the selected urgentTask to determine if it is a real one. Using the
same examination, flowcharts of all functional units of VOICE have been reviewed to
make sure the model is programmed in the designed manner.

5.1.3. Unit test
The ultimate goal of model programming is to realize the designed function.
Correspondingly, an examination should be performed to check if the program is coded
“functionally.” There are two levels of “functionality”: (1) unit functional: each unit of the
model works as designed; and (2) global functional: the entire model works as designed.
To verify the unit functionality of VOICE, this research develops a method that involves
the use of a test marker (an object with interested attributes) and corresponding check
point. Table 5.1 lists thirteen functional units of VOICE, and their test markers and check
points.
Table 5.1 Test markers and check points of functional units of VOICE

1

Functional
unit
Prioritizing

2

Processing

Test Marker(a dummy
instance of agent for test)
A task that meets the
following condition:
testTask.priority=max(all
tasks)
A task that meets the
following condition:
138

Check point (when met, unit is
functional)
urgentTask=testTask

testTask.amount(t+1)< testTask.amount(t)

Table 5.1 Cont’d
Test markers and check points of functional units of VOICE
Functional
unit

Test Marker(a dummy
instance of agent for test)
testTask.amount>0
3 Submitting A task that meets the
following condition:
testTask.submit=1
A task that meets the
4 Assigning
following condition:
testTask.assign=1
5 Requesting A task that meets the
following condition:
testTask.authority>threshold
6 Approving A task that meets the
following condition:
testTask.approval=1
All agents meet the
7 Meeting
following condition:
Agent.meeting>0
Monitoring A task that meets the
8
following condition:
testTask.mistake>threshold
9 Correction A task that meets the
rework
following condition:
testTask.redo=1
10 Overburden An staff that meets the
following condition:
staff.
workburden>staff.capacity
and a testTask
11 Requesting A task that meets the
info
following condition:
testTask.info<threshold
A task that meets the
12 Providing
info
following condition:
testTask.infofeed=1
A task that meets the
13 Conflict
mgt
following condition:
testTask.iteration>threshold

139

Check point (when met, unit is
functional)
testTask.coordinates=upperLevel.coordinates
testTask.coordinates=staff.coordinates

testTask.coordinates=upperLevel.coordinates

testTask.coordinates=lowerLevel.coordinates
Agent.meeting(t+1)= Agent.meeting(t)-1

testTask.coordinates=submitter.coordinates

testTask.mistake(t+1)<
testTask.mistake(t)
testTask.coordinates=
upperLevel.coordinates

testTask.coordinates=infoSource.coordinates
testTask.info(t+1)> testTask.info(t)
testTask.coordinates=requester.
coordinates
testTask.coordinates=coordinator.coordinates

5.1.4. Hypothetical case test
Two hypothetical cases were developed after consultation with domain experts to verify
the global functionality of VOICE. The first hypothetical case explores if VOICE works
with a one-project scenario. A single project is created, which generates a sequence of
tasks as the inputs of an illustrative construction project team (Figure 5.2).

Figure 5.2 An illustrative project team used in model verification

The values of attributes used in the simulation are shown in Table 5.2. To be noted, in
order to simplify the hypothetical case, most attributes are assigned with deterministic
values instead of random values.

If VOICE is functional, the sequence of tasks will be processed by all relevant actors and
team performance will be recorded and demonstrated at the end.

140

Table 5.2 Values of attributes used in the hypothetical cases
Attributes

Explanation

1
2
3
4
5
6
7

Project size
Work amount
Difficulty
Priority
Authority
Competence
Capacity

8

Process time

Total work amount of the project
Work amount of every task
Difficulty of the project and every task
Priority of a task
Authority level of a task
Processing efficiency of an actors
Most acceptable total work amount for an
actor
Time needed for handling nonproductive
exceptions
Threshold for a manager to accept a
submitted task
1 means manager will returned unsatisfied
tasks
Threshold of the president to hold meetings

Mistake
threshold
10 Quality
preference
11 Exception
threshold
9

Value in
simulation
1000
24
1
Uniform (1,3)
Uniform (1,3)
1
20
1
0.1
1
20

Figure 5.3 shows the performance outcomes of the first hypothetical simulation, including
work progress, effectiveness, mistakes, work related pressure, etc. Similarly, the second
hypothetical case was created to reflect a multi-project scenario. Three projects were
created for the scenario, with each project size being 300. Values of other attributes
remain the same as shown in Table 5.2. Simulation results show that VOICE is able to
model the processing of three projects, differentiating the tasks of three projects and
documenting the performance indicators individually (Figure 5.4). No warnings or errors
were detected in either hypothetical case simulations. This supports VOICE’s ability to
work functionally in both one- and multi- project scenarios.

141

a. Work progress

b. Effectiveness

c. Work related pressure

d. Average work related pressure

Figure 5.3 Simulation results of the first hypothetical case (examples)

a. Work progress

b. Effectiveness

Figure 5.4 Simulation results of the second hypothetical case (examples)

142

Figure 5.4 Cont’d

c. Work related pressure

d. Average work related pressure

5.2. Validation
Literature has different definitions of validation. For example, Kleijnen (1995) claimed
that validation is concerned with determining “whether the conceptual simulation model is
an accurate representation of the system under study”. But Oberkampf et al. (2004)
claimed that if a model can generate similar outputs as empirical data, it is validated. It is
also worth noting that a perfect validation is unachievable (Rojas-Villafane 2010). The
practical purpose of validation is to obtain sufficient confidence on the model in a hope
that the simulation results can meet the needs of the research (Rojas-Villafane 2010).
Because the main aim of this dissertation is to explain instead of to predict the influence
of cooperation in construction settings, the target of validation in this research is to ensure
VOICE to be able to reproduce the pattern of empirical data collected from case studies
rather than exact predictions.

143

This dissertation has adopted a widely accepted validation framework proposed by
Sargent (2004). Two steps were followed to validate VOICE: (1) input validation: data
and conceptual model validity ensures the inputs of a model to be accurate and reasonable,
and (2) output validation: output validity serves as a guarantee that the model will
generate similar outputs as observed in reality.

5.2.1. Input validation
Input validation aims to ensure that data and key assumptions used in VOICE can reflect
the reality to a certain extent. In the beginning, multiple sources were used in order to
triangulate the data collection. One source was surveys distributed to 34 domain experts
with 515 combined years of experience who are believed to be the most relevant process
owners in the studied cases. The second source was interviews of five key decision
makers with 119 combined years in management positions, including vice president,
principals, department heads, coordinators and managers. Third, when there were
sufficient documents, such as time sheets, they were used to supplement the opinions.
Then, descriptive statistical analysis was conducted, using the aggregated empirical data
and model assumptions (especially behavioral assumptions) to reveal data patterns such as
Probability Distribution Functions (PDFs). Last, the domain experts and/or key decision
makers in case study companies commented on and validated the statistical findings,and
the analysis was compared to document findings when applicable. This process started
early in the model development and went through numerous iterations until enough
confidence was obtained with regard to the validity of data and key model assumptions.
144

Additionally, when needed, onsite pilot studies were conducted to validate important
model inputs, such as certain individual behaviors that are difficult to capture in surveys
and interviews.

5.2.1.1. Task information validation

In VOICE, task duration is a random number drawn from a normal distribution. This was
validated by findings from the document studies and surveys administered to process
owners. On the one hand, most construction companies have well documented records
that can be used as the basis of task duration validation. In one case study, the contractor
provided seven-month time sheets, which specified the list of tasks with an internal
coding system, task duration in hours, dates, phase in a project, and responsible actors
(Table 5.3). These time sheets were organized by projects (using project codes provided
by the contractor), where tasks of the same project were aggregated and analyzed. Initial
statistical analysis found task duration follows a normal distribution.

In another case study, there were no similar time sheets. Therefore, instead of document
study, this study relied on a series of surveys to 34 domain experts to collect task
information. Figure 5.5 demonstrates two examples of survey results. Some respondents
provided task duration in hours/days directly, while others gave an estimation of the
percentage of each task for the entire work period. The latter ones can be easily converted
to hours by timing the entire work hours. Similar to the previous case, initial statistical
analysis was performed and normal distribution was found.
145

Table 5.3 An example of time sheet
#
1
2
3
4

Task description
phone call; converse; email
team
phone call; converse; email
team
status of eviction, process

9

1
1

2.16.09

1

3.16.09

1

3.17.09

1

3.18.09

1

3.19.09

1

3.20.09

2

4.14.09

0.25

36402 Tenant Eviction

2.10.09

4.24.09

phone calls; schedule; court
bailiff
phone calls; schedule; court
bailiff
phone calls; schedule; court
bailiff
phone calls; schedule; court
bailiff
phone calls; schedule; court
bailiff
36401 Tenant Eviction

10

Duration Date
1
2.09.09

5
6
7
8

Phase
231040
231040
231040
231040
231040
231040
231040
231040
231020
231020

Project
CMI.9.5
CMI.9.5
CMI.9.5
CMI.9.5
CMI.9.5
CMI.9.5
CMI.9.5
CMI.9.5
CMI.9.5
CMI.9.5

The respondent fills the hours for each task The respondent fills a percentage for each
directly

task
Figure 5.5 Surveys to collect task information

146

In an aggregated analysis, it was found that the durations of tasks in these two case studies
followed a normal distribution N (24, 4) with p-value=0.608 (Figure 5.6). The mean value
is 3 work days, which might be because respondents tend to report task durations in days
or half days instead of hours. Thus, using a normal distribution in VOICE to generate
tasks is validated.

Figure 5.6 Distribution of task duration
Other task variables, such as task dependence and difficulty, are represented as a
percentage or a range in VOICE. These variables were validated by the survey of 34
domain experts. For example, in order to validate the value of task dependence used in
simulation, the domain experts were asked to describe their perceived magnitude of
dependence, using a 5-point Likert scale. The results were further converted to a
percentage to determine how many tasks could be processed when the preceding tasks
were still underway. The results indicated this percentage follows a Poisson distribution.
147

This finding was validated by the domain experts. In the same way, all task information
was validated with the confirmation from domain experts.

5.2.1.2. Individual information validation

Individual competence was mainly validated with an analysis of personnel documents.
Due to the confidentiality of the assessment system, it was impossible to collect direct
data of individual competence. Following the literature (Sandberg 2000; Humpel and
Caputi 2001), the author proposed to quantify competence of employees by their years of
experience. This method was supported by five key decision makers in case studies. In
addition, years of experience for each position were collected, and a linear relationship
was found between years of experience and their formal levels of skill. Therefore, the use
of years of experience as an indicator of competence in VOICE has been validated. A
survey was then conducted to solicit respondents’ years of experience in current or similar
positions. Results follow a two-peak normal distribution with mean value around 15.6
years. Thus in VOICE, actors with 15 years of experience possess a standard processing
efficiency (competence=1) and competence values of other actors are proportional, based
on their years of experience.

Other individual information, including exception handling time, work quality, quality
threshold, exception threshold and preference were aggregated and converted to a range or
as a distribution function. These ranges and distribution were validated during the
interviews with five key decision makers since they have a better understanding of the
148

situation of the entire team. For example, aggregated data indicated that the quality
threshold (acceptable mistake percentage) of managers follows a Pareto distribution. The
decision makers supported this finding by asserting that only a small portion of managers
tend to be extremely strict on work quality. Similarly, other individual variables used in
VOICE were validated. In addition, several pilot studies were performed onsite by the
author to validate certain individual information that is difficult to ascertain through
surveys/interviews. An example was exception threshold of the president/VP. The author
attended all emergent meetings in a case study company in seven months, and
documented the discussion key points during the meetings as an indicator of the
president/VP’s threshold for emergent meetings. The author’s observations were then
compared to interview findings. The individual information used in VOICE reflects
realistic situations.
5.2.1.3. Organizational context validation
VOICE assumes a three level hierarchical reporting structure. This has been validated by
the formal organizational charts provided by the case study companies. The charts of both
companies can be described as a classical functional organizational structure, with three
levels: The top level is President/VP/Principals, whose responsibilities include scope
management,

proposal

control,

and

communication.

The

middle

level

is

managers/coordinators, who are responsible for work integration, quality control,
coordination, and conflict management. The lowest level is staff, who are responsible for
most specific tasks. This was also supported by findings of previous literature (Project
149

Management Institue 2008). For the other two aspects of organizational context in VOICE
(work process and information flow), simulation models were compared against official
documents for validation purposes. Specifically, job descriptions were studied to line up
tasks with actors. Work flow documents were studied to examine formal information flow
in teamwork. Because of the scope of VOICE, informal information flow was not
considered.

5.2.1.4. Behavioral assumption validation

Behavioral assumption validation aims to validate the reasonableness of agent behavioral
rules in VOICE. It includes two subtasks: (1) validating the list of behaviors, and (2)
validating mathematical functions used in the simulation.

For the first subtask, the initial list of work-related behaviors was discussed with the five
key decision makers in a series of face-to-face interviews. Their comments were
considered to reflect more practical perspectives in the development of the list of
behaviors. After several iterations, the list was finally confirmed by the key decision
makers. Then, the 34 domain experts were asked to describe their major responsibilities
and activities on a daily basis. Their answers were aggregated and classified to compare
against the list of behaviors. A testimonial was finally made that the current list of
behaviors used in VOICE (Table 4.6) reflects the most significant work-related behaviors
and activities in both cases.

150

In order to validate the mathematical functions of behaviors used in VOICE simulations, a
series of charts were prepared and validated by the five key decision makers. For example,
empirical data found that team members’ behavioral reactions to expectations (e.g.,
inferior information) can be grouped into four categories with different proportions:

"Report to supervisor"

39.58%

"Talk to responsible person"

39.58%

"Leave it alone"

10.42%

"Others"

10.42%
0%

10%

20%

30%

40%

50%

Figure 5.7 Behavioral reactions to exceptions

This chart was validated by the key decision makers in interviews: they claimed that
based on their experience, most team members will seek help from the team or solve the
problems proactively when needed. Only a small portion of team members adopt a
passive action. Thus, the rules in VOICE concerning agent’s reactions to exceptions were
validated. By similar means, the important mathematical functions about other behavioral
rules were validated.

5.2.2. Output validation
The purpose of output validation is to determine if VOICE is able to generate similar
results to observations given rational inputs. In order to realize output validation, a wellstructured scenario was modeled on VOICE to represent a representative proposal
151

development process by a project team of a studied company. Then the simulated results
were compared against the empirical data collected from the same team.

In this representative scenario, project, measured as 500 standard man work hours, is
processed by six staffs. The amount of each task follows the normal distribution found
from observed data, and values of other attributes used in the simulation, such as staff
competence, were randomly drawn from the PDFs validated in input validation phase.
Then, simulation was repeated 300 times on VOICE, and results were summarized to fit
probability distribution function for each performance indicator of VOICE. Then these
simulated PDFs were compared visually against the PDFs of empirical data. If the PDFs
are similar, it can be determined that VOICE works as observed. Statistical hypothesis
testing was not used given the small sample size of empirical data.

Several key performance indicators were selected to validate VOICE’s ability to generate
similar outputs as empirical data, including work duration, quality, effectiveness, and the
communication iterations between employees. The first indicator compared was work
duration. The studied team didn’t have record of durations of proposal development in the
history. As a result, in the interview respondents were asked about their opinions about
the work durations for proposal development. The reported minimum duration was one
week (40 hours), while the maximum duration was 2 to 3 months (assuming 400 hours).
The most likely duration, according to the respondents, is 2 or 3 weeks (assuming it is 120
hours). This ends up with a triangular distribution (40, 120, 400). Figure 5.8 (empirical)

152

demonstrates the shape of this triangular distribution by generated 300 random numbers
from the distribution. Then VOICE simulation was also repeated for 300 times, and a
distribution function was fitted as figure 5.8 (simulated). Comparing simulated results
against observed results, VOICE is able to generate a rational result of work duration.

100

200

300

Mean
Std Dev
Std Err Mean
Upper 95% Mean
Lower 95% Mean
N
Simulated

400

50

177.88884
64.766086
3.7392717
185.24746
170.53021
300

100 150 200 250 300 350 400
Mean
Std Dev
Std Err Mean
Upper 95% Mean
Lower 95% Mean
N
Empirical

153.33333
60
4.4832314
195.14583
177.50046
300

Figure 5.8 Distribution of work duration of the representative scenario
The second indicator compared was work quality, i.e., mistake percentage. The simulation
was repeated 300 times in VOICE, and the mistake percentage values were collected to fit
a distribution (figure 5.9, simulated). As shown, the simulated mistake percentage follows
a lognormal distribution, with a heavy left tail. It means in most cases, the final mistake
percentage of the proposal development is relatively at a lower level; bigger mistake
percentage only happens in several extreme cases. However, it was very difficult to
collect empirical data of mistake percentage to validate the simulated results directly,
153

mainly because domain experts tended to have very different definitions for “mistake”,
and in most cases, it was impossible to judge how much mistakes have been committed.
Therefore, this research employed an indirect way to measure the quality of work. In case
study companies, the author found that decision makers used an indicator called PF
(performance factor) to evaluate the cost performance of the projects. PF is the ratio of
actual cost versus budget. If actual cost is more than the developed budget, PF will be a
number bigger than 1. On the other hand, PF will be smaller than 1 if actual cost is less
than the developed budget. The absolute value of (PF-1) can also be used to estimate the
goodness of proposal development: if the absolute difference between actual cost and
budgeted cost of a project is bigger, it is always a signal that more mistakes are committed
during the proposal development. This research collected PF values of 79 projects, and
fitted a distribution to the absolute values of (PF-1). Result indicates a Pareto distribution
as shown in figure 5.9 (empirical). The shape of this Pareto distribution is very similar to
the lognormal distribution fitted from simulated results, with both having a heavy left tail.
It is therefore confirmed that VOICE is able to reflect the work quality of the simulated
case, although not with a direct evidence.

154

0.02

0.02

0.04

0.06

0.08

0.1

0.12

0.12

0.14

Mean
Std Dev
Std Err Mean
Upper 95% Mean
Lower 95% Mean
N

0.16

0.18

0.2

0.22

0.22 0.05

0.0375956
0.0279473
0.0016135
0.0407709
0.0344203
300

0.55

Mean
Std Dev
Std Err Mean
Upper 95% Mean
Lower 95% Mean
N
Empirical

Simulated

1.05
0.2632742
0.2315864
0.0260555
0.3151467
0.2114016
79

Figure 5.9 Distribution of mistake percentage of the representative scenario

The third indicator compared was effectiveness. In a survey, 34 respondents were asked to
provide information of their nonproductive activities, i.e., activities not associated with
direct processing of tasks, such as communication and waiting time. The author also
conducted a work sampling for a month on a consecutive basis. Every day, 8 am through
5 pm, the number of people talking face-to-face or in a meeting was documented every 15
minutes. These results were integrated to obtain PDF of effectiveness in real project teams
(figure 5.10 empirical). Then 300 simulations were conducted in VOICE to collect
simulated effectiveness values. Results demonstrate a very similar distribution (figure
5.10 simulated) to the empirical data. The mean value of the simulated effectiveness is
0.55, which is slightly smaller than the observed value 0.62. Considering the errors in the

155

survey and work sampling, such difference is acceptable. It indicates that VOICE is able
to reproduce work effectiveness.

0.15

0.50
Mean
Std Dev
Std Err Mean
Upper 95% Mean
Lower 95% Mean
N
Simulated

0.85 0.15
0.5464378
0.1216686
0.0070245
0.5602616
0.532614
300

0.50

0.85

Mean
0.617598
Std Dev
0.0754136
Std Err Mean
0.0129333
Upper 95% Mean
0.6439111
Lower 95% Mean
0.591285
N
34
Empirical

Figure 5.10 Distribution of effectiveness of the test scenario

The last experiment was aimed at validating VOICE’s ability to reproduce the key
behaviors of the system. One important behavior is the number of communication
iterations of between two team members to finish a task. For example, in order to develop
a detailed quantity take-off, the estimator needs to obtain design information from the
design team. But such information request can seldom be fulfilled in just one time. Indeed,
many “back-and-forth” communications could occur during this process. In a survey, 34
respondents reported the most likely number of iterations in their work. These results
were fitted against a Poisson distribution as shown in figure 5.11 (empirical). Simulation
results (300 data points) were also obtained, and a distribution was fitted as shown in
156

figure 5.11 (simulated). Two distributions demonstrate similar shape, with mean values
both around 4. This indicates VOICE is able to reproduce the key behaviors happened in
the tested scenario.

1

2

3

4

5

6

7

Mean
Std Dev
Std Err Mean
Upper 95% Mean
Lower 95% Mean
N
Simulated

8

9

10 11 12 1

4.0667667
1.004348
0.0579861
4.1808792
3.9526542
300

2

3

4

5 6

7

8

Mean
Std Dev
Std Err Mean
Upper 95% Mean
Lower 95% Mean
N
Empirical

9 10 11 12
3.8823529
1.9503645
0.3344847
4.5628673
3.2018386
34

Figure 5.11 Distribution of number of iterations of the representative scenario

5.3. Summary
This chapter introduced the verification and validation framework for VOICE. The
verification and validation aims to support VOICE’s ability to serve its designed purpose,
and generate results that are similar enough to the observations. Certain validations were
made indirectly as a result of an overall consideration of the validation quality and cost
associated with the validation process (Rojas-Villafane 2010). Although only limited
validation has been achieved by modeling a representative case of a studied company,
certain level of confidence has been obtained that current VOICE can be used to conduct
157

exploratory simulations since it is able to generate similar results as observed for a
representative case.

158

CHAPTER 6: CASE STUDIES
This chapter uses two case studies to illustrate the capacity of VOICE to examine
interpersonal cooperation issues in construction project teams. Each case study is
organized in the following structure:

•

Case overview: introducing the background of the case study and identifying the
problems to be addressed;

•

Model assumption: converting problems to the VOICE model and outlining the
major modeling assumptions;

•

Model development: tailoring the generic VOICE framework to specific
problems, and formulating parameters used in simulation;

•

Simulation results: presenting the simulation results

•

Interpretation of simulation results: an overall summary and interpretation of
the simulation results

6.1. Case Study 1: intra-team cooperation in estimating activities
This case study investigates how cooperative behaviors and the institutional arrangement
between members of a single project team affect management actions and team
performance in typical Design Bid Build (DBB) projects.

6.1.1. Case overview
Company D is a small developer located in Michigan focusing on design and program
development of residential projects. As a developer, this company bids for construction
projects, such as new construction or maintenance. Most jobs are Design Bid Build (DBB)
159

and each of them needs an accurate estimation before bid submission. The cost estimation
is conducted by a single team at company D: three managers work on separate
sections/crafts of the project and all report to the principals for the final estimating and
bidding decision. The following problems with D’s cost estimation effort were identified
during an interview with the principals:

•

Job acquiring capacity: In company D, potential jobs are acquired by two
principals. The estimation work is decomposed into a number of constituent
requirements and assigned to the team. It is very difficult for the principals to
“estimate” the proper workload for the team, i.e., how many jobs can be acquired
simultaneously. There is a realistic need to investigate the maximum capacity of
their team to work on estimation without overloading their employees.

•

Improper workload balancing: experience has shown that unbalanced workload
among team members may produce anxiety, inefficiency and unexpected lack of
productivity.

•

Meeting frequency: company D has a very flexible meeting policy. The
principals tend to hold more meetings; however, this might bring new
inefficiencies to the company. Complaints had begun to appear that too many
meetings are purely wastes of time.

These issues were investigated with VOICE. Because company D had very concrete needs
and data of specific projects, what-if scenario simulations were performed.

160

6.1.2. Model assumptions
In order to utilize the VOICE framework to investigate problems discussed above, the
following basic assumptions were made:

First, the job acquiring decisions of the principals may trigger a variety of individual
behavioral responses modeled with the relational behaviors of the proposed VOICE
framework. Typical behaviors include routine activities (e.g., processing tasks),
communication, and coordination (e.g., assigning tasks). However, when overloaded,
reciprocal activities may also be triggered, such as complaining about the overload. These
nonproductive activities create inefficiency and affect the capacity of the estimating team.

Second, it was assumed that an unbalanced workload among team members was primarily
attributed to the manager’s task assignment preference. Therefore, to investigate the
workload balancing issue, an investigation is needed to reveal the role of managers’ task
assignment preferences.

Third, in the estimating process, principals initiate meetings when too many technical
errors and exceptions are perceived. Meetings bring direct changes to the team by
influencing the magnitude of team members’ behavioral responses. These interactions
ultimately determine the quality of interpersonal cooperation in the estimating team and
then, the performance of estimation.

Last, although under the VOICE framework, task characteristics and organizational
context can also affect the cooperative behaviors of team members, they will not be
161

considered in this case study because they are less dynamic in Company D, compared to
the four issues addressed by the principals. Therefore, the simulation experiments only
focused on the controllable variables for a realistic recommendation.

Based on the basic assumptions, the problems can be modeled by VOICE as shown in
Figure 6.1.

Influence

Figure 6.1 Modeling case 1 with proposed conceptual framework

6.1.3. Model development

6.1.3.1. Inputs

Based on time sheets, semi-structured interviews (N=2) and online surveys (N=8)
administered to company employees as process and task owners, data needed for the
162

simulation experiments was collected and the simulation model was built as Figure 6.2.
The case study model includes three simultaneous bidding projects: a set of tasks is
generated by each project. For example, task 1 is “principal studies potential opportunity,”
task 2 is “project planning manager contacts subcontractors,” etc. Based on predefined
TART tables and user defined attributes, task attributes are embedded in the model,
shown as arrays. For example, for project 1, the array for the tasks’ work amount is
[5,5,10,15,15,10,20,20,25,10,5,5,10,15,20,15,10,10,5,5].

6.1.3.2. Process

Company D has a typical three-level organizational or report structure: Two principals
acting as the head of the company are responsible for scope management, new job
acquisition, negotiation and general management of the company. Note that, since the job
responsibilities and decision making processes of these two principals are similar, they are
considered as one role, i.e., president, in the company. Then, there are three full-time
managers with the assistance of several full-time/part-time helpers.

The individual work related behaviors and decision processes were also embedded in the
simulation model. Specific inputs are shown in Figure 6.2. In the following section, two
examples are given to demonstrate how VOICE helps with the investigation of the
influence of cooperative behaviors on team effectiveness. More details about the model
for case study 1 are provided in Appendix 1.

163

Figure 6.2 Snapshot of the simulation (case study 1)

6.1.3.3. Outputs

According to the needs of the four issues, the performance indicators monitored in the
simulation experiments include time (duration of estimation), quality (mistakes in
estimates), effectiveness and work related pressure.
164

6.1.4. Simulation results

6.1.4.1. Experiment 1: Influence of job acquiring decision

In the past, the estimating team had been able to work on three projects concurrently. To
test the maximum capacity, three-project and four-project scenarios were compared to
analyze the influence of extra work. The extra project (project 4) is assumed to have the
same parameters and processes as project 3, and all human factors, such as work
efficiency, stay the same under the two scenarios. Figure 6.3 shows the simulation results,
where the Y axis refers to the remaining work, and the X axis is the elapsed time
measured in hours, whereby one tick equals one man-hour of work in reality. Therefore,
the project’s bidding work progress is demonstrated.

The results indicate that when bidding three projects simultaneously, all bidding work can
be finished before the preset deadline (Figure 6.3). However, when bidding for four
projects, three of them cannot meet the deadline, even though only one project is added to
the work list. More importantly, the increase in work time is not a linear relationship with
the number of projects: the total work hour spent on three projects is 671; it becomes 1285
when another project is added which is almost twice of the original number. The
relationship between increased job load and estimated time for processing them can
hardly be discovered by experiment because of the interferences among projects.
Capturing the quantitative relationship is very helpful to answer a variety of practical
questions like: “how many additional projects can a construction team work on if it hires
165

3 new staff members?”

Such questions can hardly be answered by even the most

experienced practitioners. VOICE thus could be used as a supporting tool for job
acquiring decision makings.

(1) Bidding for 3 projects simutaneously

(2) Bidding for 4 projects simutaneously
Figure 6.3 Real time bidding work progress (scenario 1)
166

6.1.4.2. Experiment 2: Influence of task assignment preference

An unbalanced workload among team members had been attributed to managers’ task
assignment preferences. Therefore, two major preferences were investigated in a what-if
scenario simulation: (1) assigning tasks to staff with the least workload, and (2) assigning
tasks to the faster staff. It was found that assigning jobs to faster staff does not always
lead to faster completion. Figure 6.4 demonstrates that consistently assigning jobs to
faster staff increased total work time for all projects. Figure 6.5 shows the work related
pressure of staff. Under preference 1 (considering the workload), the biggest stress
indicator is 17.6 for Helper 4; but under preference 2 (always assign jobs to faster helpers),
this indicator is 21. According to the literature, greater work related stress reduces work
efficiency, and helpers may refuse to work under some conditions, which means the
manager must spend extra time to reassign the job. Managers’ coordination behaviors
affect work performance by affecting employees’ work-related pressure.
Remaining work

168

223

280

Time
(1) Assign job while considering helpers’ workload
Figure 6.4 Real time bidding work progress (scenario 2)
167

Figure 6.4 Cont’d
Remaining work

307
301

210

Time
(2) Aways assign job to faster helpers

Work pressure
17.6
15
12

Time
(1) Assign job while considering helpers’ workload
Work pressure

21
15
13.5

Time
(2) Always assign job to faster helpers
Figure 6.5 Work related stress of helpers (scenario 2)

168

6.1.4.3. Experiment 3: Influence of meeting

In Company D, principals have the right to hold emergent meetings when they perceive
too many management exceptions. These exceptions include but are not limited to:
approval requests from helpers and managers, reports on job burnout, lack of necessary
information, or conflicts between different departments. When the management
exceptions exceed the principals’ threshold, the principals initiate emergent meetings.
During the meetings, all employees must pause their work for a given time, although
(depending on how many problems must be solved), after the meetings, the problems are
solved. For example, an assistant estimator will find it much easier to obtain worker
payment information from the accounting department during the meeting; or the
principals will announce the rules of determining profitability during the meeting so that
employees do not need to ask for such information repeatedly. The principals want to
know, however, the impacts of the frequency of emergent meetings.

Two strategies were considered: (1) more emergent meetings: the threshold of principals
was set to 5, which means they are more sensitive to management exceptions; (2) fewer
emergent meetings: the threshold was set to 50, which means principals are more tolerant
of exceptions. More meetings does not necessarily lead to longer total work time: even
though holding emergent meetings may be perceived as “wasting” time, the simulation
results show that strategy 1 (more emergent meetings) reduces the total work time of all
three projects (Figure 6.6). The reason is more emergent meetings benefited the company

169

by increasing the accumulative work effectiveness (measured as a ratio of effective work
time versus ineffective work time, where total time is the sum of effective work time,
ineffective work time and idle time). Additionally, the effectiveness indicators under
strategy 1 are more stable than under strategy 2 (Figure 6.7). This means that holding
more emergent meetings does not affect employees’ work; rather, it improves efficiency.
Therefore, we advise the principals of company D to be stricter about management
exceptions, and hold emergent meetings as needed. A further examination found there is a
trade-off between time spent on meetings and additional efficiency gained from them.
Figuring out a better frequency of meetings is a difficult task for decision makers because
of the complex interactions of human behaviors, task characteristics and work processes.
VOICE could be used for this purpose because in its simulation these interactions have
already been captured as part of the emergent process.

(1) More meetings
Figure 6.6 Real time bidding work progress (scenario 3)
170

Figure 6.6 Cont’d

250

(1) Less meetings

%

(1) More meetings

Figure 6.7 Accumulative effectiveness of bidding work (scenario 3)

171

Figure 6.7 Cont’d

71%

(2) Less meetings

6.1.5. Interpretation of simulation results
Case study 1 examines the cooperation related issues of an estimating team on DBB
projects. In particular, the simulation experiments investigate the implications of a variety
of managerial actions. Results indicate that cooperation related behaviors and processes
are a critical connection between management actions and team performance in estimation
development. For example, in one simulation experiment, two types of coordination
preferences of managers are compared, i.e., whether current workloads of team members
are considered when assigning new tasks to them. Results indicate that if a manager
always assigns tasks to faster staff, in hopes that the entire job would be finished more
quickly, the ultimate result is often the opposite. The entire estimation job may actually

172

last longer if the workload is not well balanced among team members. Certain staff might
become over stressed and a variety of behavioral responses, such as complaining, may be
triggered. These responses add up to nonproductive time because additional efforts are
needed to solve the conflicts. In another experiment, the function of meetings is examined.
It reveals that in some cases, more meetings are helpful to improve team productivity,
although the meetings themselves take time. One possible explanation is that meetings
may enhance mutual understanding between team members, resolve conflicts, and reduce
unnecessary communication. The entire team is gathered in one place and problems may
be solved effectively. But if the decision maker over relies on meetings to solve perceived
exceptions (i.e., situations not expected or planned), the team might become excessively
inefficient because too much time would be spent on meetings. Clearly, according to the
findings in this case study, the relationship between managerial actions and ultimate team
performance is nonlinear - any managerial actions taken by the decision maker could
cause a series of unexpected behavioral responses from the team, which in turn alters the
expectations of the team performance. This relationship depends on the characteristics of
the jobs, particular organizational context, and behaviors of every member on the team.

In case study 1, no significant empirical data existed. A what-if scenario simulation
examined the possible outcomes of adopting certain managerial actions/strategies.
However, these managerial actions/strategies were not actually applied in the company
when the case study was ongoing. For example, the simulation compared the results of

173

different task assignment preferences. But in reality, it is very difficult, if not impossible,
to actually change managers’ behaviors and to observe the consequences.

Therefore, face validation was used in this case study to validate the findings of the
simulation experiments. As addressed by Banks and colleagues (2005), face validation is
used when there is a lack of empirical data, or for the purpose of reducing development
cost. It is achieved by “consulting people knowledgeable about system behavior on model
structure, model input, and model output. Use any existing knowledge in the form of
previous experience” (Banks et al. 2005).

Two principals of company D were interviewed, focusing on their perceptions of the
model’s validity and the simulation results. All possibilities reflected in the simulations
were fully discussed. They both agreed with the reasonableness of the simulation findings.
In particular, they agreed that: (1) adding one extra job might influence the completion of
other undergoing jobs; (2) workload is a critical decision point for managers and thus, the
proper level of work balancing is needed; and (3) meetings benefit the company by
solving issues, enhancing mutual understanding, clarifying needs, and encouraging
communication. One principal indicated that the VOICE simulation experiments were
useful for investigating performance enhancement opportunities in the company.

6.2. Case study 2: Inter-team cooperation in project proposal development
Case 1 investigated interpersonal cooperation within a single project team. Case study 2
extends the investigation to interpersonal cooperation in a cross-functional context where
174

the institutional and psychosocial conditions become more complex. In particular, VOICE
is utilized to examine the implications of goal incongruence in cross-functional
cooperation during project proposal development.

6.2.1. Case overview
The case study was conducted with a large construction company (denoted as Company
Z). The major focus of Company Z is the construction and maintenance of power plants,
nuclear plants and refinery facilities. In order to enhance its competitive ability in the EPC
market (Engineering, Procurement and Construction), Company Z acquired an
engineering design firm (denoted as ZENG) a couple of years ago to design of all of Z’s
new EPC jobs. At Z, proposal development is the sole responsibility of a Z’s project
proposal team (i.e., estimating team). But because of the specialty of work, Z’s proposal
team highly relies on the technical and quantity information from ZENG to form estimates
and develop proposals.

A series of face-to-face interviews with those in key roles at Company Z (including vice
president, department heads, coordinators and estimators) highlighted cooperation
difficulties between Z’s proposal development team and the engineering team of ZENG.
Z’s estimators found it increasingly difficult to obtain necessary information from their
ZENG counterparts. Problems included “late reply” and “inaccurate or incomplete
information” from the engineering team, “lack of trust,” “lack of integration of processes,”
“different technical terms and knowledge set,” etc. As a result, frustration was spreading

175

in the proposal team and had led to serious performance issues. When asked about the
reasons, although specific answers were different, most respondents attributed the current
cooperation issues between the proposal and engineering team to “unclear statement of
the objectives,” “misuse of definitions” and “very different understanding of needs.” This
suggests that the “incongruent targets between proposal team and the engineering team”
was the biggest driving factor of the inefficient cooperation between the two teams. A
survey (N=34) to process owners in company Z confirmed the above observations and
highlighted key concerns/interests in the company, which constitute the major themes of
the simulation experiments using VOICE:

•

Effects of goal incongruence: 65% of respondents believed the cooperation
issues between the proposal team and engineering team to be based on “different
focuses” of the two teams. In particular, the engineering team’s goals were
structural integrity, constructability and robustness. The proposal team’s goals
were accurate and efficient proposals. Such goal differences could lead to a
divergence in behaviors.

•

Additional implications of time pressure: 91% of the respondents claimed time
to be the first priority of proposal development. They believed that a sense of time
pressure can significantly affect the quality and efficiency of the cooperation
between two teams.

•

Additional implications of task dependence: 78% of the respondents claimed
task dependency to be a critical variable in the cooperation between the proposal
176

team and the engineering team. It was believed that a higher level of task
dependence requires more cooperation between two teams.
•

Additional implications of micro-management: the management of Company Z
wanted to understand if the goal congruence is mitigated or amplified by the
management style of managers, especially their preference for micro-management.

6.2.2. Model assumptions
In order to model this case in VOICE, a set of assumptions were made. First, it was
assumed that expectations in this case are mainly generated by information exchange.
When a project proposal team member is working on proposal development, he/she
requires intensive information from the engineering team. An exception occurs when
available information is insufficient for the responsible proposal team member to perform
the task (Galbraith 1973), which leads to additional coordination, communication or
reassignment activities at the immediate expenses of time and cost or quality. Goal
congruence plays a vital role in this process, which is demonstrated in the difference of
the perceptions of behavioral standards and ranking of management criteria (Thomsen et
al. 2005). Goal congruence can affect the quality and amount of the appropriate
information contributed by the engineers because a higher magnitude of goal congruence
is anticipated to enhance the understanding among team members (Witt 1998). Thomsen
et al. (2005) model goal congruence as a percentage, with 100% being the most congruent
condition and 0% being the least. This case study uses the same definition and assumes a

177

linear relationship between goal congruence and information quality/amount exchanged
between an engineer and a project proposal team member.

Second, it was assumed that a variety of micro-level behavioral and instructional
responses occurs when goals between two teams are incongruent. The most immediate
effect is the behavioral responses of team members, including:(1) non-conformance: Kunz
et al. (1997) found that if no identical goal exists, team members tend not to follow the
approaches which others might favor or prescribe. The final performance may be
imperfect if such non-conformance is not remedied; (2) passive responses: team members
may react to the consequences of goal incongruence, including correcting or reworking
tasks with perceived errors and counteracting against nonproductive non-conformance.
These behavioral responses will improve the quality of decision making, but at the direct
expense of time and cost (Witt 1998); and (3) proactive responses: new evidence from
psychosocial research (e.g., experimental findings) finds that goal incongruence at the
intermediate level encourages actors to study a wider range of alternatives, and finally,
mutually agreed upon solutions might be found (Jehn 1995; Amason 1996). It also
facilitates the understanding and interpretation of the needs pertaining to the considered
solutions and encourages actors to participate and fully enage in the pursue of better
achievement of common goals (Kunda 2006). In addition to behaviroal responses,
evidence reveals that goal incongruence may lead to a variety of institutional reactions,
including

“coordination

effort,”“selective

178

authority

delegation,”

“monitoring,”

“steamrolling” and,“politicking.” Thomsen et al. have an excellent discussion on the
details of these institutional reactions (Thomsen et al. 2005).

Third, it was assumed that a series of factors may intensify or mitigate the effects of goal
incongruence. This has been supported by numerous organizational and psychosocial
findings. The most influential variable is task characteristics. Evidence suggests that a
high level of interaction and task dependence intensifies the influence of goal
incongruence (Schmidt and Kochan 1972; Gladstein 1984; Jehn 1995). The other
influential factor for goal incongruence is individual personal traits, such as competence,
experience and managerial preferences (Mintzberg 1983; Pfeffer 1994; Smith et al. 1994;
Burton et al. 1998; Thomsen et al. 2005). For example, Pfeffer (1994) found the
competence level of an individual to determine the manner of coping with conflicts from
goal incongruence – people with higher expertise tend to appeal to a higher authority
instead of negotiating with the counterpart. Thomsen et al. (2005) suggest that managerial
preference may mitigate or amplify the effects of goal congruence.

Ultimately, the consequences of goal incongruence – behavioral and institutional
responses, influential variables including task and individual characteristics – determine
team performance, including time (Thomsen 1998), quality (Thomsen et al. 2005) and
others. Under the proposed conceptual framework, behavioral and institutional responses
triggered by goal incongruence can be modeled by the relational behaviors of VOICE.
Task characteristics and individual attributes serve as the influential factors.

179

Organizational context forms the context of the micro-level behavior processes. These
interactions ultimately determine the quality of inter-team cooperation and then the
performance of proposal development. Figure 6.8 summarizes the modeling of case 2 with
the proposed framework.

Influence

Figure 6.8 Modeling case 2 with the proposed conceptual framework

180

6.2.3. Model development

6.2.3.1. Inputs

A survey was conducted to 34 process owners in company Z to collect data with respect
to task characteristics and individual traits. Then probability distribution functions (PDFs)
were fitted to the data. Table 6.1 summarizes the PDFs of key variables. For more details
of the input data, please refer to Appendix 2.

Table 6.1 Fitted PDFs of key variables of individuals and proposal development task
Task characteristics
Work amount
Priority
Difficulty
Authority level
Dependence
Information
Starting address

PDF
Normal (24, 4)
Uniform (1, 3)
Uniform (1, 3)
{1,2,3}~{60%, 30%, 10%}
Poisson (4.06)
Normal (0.2, 0.15)
Ceiling (Uniform (0,5))

Individual information
Competence
Capacity
Processing error
Position level
Assigning preference
Exception handling time
Quality/Info threshold
VP’s exception threshold
Reactions to exceptions

PDF
Two peaks normal
Normal (120, 8)
Normal (0.05,0.01)
1,2,3
3
Lognormal(0.335,0.769)
Johnson SI (2.29, 0.92)
20
Report 45% Return 45% nothing
10%

Monte Carlo simulation was performed in this case study to fully investigate the
implications of goal congruence. Therefore, task dependence was used as a probability
that determines whether a newly generated task can be processed or not when preceding
181

tasks are ongoing. Task information refers to a percentage of the amount of information,
where 100% means perfect fulfillment of the information need. Because the assigning
preference of the manager/coordinator is not the interest of this case study, it assumes that
all coordinators will assign tasks to the staff with the least work pressure, which makes
sense in most cases.

6.2.3.2. Process

Based on the responsibility description, there are four roles in a proposal team: (1) Vice
president of proposal team, whose responsibilities include the overall control of
estimating work of all projects, the validation and approval of the final proposal, and
initializing and organizing meetings within the proposal team or across teams; (2)
Proposal coordinators, who integrate the craft estimates, check and validate the estimates
submitted by discipline estimators, coordinate tasks, and solve the conflicts between
estimators and/or between estimating team and engineering team. The estimating
coordinator’s job is done on a project basis; (3) Estimators of all disciplines. Their jobs
include quantity take-off, productivity study (estimating the work hour/quantity), pricing
(quoting prices from vendors), validation of estimate (comparing estimates with historical
similar jobs) and estimate entry (entering the final craft estimates to the information
system); and (4) Engineers, who do not participate in the proposal directly, but provide
necessary design information to the proposal development team. The simulation considers
a simple scenario, i.e., five proposal coordinators and corresponding staff members are

182

considered in the simulation. Such simplification will not affect subsequent experiments
and analyses, since it well represents the work settings of the proposal team. Similarly, the
engineering team’s structure is also simplified to capture the actual work environment
(Figure 6.9).

Figure 6.9 Snapshot of the simulation (case study 2)

6.2.3.3. Outputs

Four performance indicators are selected as the index in the experiments, including
efficiency, effectiveness, quality and work related pressure.

183

6.2.4. Simulation results

6.2.4.1. Experiment 1: Direct influence of goal congruence on team performance

This experiment is interested in the influences of goal congruence on the performance of
the proposal development in two situations: (1) without time pressure, and (2) with time
pressure. The first scenario aims to reveal the pure influence of goal congruence, while
the second scenario is also interested in the additional implications of time pressure
perceived by team members. Following Pruitt and Drews’ findings (1969), a sigmoid
function was used to measure time pressure.

Influence on efficiency

The simulation results (Figure 6.10) indicate there is a strong linear correlation between
the magnitude of goal congruence and the efficiency of the proposal development without
time pressure (p-value<0.0001). A quadratic or cubic ANOVA test does not improve the
corresponding F-ratio. This conclusion contradicts some previous studies (such as
Thomson et al, 2005) where a nonlinear relationship was observed. A possible
explanation is that in proposal development, the quality and amount of the information
exchanged is strongly tied to the magnitude of goal congruence, which amplifies the
influences of goal congruence on work efficiency (influences of other factors are weaker).

Figure 6.10 indicates that the efficiency of proposal development becomes more stable
under different levels of goal congruence (p-value=0.893) when time pressure is

184

noticeable. When the level of goal congruence is smaller than 0.8, values of efficiency
with time pressure are significantly larger than those without time pressure. The
distinction between the values of efficiency becomes ignorable when the level of goal
congruence is greater than 0.8 (Table 6.2). This finding means time pressure can
significantly improve the efficiency of proposal development under a lower level of goal
congruence. This observation has been ignored in practice – according to the interviews to
the key decision makers of company Z, it was believed that sharing goals between the
proposal team and the engineering team can always improve productivity of the proposal
development. However, this simulation experience found if there is a specific deadline for
proposal development, proposal team member would somehow work really hard to meet
the deadline, no matter how well the goal has been shared between them and engineers.
As a result, efficiency remains in a certain level. VOICE revealed some facts that could be
easily missed by key decision makers.

Table 6.2 ANOVA tests under different levels of goal congruence (efficiency)

185

8.89

0.073

3.04

0.0042

0.7879

0.0866

1.0

13.67

0.9

0.0005

0.8

18.86

0.7

<0.0001

22.89
<0.0001

0.6

31.31

56.45
<0.0001

0.5

<0.0001

92.59
<0.0001

0.4

62.26

0.3

<0.0001

0.2

F-ratio

0.1

p-value

GC

Efficiency of proposal development
Without timepressure

With timepressure

6
Work/std.time

5
4
3
2
1
0
0.1

0.2

0.3

0.4 0.5 0.6 0.7
Goal congruence

0.8

0.9

1

Figure 6.10 Influence of time pressure on efficiency of proposal development

Influence on effectiveness

The influence of goal congruence on effectiveness of the proposal development is
nonlinear when there is no time pressure (Figure 6.11). An ANOVA analysis indicates a
quadratic relationship gives better p-value (p-value<0.0001). Increases to work
effectiveness will slow down when goal congruence becomes bigger, and when goal
congruence is over a certain level (0.8 in this case), the effectiveness becomes stable. It
means the marginal diminishing effect exists in the additional contribution of goal
congruence to work effectiveness. A realistic interpretation of this finding is that when
goal congruence is greater than a certain level, the additional effectiveness gained is
attributed more to other factors than to shorter communication and routine time.

186

However, effectiveness becomes more stable when time pressure matters (Figure 6.11; pvalue=0.782), meaning the influence of goal congruence becomes ignorable. Again, the
distinction between the effectiveness is significant when the level of goal congruence is
smaller than a certain level (0.6). This indicates that time pressure offsets the influence of
goal congruence. This finding has provided an explanation to a previous observation, that
goal congruence doesn’t benefit efficiency of the proposal development when time
pressure matters. It is likely that in order to meet the deadline, the proposal team tends to
reduce nonproductive time such as quality monitoring, meetings and communication. As a
result, the effectiveness of the proposal development keeps in a relatively high level. This
experiment has revealed the link between human behaviors and the global team
effectiveness, which could be easily ignored in practice.

Effectiveness of proposal development
Without time presure

With timepressure

100%
Effectiveness

80%
60%
40%
20%
0%
0.1

0.2

0.3

0.4 0.5 0.6 0.7
Goal congruence

0.8

0.9

1

Figure 6.11 Influence of time pressure on effectiveness of proposal development
187

Table 6.3 ANOVA tests under different levels of goal congruence (effectiveness)

10.56

3.37

0.58

0.01

0.80

1.49

0.0019

0.0714

0.4487

0.9243

0.3743

0.2277

1.0

12.23

0.9

0.0009

0.8

32.13

0.7

<0.0001

0.6

92.47

0.5

<0.0001

0.4

169.20

0.3

<0.0001

0.2

F-ratio

0.1

p-value

GC

Influence on quality

No significant relationship was observed (p-value=0.5521) for the influence of goal
congruence on the quality of proposal development when there is no time pressure (Figure
6.12). This observation contradicts previous studies, such as Thomson et al. (2005). The
reason is that proposal development is highly quality-driven: in the studied case, the
coordinators were responsible for the accuracy of the developed proposal and always set a
rational expectation for the estimates. If the estimates highly diverged from a reasonable
range according to the historical data, the coordinators would request a re-work or
correction in most cases. This may sacrifice the efficiency of proposal development for
additional communication and routine activities, but it ensures the quality of the final
proposal (Figure 6.12).

This experiment demonstrated that time pressure significantly increases the risk of
committing more mistakes in proposal development (Figure 6.12). As shown, the average
mistake percentage tripled under time pressure. This observation is true at all levels of
188

goal congruence with a small statistical significance level (Table 6.4). This means that
improvements to the efficiency and effectiveness of proposal development sacrifice the
expected quality of the delivered proposal. This observation is corresponding to previous
findings about the change to the efficiency and effectiveness when time pressure matters.
Because there is a deadline to meet, the proposal team members tend to reduce time for
quality monitoring by relaxing their quality threshold. The consequence is the team will to
commit more mistakes. This is another example that how human behavior and attitude can
affect team performance under different situations. Such a finding builds on an overall
consideration of the micro-level processes and the transition to the global performance,
which can be hardly captured by human experience.

Quality of proposal development
Without timepressure

With timepressure

Mistake percentage

30%
25%
20%
15%
10%
5%
0%
0.1

0.2

0.3

0.4 0.5 0.6 0.7
Goal congruence

0.8

0.9

1

Figure 6.12 Influence of time pressure on quality of proposal development

189

Table 6.4 ANOVA tests under different levels of goal congruence (quality)

90.14

62.55

73.72

43.89

75.48

66.21

<0.0001

<0.0001

<0.0001

<0.0001

<0.0001

<0.0001

1.0

85.63

0.9

<0.0001

0.8

72.52

0.7

<0.0001

0.6

70.41

0.5

<0.0001

0.4

62.41

0.3

<0.0001

0.2

F-ratio

0.1

p-value

GC

Influence on work related pressure

Another finding of this experiment reveals that goal congruence may hardly affect the
work pressure of employees when there is no time pressure (p-value=0.6769). Figure
6.13 shows that the values of the averaged work pressure of all proposal team members
stay within a stable range between 0 to 50. A further investigation was conducted for each
team member. Results indicate for individuals, the level of goal congruence remains
ignorable comparedto work pressure, most likely because work pressure is mainly
determined by task characteristics (such as work amount and difficulty). Improved goal
congruence is not helpful fpr reducing the accumulated work amount of the proposal team
members.

An experiment was also conducted to check if the averaged work related pressure changes
when time pressure matters. The result offer a negative answer – there is no significant
difference between work related pressures under the two scenarios, i.e., with/without time
pressure (Figure 6.13). This finding has revealed a fact that couldn’t be easily found, that
work pressure is mainly dependent on the technical characteristics of tasks, such as task
190

amount and difficulty. Although an F test is passed when goal congruence equals 0.4
(Table 6.5), it can be interpreted as a systematic error due to the relatively small size of
the sample.

Average work related pressure
Average work related pressure

Without timepressure

With timepressure

50
40
30
20
10
0
0.1

0.2

0.3

0.4

0.5
0.6
0.7
Goal congruence

0.8

0.9

1

Figure 6.13 Influence of time pressure on averaged work pressure

Table 6.5 ANOVA tests under different levels of goal congruence (averaged work
pressure)

191

0.042

0.18

0.055

0.8377

0.6716

0.8147

1.0

0.087

0.9

0.7682

0.8

0.65

0.7

0.4205

13.01
0.0003

0.6

0.098

0.98
0.3227

0.5

0.7549

0.12
0.7292

0.4

0.50

0.3

0.4793

0.2

F-ratio

0.1

p-value

GC

6.2.4.2. Experiment 2: Additional implications of task dependence

This experiment investigates whether task dependence has additional implications for
performance, considering the existing effects of goal congruence. Recent findings of
organizational science have revealed that task dependence may intensify or offset the
effects of goal congruence (Schmidt and Kochan 1972; Gladstein 1984; Jehn 1995;
Thomsen et al. 2005). To examine this, a series of sensitivity analyses were conducted.

Influence of goal congruence + task dependence

Figure 6.14 shows the combined influence of goal congruence and task dependence. It
indicates that task dependence exerts stronger influences on the efficiency, effectiveness
and quality of proposal development (p-values<0.0001) compared to goal congruence.
One interpretation can be made from a technical standpoint: higher task dependence
means tasks can hardly be processed concurrently and thus, it is associated with longer
waiting times. This further reduces the efficiency and effectiveness of proposal
development. It also suggests that if tasks are more dependent on one another, there is a
bigger chance that more mistakes will be committed. However, like the finding in
experiment 1, task dependence can hardly affect the work pressure of team members.

Amplifier of goal congruence

Then, a series of ANOVA tests were performed to check if task dependence intensifies or
mitigates the effects of goal congruence, as concluded by the literature. The result shows

192

that the F ratio tends to become bigger when level of task dependence increases (Figure
6.15). Since the F ratio is an indicator showing the significance of difference, it can be
interpreted that the difference of efficiencies under different levels of goal congruence
are amplified. What’s worth noting, to reach this finding requires an overall
consideration of multiple factors including behavioral (e.g., goal congruence) and
technical (e.g., task dependence) factors. It can be regarded as an emergent phenomenon
since it is the consequence of the interactions of multiple factors.

0.8
4

0
0

0

0
1

1

1 1

0

0

12

0.05

0.01

1

0
1

0

0

1
1

0

Figure 6.14 Influence of goal congruence and task dependence
193

This conclusion is consistent with previous studies, such as the findings of Thomson et al.
(2005). Similar findings were obtained with respect to effectiveness of proposal
development (results are not shown in this thesis). However, the effects of goal
congruence on the quality of proposal development can hardly be altered by task
dependence. This is also associated with the quality expectation set by the coordinators.

F ratio

Magnitude of difference w.r.t efficiency
300
250
200
150
100
50
0

0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
1
F ratio 34.49 87.32 140.8 129.3 120.5 85.66 129.3 141.3 153.9 115.7 259.5
Task Dependence

Figure 6.15 Influence of task dependence on changes to efficiency under different levels
of goal congruence

6.2.4.3. Experiment 3: Additional implications of micro-management preference

Management styles can be grouped into two categories according to how coordinators
involve themselves in the decision making and managerial actions; the categories are
micro-management and not (Mintzberg 1983; Burton et al. 1998). Micro-management is
the custom of being heavily engaged in the daily affairs and specific tasks of subordinates
while the opposite is giving a degree if autonomy to subordinates. This experiment
194

examines the additional implications of micro-management. The magnitude of micromanagement is measured with the acceptable number of iterations for information
exchange before raising the issue to coordinators (Pfeffer 1994; Kristof-Brown and
Stevens 2001). A smaller acceptable number of interactions means the coordinators prefer
to micro-manage.

Influence of micro-management + goal congruence

Figure 6.16 demonstrates the combined influences of goal congruence and micromanagement preference (acceptable number of iterations) on performance. Micromanagement and goal congruence between teams together can alter the shape of
performance landscapes.

0.7
1.3

0.9
1

0.3
1
0

0
0 15

0 15

Figure 6.16 Influence of goal congruence and micro-management

195

Figure 6.16 Cont’d

0.05
3

15
0.036

2
15

0.2

0.2
1 0

1

0

A further examination with Figure 6.16 and ANOVA analysis (Table 6.6) found that
micro-management’s influence shows different features under different levels of goal
congruence:

•

Efficiency: The influence of micro-management on efficiency changes according
to level of goal congruence. When goals are less congruent between two teams
(0.1 and 0.2), micro-management can help improve efficiency. But when goals are
highly congruent between two teams (0.8 and 0.9), too much micro-management
indeed hurts efficiency. Otherwise, there is no significant relationship between
micro-management and efficiency. The effects of micro-management on
efficiency might be opposite depending on the level of goal congruence. In
practice, it means that if the proposal team and the engineering team share the
same goal, the coordinator should give more freedom to the team members to
handle exceptions in order to improve productivity. In contrast, the coordinator
should get involved more in the team’s daily issues if the goal is different between
196

two teams. This is the consequence of the interaction between goal congruence
and micro-management, and affected by the emergent process from these bottom
behaviors to overall team performance. It supports that behavioral and attitudinal
factors can significantly affect project team performance. Such relationship is
difficult to conceal by pure experience.
•

Effectiveness: The effect of micro-management on effectiveness also depends on
level of goal congruence. When goals are less congruent, such as at a level of 0.1
or 0.2, micro-management improves effectiveness. Otherwise, micro-management
sacrifices effectiveness. This indicates that micro-management helps with
effectiveness only when goals are incongruent. Again, it is an example of how
behaviors and attitudes of project team members may lead to unexpected team
performance.

•

Quality: Result shows that autonomy sacrifices quality in most situations. Micromanagement can always help reduce mistakes. However, this is not true when the
goals of two teams are highly congruent (e.g., greater than 0.9). In this case,
micro-management will slightly increase the chance of committing more mistakes.
This indicates that when teams share the same goals, micro-management leads to
mistakes. The relationship between quality of the proposal development and
micro-management is nonlinear, depending on the level of goal congruence. Such
nonlinearity can be regards as an emergent property of the cooperation between

197

the proposal team and the engineering team since it is a result of the interactions of
cooperative processes, and lead to unexpected finding.
•

Work related pressure: the ANOVA indicates there is a significant relationship (pvalue<0.0001) between micro-management and work related pressure at each level
of goal congruence: less micro-management or higher level of autonomy for the
staff means a higher level of work related pressure.

Table 6.6 p-values of micro-management’s influence under levels of goal congruence
Congruence Efficiency Effectiveness Quality Pressure
0.1
<0.0001*
<0.0001* <0.0001* <0.0001*
0.2
0.0002*
<0.0001* <0.0001* <0.0001*
0.3
0.0064
<0.0001* <0.0001* <0.0001*
0.4
0.3628
<0.0001* <0.0001* <0.0001*
0.5
0.3477
<0.0001* <0.0001* <0.0001*
0.6
0.479
<0.0001* 0.0134* <0.0001*
0.7
0.1075
<0.0001*
0.8255 <0.0001*
0.8
0.1533
<0.0001*
0.3698 <0.0001*
0.9
0.0003*
<0.0001*
0.2294 <0.0001*
1.0
0.0024*
<0.0001* 0.0001* <0.0001*

Influence of micro-management + task dependence
The experiment also examined the combined impacts of micro-management and task
dependence on performance. Figure 6. 17 demonstrates the results of uncertainty analysis
based on 52,800 simulations.

198

0.5
5

15

15

0
1

1
0

0

0.1

0

14

0.05

0.01
1

15

0

0
0 15

1

0

Figure 6.17 Influence of goal congruence and micro-management

An examination of Figure 6.17 and the ANOVA reveal the differing effects of micromanagement under different levels of task dependence (Table 6.7).

•

Efficiency: the influence of micro-management becomes less noticeable when task
dependence is considered. Only when tasks are very independent (task dependence
is 0 through 0.2), is micro-management able to improve efficiency; otherwise, it
exerts no influence. This indicates that micro-management is beneficial only when
tasks are highly dependent. This finding has demonstrated the interaction between
199

technical factor (task dependence) and behavioral factor (micro-management), and
highlighted the influence on project team performance. It supports that different
aspects of cooperation in construction settings may act together to generate
unexpected performance.
•

Effectiveness: similar to efficiency, the influence of micro-management on the
effectiveness of proposal development is not significant when task dependence is
considered. Again, this finding requires the consideration of both technical aspect
of proposal development and behavioral aspect of management. VOICE provides a
possibility to quantify the combined influences of interdependent factors of
cooperation.

•

Quality: autonomy sacrifices quality. When coordinators prefer the autonomy of
team members, the team will commit more mistakes. Worth noting, however, is
that the opposite trend occurs when task dependence equals 0, and is due to the
abnormal data points in the simulation.

•

Work pressure: ANOVA does not show a significant relationship between micromanagement and work related pressure under most task dependence levels. Only
when tasks are highly independent (dependence is smaller than 0.4) do the results
show that micro-management can reduce work related pressure.

200

Table 6.7 p-values of micro-management’s influence under levels of task dependence
Dependence Efficiency Effectiveness Quality Pressure
0.0
<0.0001*
<0.0001*
0.003* <0.0001*
0.1
0.0027
<0.0001* <0.0001* <0.0001*
0.2
0.0343
0.0238* <0.0001* <0.0001*
0.3
0.0789
0.0862 <0.0001* 0.0253*
0.4
0.1237
0.1658 <0.0001* 0.0448*
0.5
0.3364
0.0822 <0.0001*
0.1864
0.6
0.6238
0.1561 0.0134*
0.3762
0.7
0.1595
0.1231 0.0013*
0.4579
0.8
0.4423
0.0684 0.0002*
0.2688
0.9
0.5864
0.0402* 0.0002*
0.3357
1.0
0.3522
0.0355*
0.0052
0.4238
6.2.5. Interpretation of simulation results
This case study extends the investigation to interpersonal cooperation in a cross-functional
context where psychosocial conditions become more complex, especially as the
implications of goal congruence become obvious. Several findings are as follows.

First, the simulation finds goal congruence to be an influential factor for team productivity,
but negligible to the work quality and work pressure of the project team. First, a higher
level of goal congruence between the proposal team and engineering team significantly
improves the efficiency and effectiveness of proposal development. A likely interpretation
is that enhanced goal congruence improves the mutual understanding of objectives,
definitions and needs between two teams, and encourages proactive participation of the
engineers in proposal development.

This in turn reduces the need for additional

coordination, and increases the quality of each information exchange between engineers
and proposal team members. To be noted, further investigation finds that time pressure
201

can alleviate the effects of goal congruence, in that efficiency and effectiveness under
different levels of goal congruence show little difference, unlike that observed in
scenarios without time pressure. This is because, under serious time pressure, team
members tend to be more tolerant of the quality of obtained information and/or tasks, and
coordinators are inclined to reduce the frequency of coordination. These behavioral
changes indeed improve the efficiency and effectiveness of proposal development,
although at the expense of quality. Such findings highlight the positive role of aligning
goals between different teams in improving team productivity. On the other hand,
however, goal congruence exerts no immediate influence on the quality of proposal
development and does nothing to the work pressure of team members. Proposal
development is highly quality-driven: coordinators are responsible for the accuracy of the
developed proposal and always set a rational expectation for the estimates. If the estimates
greatly diverge from a reasonable range according to the historical data, the coordinators
will request a re-work or correction in most cases. This ensures the quality of the final
proposal at an immediate cost of the efficiency of proposal development, due to additional
communication and routine activities. Although contradicting previous findings, such as
the work done by Thomson et al. (2005), this conclusion makes reasonable sense in this
case study. Another finding reveals that goal congruence hardly affects the work pressure
of the employees. This indicates that work pressure is mainly determined by task
characteristics, such as the work amount and dependence. Therefore, improved goal

202

congruence is not very helpful in improving work quality or balancing employees’ work
load.

Second, in certain cases, micro-management can provide several benefits to a project team,
including increased efficiency and effectiveness, improved work quality, and reduced
work pressure of team members; but this is true only when certain behavioral and
technical conditions are met.

The simulation results find that micro-management

improves the efficiency of the project team only when goal congruence between the
proposal team and the engineering team is at a relatively low level. In other words, when
the two teams obviously diverge in targets, micro-management may help improve
productivity. However, when goals between teams gradually become congruent, the
benefit obtained from micro-management starts to decrease. When goals are highly
congruent (e.g., greater than 0.8), micro-management indeed reduces the efficiency and
effectiveness of teamwork. On the other hand, micro-management has relatively
consistent effects on other performance indicators, including work quality and work
pressure of team members. It was found that micro-management can improve the quality
of project teams in all cases, except when goals are highly congruent. This indicates that
more frequent involvement from a higher level of management can help quality control.
The other finding is that micro-management balances the workload between employees
and thus reduces the work pressure. These findings indicate that the effects of micromanagement are very contingent on the behavioral and technical conditions of a project

203

team, suggesting that the adoption of micro-management should be built on a prudent
evaluation of the environment.

Third, task dependence can significantly affect the productivity and work quality of the
project team; it is able to alter the effects of goal congruence and micro-management.
Task dependence is a crucial factor for understanding inter-team cooperation in proposal
development. On the one hand, the simulation results find task dependence to be a
significant predictor of efficiency, effectiveness, and quality. If tasks are more dependent,
the team is less productive and commits more mistakes. This is understandable from an
empirical perspective, since dependence often means additional efforts for communication
and coordination, a bigger chance of mistakes and conflicts. On the other hand, task
dependence may affect the effects of goal congruence and micro-management. Simulation
results found that the efficiency difference between levels of goal congruence becomes
bigger when tasks are more dependent. In contrast, the effects of micro-management are
more significant when tasks are more independent. This finding highlights task
dependence to be a vital point of decision making in project team management, especially
when managerial and/or behavioral changes are planned.

6.3. Discussion

6.3.1. General findings of simulation studies
The simulation experiments were conducted based on two case studies, as described in
previous sections of this chapter. These experiments investigated realistic issues
204

pertaining to interpersonal cooperation in construction project teams. In particular, three
categories of topics were investigated in these simulation experiments, namely technical,
behavioral and managerial:

•

Technical: focused on the technical components of interpersonal cooperation and
their influences on project team performance. Typical topics include the
implications of task characteristics, such as task dependence.

•

Behavioral: investigated the attitudinal preferences and behavioral responses of
project team members to different situations, and consequential influence on team
performance. Typical topics include trust among team members, goal congruence,
and influence of sense of time pressure.

•

Managerial: investigated the effects of managerial actions/strategies under the
general framework of cooperation in construction project teams. Typical topics
include influence of different job acquiring strategies, coordination and
communication activities, and managerial preferences, such as micro-management.

The simulation results highlighted the role of interpersonal cooperation in project team
performance, either with the inter- or intra-team scenario. In addition, it was found that
four constructs of interpersonal cooperation--tasks, individuals, behaviors, and
organizational context--all play unique roles in forming cooperation and in turn, they
affect efficiency, effectiveness, quality and other performance indicators. The findings of
simulation experiments are used to form operational recommendations as follows.

205

6.3.2. Operational recommendations
Based on the simulation findings, a set of operational recommendations are made as
follows. It is worth noting that these recommendations are based on very specific case
studies, and therefore, should be considered when similar scenarios are met.

First, it should be noted that any managerial decisions made by the team decision
makers will trigger a series of behavioral reactions of the team members, which
might lead to unexpected consequences. Behavioral reactions could be serious enough
to alter the effects of original plans. Sometimes, the effects of certain managerial actions
are even counterintuitive. A lesson learned from case study 1 is “balancing,” which plays
a vital role in improving the performance of the team in the project planning phase.
Specifically: 1) decision makers should balance the total work amount to a level
commensurate with the team’s capacity. Otherwise, in some cases one additional job
could risk the completion of all other jobs; 2) managers should strive to balance the work
load between team members. Otherwise, unexpected behavioral responses might occur,
and additional time will be taken for unnecessary coordination; and 3) the function of
meetings in enhancing cooperation between team members should be recognized, but the
frequency of meetings should strike a balance between the benefit gained and cost of time.

Second, improving goal congruence between functional units and applying micromanagement should be considered simultaneously in order to improve team
performance. On the one hand, encouraging team members to share the same goal is

206

always helpful for team productivity. A higher level of goal congruence can significantly
improve efficiency and effectiveness in proposal development. This directly results in
shorter work duration and less cost. But goal congruence helps the quality of proposal
development very little. On the other hand, the simulation revealed that micromanagement can always help reduce mistakes in the developed proposal and therefore,
improve quality. However, it reduces efficiency and effectiveness of the team, especially
when goal congruence is improved. In short, improving goal congruence and applying
micro-management might offset consequences. As a result, the decision maker must make
a strategic decision between the two approaches, according to the specific environment.
For example, if the job is highly time driven, a feasible strategy to improve team
performance could be to improve the mutual understanding between different functional
units on the objectives through more face-to-face communication. But if the job is quality
driven, a better strategy might be more involvement of management in routine jobs,
adding more quality monitoring procedures.

Third, decision makers should notice the importance of the technical characteristics
of the jobs, especially task dependence, in the effects of any decision on team
performance. Evidence from the simulation experiments has proven task dependence to
be an amplifier of the effects of goal congruence but at the same time, it mitigates the
influence of micro-management. This finding has a practical value for decision makers. In
the case when tasks are more dependent, the decision makers should make a greater effort
to improve goal congruence rather than use micro-management; but when tasks are
207

relatively independent, micro-management is a better choice for improving team
performance. In this way, the benefits gained from improved goal congruence or micromanagement can be utilized to the maximum.

The results demonstrate that VOICE is an effective decision support tool for project team
management, not simply a theoretical investigation tool. It converts realistic scenarios to
executable models, where inputs are drawn from empirical evidence. Furthermore,
because ABM is utilized, and a GUI is used in VOICE, practitioners will have little
difficulty applying VOICE to real world management practices.

6.4. Summary
In order to illustrate how VOICE can be used to solve realistic cooperation issues in
construction project teams, two case studies were introduced in this chapter. Inter-team
and intra-team cooperation problems of the two cases were framed using the proposed
framework. A set of what-if scenario simulations and exploratory simulations was
conducted to quantify the relationship between cooperative behaviors or processes and
project team performance. Operational recommendations were offered based on the
simulation results.

208

CHAPTER 7: CONCLUSIONS
This chapter reviews the entire work and provides a constructive and precise evaluation of
this research. It identifies the contributions of this research, and addresses the limitations
of this research and future agenda to fill the gaps.

7.1. Review of the research
This research stemmed from the recognition of the importance of enhancing cooperation
in the construction industry. Accumulating evidence attributes performance issues of the
construction industry to the lack of cooperation between project participants, which is
described as a reciprocal effort of two or more individuals or organizations acting toward
the accomplishment of the common interest. An in-depth review of this research found
that existing industrial and academic efforts should be extended for fully understanding
cooperation in construction settings. The theoretical gaps can be grouped as:

•

A lack of effort to investigate cooperation at individual levels;

•

A lack of attention to the behavioral and attitudinal aspects of cooperation in a
construction setting;

•

A lack of solid evidence for the relationship between cooperation and construction
project team performance; and

•

A lack of an effective research method to capture the complexity of the studied
theme.

In response to the importance of this theme, this research investigated cooperation in
construction project teams from an innovative perspective that has been ignored by most
209

efforts. In particular, it investigated cooperation at the individual level – this research
views interpersonal cooperation as the most fundamental cooperation process that
determines the quality of a construction project team. As a result, the core of this research
is to capture the transition from local dynamics of micro-level cooperative behaviors to
global performance of project teams. Unlike most of previous studies in the same topic, it
emphasizes the diversity of cooperative behaviors of project team members and the
realization of them in different scenarios.

To achieve the research goal, a comprehensive conceptual framework was developed
based on an extensive review of literature pertaining to cooperation theory, construction
project teams, and exploratory simulation. The proposed framework captures all relevant
components of cooperation in a construction setting, including technical factors,
individual traits and behaviors, and organizational context. Team performance, e.g.,
productivity and quality, is assumed to be the result of the interactions among these
components of cooperation. On the basis of the proposed conceptual framework, an ABM
simulation platform, named VOICE (Virtual Organizational Imitation for Construction
Enterprises), was developed for a rigorous theory-based investigation of the micro-level
behavioral and attitudinal factors that shape local interpersonal cooperation processes in
construction project teams, and their impacts on team performance.

VOICE was used to conduct two in-depth case studies. In the first case study, intra-team
cooperation between members of an estimating team in typical DBB projects was

210

explored. The big question was “How do managerial actions affect cooperative behaviors
in a single team and in turn determine team performance in typical Design Bid Build
(DBB) projects?” Specifically, the implications of certain managerial actions - job
acquiring strategies, coordination activities, and team communication - were investigated
under different scenarios. The second case study extended the investigation to a crossfunctional context where institutional and psychosocial conditions become more complex.
The simulation experiments were performed based on a real scenario between a proposal
team and an engineering team in an EPC construction enterprise. The question
investigated was “What is the role of goal congruence in the cooperation between
members of different teams during project proposal development?” The effects of goal
congruence were examined, based on a series of uncertainty analyses. Moreover, the
implications of time pressure, task dependence and micro-management were investigated
to determine if additional influences were exerted on inter-team cooperation.

The

findings of the two case studies and corresponding analyses have validated the basic
assumption of this research, that cooperation at an individual level plays a vital role in
determining the performance of a construction project team in certain cases. This
investigation could be nontrivial because there are numerous complex interactions to be
captured between components of cooperation. These interactions constitute the transition
from local dynamics to global efficacy of a construction project team.

This research does not aim to tune down the importance of previous research efforts.
Instead, it attempts to add knowledge to construction studies, from an innovative
211

perspective, associated with the nature and implications of interpersonal cooperation in
construction project teams. It still falls in the research category of enhancing cooperation
in construction settings.

7.2. Contributions of this research
This research has made a variety of contributions to the body of knowledge with respect
to the cooperation in construction settings. In particular, it builds a stepping stone for the
knowledge transformation of cooperation research in construction by providing an
innovative investigation approach, and offers a decision support tool for construction
companies. The specific contributions are summarized as follows.

First, this research provides an alternative method for future cooperation studies in
construction, i.e., a simulation model to investigate cooperative behaviors at the
individual level and micro-level processes in the construction context. The importance
of cooperation in forming high performance construction project teams has been well
recognized. But there is a lack of comprehensive framework to capture all relevant
elements of cooperation at the individual level in construction project teams. Furthermore,
it remains a problem that how to realize a continuous knowledge discovery with an
effective investigation method other than surveys and interviews. This research developed
a simulation model to provide a more comprehensive view for future cooperation studies
in construction. Specifically, the model captures the most fundamental and specific
processes that constitute interpersonal cooperation in construction project teams,

212

including technical factors (task characteristics), behaviors (cooperative behaviors),
individual traits and organizational context (reporting structure, work process and
information flow). It views cooperation as a synthesized result of the dynamic interaction
among these factors. For instance, various simulation experiments conducted in this
research have proven cooperation to be a manifestation of the interactions among
individuals. Such a method provides innovative insight into the constructs and mechanics
of cooperation in construction, and allows the understating and investigation of the
dynamics of cooperation that we couldn’t access before. In a short, this research models
for insights, not numbers.

Second, the case studies contribute to the cooperation research in construction by
highlighting the role of behavioral and attitudinal aspects of cooperation in
construction settings when certain conditions are met. Industry has attempted to
understand and solve cooperation related issues from a technical point of view. Efforts
have included new contracting strategies and advanced technologies made to foster the
collaboration between project organizations. This research proposed an alternative way to
encourage cooperation in construction projects, and paid particular attention to the
behavioral and attitudinal processes of each project team member. The two case studies
found in certain cases project team members tend to adopt very different behavioral
actions towards changes in circumstance, which is affected by their psychosocial
conditions. These behavioral reactions to the contingencies will, in turn, determine the
quality and efficacy of project teamwork. Some experiments indicated that a slight change
213

to the behavioral preference of the actors, such as the micro-management preference of
managers/coordinators in project teams, can significantly affect the final productivity and
quality of the team work.

Third, this research proves ABM to be an effective investigation approach in the
studied theme, especially in an era when complexity dominates. The interactions
among components of interpersonal cooperation and project team performance is quite
complex, affected by multiple other institutional and environmental factors (such as task
dependence). There is no simple answer to the nature of cooperation in construction
settings as claimed by previous evidence; instead, the complex interactions of cooperative
behaviors and processes should be taken into account. Traditional “top-down” methods
have proven insufficient to the studied theme because rules of the entire system are
posited even before the investigation. In a temporary and virtual context, considerable
information pertaining to cooperative behaviors and processes is missing from the
analytical process. ABM, on the other hand, fits into the studied theme very well. It
situates an initial population of autonomous heterogeneous agents, i.e., project team
members in this case, defines decentralized behaviors, and allows them to interact
according to local rules; thereby, it generates (or “grows”) the macroscopic regularity
from the bottom up. This is a natural reflection of the cooperation processes in
construction project teams. Results indicate that ABM can effectively capture the
complexity embedded in cooperation processes of project teams, and is able to reproduce
the transition in a bottom-up manner. Plus, attributed to the flexibility of ABM, the model
214

architecture can be easily tailored to different scenarios. Uncertainty analysis is also a
simple function in ABM. The findings of this research add additional credibility to the
application of ABM in the construction engineering and management field.

Last, this research provides construction companies with an innovative decision
support tool, VOICE. It is very easy for decision makers in any construction
organizations to measure the performance of a construction project. Also, it is not a
nontrivial job to evaluate the micro-level team processes in construction settings such as
processing project tasks. However, the link between local processes and global project
team performance is contingent on many contextual factors. It is therefore difficult for
most decision makers to estimate the likely performance outcome of a team, given
complex initial conditions and relationships among these conditions. VOICE is developed
in this research to capture the transition from local dynamics of cooperation processes and
behaviors to global performance of construction project teams. The two case studies
demonstrated that VOICE can be used to examine the influence of local processes, such as
micro-management, specific coordination, trust among team members, communication
and conflicts, on a project team’s performance, including efficiency, effectiveness and
quality.

215

7.3. Potential limitations of this research
Several weaknesses of this research might limit the reasonableness of the findings and the
applicability of the conceptual framework. These weaknesses mainly come from the
limitation of the research methodology.

First, the conceptual framework proposed in this research still needs external
justification to increase its reasonableness and applicability. Interpersonal cooperation
in construction settings is a multidisciplinary theme building on the knowledge of
multiple disciplines, including psychology, organizational science, construction
engineering and management. The conceptual framework proposed by this research is
mainly based on a literature review solely conducted by the author. Due to limitations of
the author’s expertise, it is likely that many important aspects/variables are missing from
the framework. Furthermore, although the model has been validated against empirical
evidence, the possibility still exists that certain assumptions used in the simulation model
come from author’s inappropriate interpretation of the literature and evidence. Taking the
list of cooperative behaviors in VOICE as an example, the author defines and models
these behaviors according to a straightforward interpretation of the literature and evidence
from the case studies. The rules, at first glance, might make fair sense for the simulation
experiments; but there is no guarantee that all necessary psychosocial processes are
captured well. Behaviors are investigated separately in VOICE, but they might be
interdependent in reality – coordination frequency of a coordinator might be the result of
trust between him/her and other team members; micro-management preference might
216

depend on the manager’s competence. Overlooking these interlinks among cooperative
behaviors could distort the analysis results. Simplification is a natural requirement of a
modeling study, but such simplification should ensure that the fundamental structure of
the real system is preserved. This research needs to do more work to justify this.

Second, more empirical evidence is required to improve the generalizability of the
findings. As a small scale research, the empirical evidence used in the simulation model
development and investigation is from only two case studies. Moreover, both case studies
are only about cooperation in the project planning phase (one discusses the estimating
effort, the other discusses proposal development). This is mainly due to the limited
resource that the author could access. Indeed, a construction project is a multi-phase effort,
from project planning to project closeout. The cooperation in any project team could
occur in home office and project sites. To fully understand the nature and implications of
cooperation in construction settings, a representative project phase should be considered.
In addition to “phase” consideration, the impacts of project type, contracting strategies,
delivery systems and other project and organizational characteristics pertaining to a
construction project are not ignorable. This research simplifies the external environment
and focuses on a narrow scope, which limits the generalizability of the findings.

Third, the selected cooperation related issues/problems investigated in the simulation
experiments hardly satisfy the full need for reality-based data and thus, hurts the
model’s usefulness. Due to the need for accessible resources and length limitation of this

217

dissertation, only selected cooperation related issues/problems have been investigated in
this research. Although these issues/problems are directly identified by practitioners, in
that they (issues/problems) are critical to the success of their teams, cooperation related
issues/problems in reality could be much richer and more diverse. Critical points of
decision could happen in all links of cooperation processes, including organizational,
procedural, behavioral, and cultural implications. In fact, numerous studies have already
proven the importance of cooperation components in different circumstances. Therefore,
future work should incorporate more cooperation related issues/problems in construction
settings to improve the usefulness of the developed model.

7.4. Future agenda
A future agenda is proposed to address the limitations as discussed previously:

•

Refine the current conceptual framework for interpersonal cooperation in
construction setting to capture a richer pool of behaviors and reflect
interdependent psychosocial processes. Talents from all relevant disciplines –
psychology, sociology, organizational science and other engineering domains –
will be introduced for future research efforts. The participation of subject matter
experts would help refine the current conceptual framework to reflect realistic
interdependent psychosocial processes to a better degree. In addition, the current
list of cooperation behaviors will be extended, based on new evidence, to develop
a bigger pool of behaviors for the simulation model. The ultimate goal is to

218

develop a “library of cooperative behaviors” specified for construction
organizations, with detailed representation framework and models. By smart
selection and combination of these behaviors, future use of VOICE might be to
model the comparable complexity in reality.
•

Extending the investigation to consider complete project phases and scenarios.
More empirical evidence will be collected and integrated into the framework in the
hope that the cooperation processes in an entire project life cycle could be
captured by the investigation. Moreover, representative scenarios, such as
cooperation issues in different project delivery systems, are also expected in future
investigations. In such, VOICE can be enhanced to be a generic investigation tool
pertaining to cooperation in construction settings. Notably, the basic conceptual
framework might need to be tailored to reflect the differences between project
phases and scenarios.

•

More advanced analysis methods will be used to enhance the analytical ability
of VOICE. At present, VOICE can support uncertainty analysis. This is helpful to
examine the variation of simulation outputs, and uncover the relationship between
determinants and observations. However, uncertainty analysis helps little to
compare between critical determinants of problems in terms of their contribution
to uncertainty, especially when the number of variables considered in the study is
quite large. Global sensitivity analysis (GSA) is a better option for the analytical
function of VOICE. As an advanced uncertainty analysis method, GSA has
219

received increasing interest of scholars from different disciplines, especially in
environmental science. It decomposes the unconditional variance of the model into
terms due to individual factors, plus terms due to interaction among factors (Chen
et al. 2005), and therefore, it performs better in revealing the structure of a
problem. The challenge of introducing GSA into VOICE is the development of the
interfaces across platforms.

This dissertation builds a stepping stone for knowledge transformation about cooperation
in the construction context. It is expected that better approaches to improve the
performance of construction projects and project teams can be discovered based on the
contributions of this dissertation.

220

APPENDICES

221

APPENDIX 1: INFORMATION ABOUT CASE STUDY 1
A1.1 Data collection instruments
Surveys
(A sample of three survey templates; N=8)
To whom it concerns,
Thank you for considering participates in this research study!
Contemporary construction project teams are mostly labeled by adversarial relationships,
a lack of transparency, and mistrust. This affects team’s capability to deliver projects to
the client in a reliable and predictable manner. As addressed by many industry wide
reports, a real challenge of the construction business is how to encourage project
participants to work together effectively. A major source of the ineffectiveness of
construction project teams is the lack of interpersonal cooperation between project
participants, which is described as a reciprocal effort of two or more individuals act
towards the accomplishment of the common interest. Scientific and industrial evidence
has proven that interpersonal cooperation across the companies and within the
construction company both influence the performance of a project team. Interpersonal
cooperation, on the other hand, is a complex consequence of many interacted factors,
including personality, work settings, organizational hierarchy, and daily work-related
activities etc. To fully understand the interpersonal cooperation in construction project
teams, it demands a closer investigation to these relevant factors. This study aims to
investigate and reproduce how interpersonal cooperation is built in construction project
teams, and what is its influence on team performance using computer simulation.
We appreciate and highly value your cooperation in this research. The investigation
results will only be used for the research and education purpose in School of Planning,
Design and Construction (SPDC) at MSU; all the information will be classified and kept
confidential subjected to Federal and MSU regulations (for more detail, please refer to
“Informed Consent Form”).If you have any concerns or questions about this study, such
as scientific issues, how to do any part of it or to report an injury, please contact the
researcher.
By clicking “accept informed consent form”, you will be transferred to the survey. Thank
you for your cooperation.

PERSONAL PROFILE

222

How many work experience do you have? (

)

What department are you in?
Board of Management
Project Planning (Estimation, Scheduling etc.)
Project Control
Project Operation (e.g., Project Manager)
Finance/Accounting
Human Resource
Other:

What level of position are you currently in?
President or equivalent
Vice president or equivalent
Department head
Senior
Junior
Other:

WORK RELATED BEHAVIORS

When doing your job, do you follow the following procedures?
223

Never

Seldom

Half-half

Seeking for
project
information
Analyzing the
potential
project
Set priority
among
different tasks
Check the
quality of the
document/task
submitted by
subordinates
Approve tasks
of
subordinates
Assign tasks
to
subordinates
Check the
availability of
necessary
information
Check
whether there
are too many
exceptions in
the entire
company
(e.g., too
many
complaints)
Initiate
meetings
Make record
for future
reference
224

Most time Always

WORK RELATED PREFERENCE

If self-assess your work performance quality, what is your assessment?
Excellent
Good
Average
Not so good
Other:

Mistakes in submitted job are inevitable. If measured as a percentage, how much
percentage is unacceptable for you?
1%~3%
3%~5%
5%~10%
10%~20%
Other:

Continued: If you found too many mistakes existing in the work submitted by your
subordinates, what would you most likely do?
Return it to whom prepared it
Re-assign it to another subordinate, e.g., who has better performance quality
Correct the mistakes by self

225

Initiate an emergent meeting for it
Depends on how many mistakes
Other:

When necessary information is missing for your current work (e.g., lack of quotes
from subcontractors), what would you most likely do?
Query information source
Ask subordinates to find necessary information
Make judgment based on experience
Looking into common source, e.g., means book
It depends
Other:

Many management exceptions happen every day, e.g., too many complaints about
overloaded work, lack of subcontractors quotes, or too many work performance
quality issues. If we simply count these management exceptions, what is the number
of exceptions is too big for you (daily)?
1~5
5~10
10~15
15~20
20~25
25~30
>30
226

Other:

Continued: If too many management exceptions happened, what would you most
likely do?
Initiate an emergent meeting
Talk to responsible person
It depends on the seriousness
Do nothing
Other:

Continued: If you decide to initiate an emergent meeting towards management
exceptions, how long of such typical meeting (not routine meeting)?
<1 hour
1~2 hours
2 hours ~half day
half day - 1 day
> 1 day
It depends on the seriousness
Other:

227

Task information

No.

Task

Processing/lead time
(hour)
Min

Likely

Max

Priority (0~10)
(10: most urgent)

Complexity (0~10)
(10: most difficult)

YOUR COMMENTS

Please write any comments in the space provided below regarding this survey. Any
feedback will be greatly appreciated
(

)

Interviews
(Only present key questions to two principals; N=2)
Q1: The major issues in your company? Are they cooperation related?
Q2: Can you describe the responsibilities of your employees?
Q3: Can you describe representative activities of principals/managers/staff on a daily
basis?
Q4: How many projects can you estimate at the same time based on current resources?
Q5: Do you agree that individual competence/experience/decision process can affect the
work efficiency and effectiveness? Linear relationship?
Q6: To accelerate the estimating, what would you or mangers do?
Q7: How many management exceptions can you bear? (e.g., too much unnecessary
communication) What will you do to solve management exceptions? (Meeting?)
Q8: Frequency and normal duration of your meeting? Do you agree that meetings can
enhance the communication, reduce unnecessary actions and misunderstandings?
Q9: Can you verify/validate the model inputs as presented?
Q10: Can you verify/validate the list of work related behaviors as presented?
Q11: Can you verify/validate the simulation findings as presented?
228

Company Documents
(Only present the list of documents due to confidentiality)
D1: Organizational chart
D2: Estimation flowchart
D3: Time sheets (Jan 2010 through August 2010)
D4: Job description
D5: Other relevant management artifacts

A2.2 Model parameters
Projects and Tasks
Based on the data instruiments, necessary model inputs were collected. The folowing
tables show the parameters of three projects used in simulation.
Table A.1 Task parameters of Project 1 (size=117)
Task
Task amount
Task difficulty
Task priority
Dependency
information
Authority level
# of successor
Actor address
Task
Task amount
Task difficulty
Task priority
Dependency
information
Authority level
# of successor
Actor address

1
5
1
1
1
0
1
1
0
11
5
1
1
1
0
1
1
1

2
5
1
1
1
0
1
1
1
12
5
1
1
1
1
3
1
1

3
10
1
1
1
1
1
1
0
13
10
1.2
1.5
1
0
3
1
1

4
15
1.1
1.1
1
1
2
1
0
14
15
1
1
1
0
1
1
3

229

5
15
1.1
1.1
1
0
2
2
0
15
20
0.5
1
1
0
1
1
3

6
10
1.2
1
1
1
2
1
1
16
15
1
2
1
1
1
1
2

7
20
1.2
1
1
1
1
2
2
17
10
1
1
1
1
1
1
0

8
20
1.5
1
1
0
1
1
1
18
10
1
2
1
0
1
1
2

9
25
1.2
1.2
1
0
1
1
1
19
5
1
2
1
0
1
1
2

10
10
1
1
1
0
1
1
1
20
5
1
1.5
1
0
2
0
2

Table A.2 Task parameters of Project 2 (size=115)
Task
Task amount
Task difficulty
Task priority
Dependency
information
Authority level
# of successor
Actor address
Task
Task amount
Task difficulty
Task priority
Dependency
information
Authority level
# of successor
Actor address

1
5
1
1
1
0
1
1
0
11
10
1
1
1
0
1
1
1

2
9
1
1
1
0
1
1
1
12
10
1
1
1
1
3
1
1

3
10
1
1
1
1
1
1
0
13
10
1.2
1
1
0
3
1
1

4
10
1
1
1
1
2
1
0
14
15
1
1
1
0
1
1
3

5
15
1
1
1
0
2
2
0
15
10
0.5
1
1
0
1
1
3

6
12
1
1
1
1
2
1
1
16
15
1
1
1
1
1
1
2

7
20
1.2
1.2
1
1
1
1
4
17
10
1
1
1
1
1
1
0

8
10
1.2
1
1
0
1
1
1
18
10
1
2
1
0
1
1
2

9
25
1.2
1.2
1
0
1
1
1
19
5
1
2
1
0
1
1
2

10
10
1
1
1
0
1
1
1
20
10
1
2
1
0
2
0
2

6
10
1.2
1
1
1
2
1
1
16
15
1
1
1
1
1
1
2

7
20
1.2
1
1
1
1
1
4
17
10
1
1
1
1
1
1
0

8
20
1.2
1.2
1
0
1
1
1
18
10
1.1
2
1
0
1
1
2

9
30
1.2
1.1
1
0
1
1
1
19
5
1
1.8
1
0
1
1
2

10
20
1
1
1
0
1
1
1
20
8
1.1
2
1
0
2
0
2

Table A.3 Task parameters of Project 3 (size=137)
Task
Task amount
Task difficulty
Task priority
Dependency
information
Authority level
# of successor
Actor address
Task
Task amount
Task difficulty
Task priority
Dependency
information
Authority level
# of successor
Actor address

1
15
1
1
1
0
1
1
0
11
5
1.1
1.1
1
0
1
1
1

2
15
1.2
1
1
0
1
1
1
12
5
1
1
1
1
3
1
1

3
10
1
1
1
1
1
1
0
13
10
1
1
1
0
3
1
1

4
15
1.1
1.2
1
1
2
1
0
14
15
1
1
1
0
1
1
3
230

5
15
1.1
1.1
1
0
2
2
0
15
20
0.8
1
1
0
1
1
3

Individual information
Individual information is shown in figure 6.2

A2.3 Model development
The following figure demonstrates the sketch of organizational chart of case study 1:
Principals

Manager 1

Manager 2

Staff 1

Manager 3

Staff 4

Staff 2

Staff 5

Staff 6

Staff 3

Figure A.1 Organizational chart of case study 1
Specific model codes can refer to appendix 3: sample codes.

A2.4 Outputs
Charts were generated and documented as shown in chapter 6.

231

APPENDIX 2: INFORMATION ABOUT CASE STUDY 2
A2.1 Data collection instruments
Surveys
(Survey template; N=34)
Introduction
Obtaining accurate and timely information from the engineering teams at the time of
proposal development has historically produced difficulties and process inefficiencies.
This survey aims to investigate the causes and possible remedies. Your opinion is
extremely important for the success of this survey. Please take 15 minutes to answer the
following questions (most are multi-choice questions). Your confidentiality and privacy
will be the first priority, and your identification won’t be stored or appear in any
document.

A. Background information
Would you please provide background information about yourself?
How many years of experience do you have as an estimator? (
How many years of experience at Z? (

)

)

What is your current position? (

)

Number of years in current position? (

)

What craft is your major area of focus? (

)

B. Problem identification
Communication with engineering teams at the time of proposal development has
historically produced difficulties and process inefficiencies. Could you please provide us
your opinions related to these concerns?
232

Overall, to what extent do you feel that your performance is affected negatively because
of communication issues with the engineering team? Please circle one.
No Affect (-1-2-3-4-5-) Significant Affect
Overall, to what extent do you feel that your contact person from the engineering team is
meeting your expectations? Please circle one.
Not meeting any of my expectations (-1-2-3-4-5-) Meeting all of my expectations
Overall, to what extent do you feel your engineering team counterpart is approachable
regarding supporting you in performing your assigned tasks?
Not at all Approachable (-1-2-3-4-5-) Very Approachable
Overall, in your opinion, to what extent does the engineering team meet their
responsibilities in supporting the estimating process? Please circle one.
Not meeting their responsibilities (-1-2-3-4-5-) Meeting all of their responsibilities
I feel the following problems have affected my work efficiency:
Late reply from engineering team
Inaccurate information from engineering team
Different technical terms, labels and knowledge
Significant incomplete information from engineering team
Others:__________________________________________________
I think the following influence the difficulties and inefficiencies with the engineering
team at the time of estimate development:
Different focuses
Different job responsibilities
Different award/evaluation systems
Different work objectives
Different location

233

Different organizational culture
Two separate entities at Z
Work procedure issues
Lack of effective communication methods/technologies
Lack of rules or standard procedures to solve problems
Lack of personal social contact
Lack of coordinating efforts from upper levels
Others:__________________________________________________
How would you change the current work process to better support your estimating
task?____________________________________________________________________
What other issues do you feel are important regarding communication issues with the
engineering team? _______________________________________________
C. Task-related questions
Research suggests the features of a person’s work can affect commnucation. Could you
please provide us your opinions about your work?

To what extent is your task performance dependent on receiving accurate information
from engineering team? Please circle one.
Not at all Dependent (-1-2-3-4-5-) Very Dependent
To what extent does your task performance impact others in your team?
Very Little Impact (-1-2-3-4-5-) Significant Impact
To what extent do you agree that working at the same location with the engineering team
could improve communication? Please circle one.
No improvement (-1-2-3-4-5-) Significant Improvement
To what extent do you work closely with others in your team to perform your tasks?

234

Work independently (-1-2-3-4-5-) Work closely with others
I typically understand my work objectives and the process to perform my tasks
Do Not Understand (-1-2-3-4-5-) Understand
Please describe typical tasks you perform on a daily basis (including a brief description,
normal duration and if this task is dependent on engineering team).
Task Name

Task Description

Task
Duration

Dependent on
engineering team?

D. Personal characteristics-related questions
Personal characteristics play an important role in interpersonal relationships. Would you
please share your opinion of your personal characteristics?
To what extant you agree with the following statements. Unless noted otherwise use: (1:
Strongly DISAGREE; 5: strongly AGREE):
I frequently do extra things I know I won't be rewarded for, but which make my
cooperative efforts with other people more productive. (-1-2-3-4-5-)
I willingly help other individuals, even at some cost to my personal productivity. (-1-2-34-5-)
I prefer to work with others in a group rather than working alone. (-1-2-3-4-5-)
When making decisions at work that affect others, I try to take their needs and feelings
into account. (-1-2-3-4-5-)
I enjoy discussing my current job with people outside work. (-1-2-3-4-5-)
I rate my job satisfaction as (-1-2-3-4-5-)
I assist others with difficult/heavy workloads, even though it is not part of my job. (-1-23-4-5-)
235

My contact person from the engineering team is typically qualified with sufficient
experience and depth of technical knowledge. (-1-2-3-4-5-)
My contact person presents him / her self as an important resource to the team. (-1-2-3-45-)
From my experience I rate my engineering counterpart’s commitment to the process as
Unacceptable (-1-2-3-4-5-) Acceptable
My engineering counterpart always tries to take my needs and time deadline requirements
into account. (-1-2-3-4-5-)
In your opinion, does your engineering counterpart take ownership of the quantity
estimate provided to you for inclusion in an estimate? Not at all (-1-2-3-4-5-) Yes

E. Interaction related questions
The following questions are about how you communicate with the engineering team, and
your personal response to certain conditions when working on major estimates.
In my work I use the following ways to communicate with engineering team:
Phone call
Email
Fax
Face-to-Face meeting
Mail
Others:__________________________________________________
The most important ONE communication way:
Phone call
Email
Fax
Face-to-Face meeting
236

Mail
Others:__________________________________________________
The frequency I officially communicate with engineering team is ___ times / week when
we are developing a proposal.
Overall we (me and person from engineering team) spend ____minutes in each
conversation.
The frequency I socially and informally communicate with engineering team is ___times /
week.
Overall, on average, I typically wait ____hours or _____days to get response from
engineering team.
Overall, it typically takes _____ iterations (times “back-and-forth”) until I get the final
information I need.
How often has your engineering counterpart notified you when he/she could not meet a
commitment? Never (-1-2-3-4-5-) Always
If you cannot get needed information, which of these actions do you typically take? Please
mark (1, 2, 3):
Contact the same person again
Seek help from another person in engineering team
Seek help from someone in my team
Contact the supervisor of engineering team
Try to figure out the needed information by myself
Schedule a face-to-face meeting
Others:__________________________________________________
Once you get the needed information, what would you do:
Use the information directly
Carefully check the information before using it
237

Others:__________________________________________________
Measured in a percentage (100% is the best), what percentage of quality of the
information is acceptable in your work? _______(e.g., 90%)
If you believe lack of qualified tasks of others has just affected your work, what action do
you typically resort to first:
Raise issues to supervisor
Try to talk to responsible person
Leave it alone
Others:__________________________________________________
How often do you have official meetings with engineering team? ____ (e.g., 1/month)
Regarding improved communications, can you briefly describe the positive results you
have experienced from face to face meetings with your engineering counterpart?
_______________________________________________________

F. Organization and procedure related questions
Current procedures lead to decisions and outcomes favorable to me.
Never (-1-2-3-4-5-) Always
Members of the estimation team are encouraged by supervisors to take proactive action to
resolve engineering communication issues?
Never (-1-2-3-4-5-) Always
Help is available from the organization (estimation and/or other teams) when I have a
problem.
Strongly Disagree (-1-2-3-4-5-) Strongly Agree
Rather than being totally dependent on my engineering counterpart to come through when
I need assistance, I try to have a backup plan ready.
Strongly Disagree (-1-2-3-4-5-) Strongly Agree
238

I keep close track of my interactions with my engineering counterpart, taking note of
instances where he/she does not keep up her/his end of the bargain.
Never (-1-2-3-4-5-) Always
F. Open Question
What else do you want to say regarding the communication difficulties with the
engineering team, such as causes, solutions, or your feelings? (

)

Interviews
(Only present key questions to five key decision makers; N=5)
Phase 1:
Q-1: What are the major issues in your team? Are they cooperation related?
Q-2: Amongst the four identified cooperation related issues, which one is considered as
the most critical one?
Q-3: Why is the communication/information exchange issue between proposal team and
engineering team is important to you?
Phase 2:
Q-1: In general, how long it takes to develop a proposal?
Q-2: Do you and/or your colleagues have to work overtime to get your work done? Do
you feel pressure or frustration?
Q-3: What are the major responsibilities of your work?
Q-4: Can you verify/validate the list of work related behaviors as described?
Q-5: To what degree you and/or your colleagues’ job are dependent on the engineering
team?
Q-6: Do you think the performance of proposal team is affected negatively by engineering
team?
Q-7: Can you describe problems/difficulties with engineering team?
Q-8: What are the reasons for the problems you just described?
239

Q-9: How would you change current work process to better support your estimating task?

Company Documents
(Only present the list of documents due to confidentiality)
D-1: Organizational chart
D-2: Proposal development flowchart
D-3: Other relevant management artifacts (e.g.., excel pivot table used for communication
purpose)

Work Sampling

In order to collect actual data about work effectiveness, two approaches were used: 1)
survey: in a survey 34 respondents were asked to provide information of their
nonproductive activities, i.e., activities not associated with direct processing of tasks, such
communication (e.g. phone calls, emails and other communications that are not face-toface) and waiting time. Frequencies and durations of these activities were finally obtained;
and 2) work sampling: the author conducted a work sampling to one of the case study
companies for a month on a consecutive basis. Every day, 8 am through 5 pm, the author
documented number of people talking face-to-face or in a meeting at every 15 minutes.
The results were integrated finally to study the face-to-face communication behaviors of
the team members (Figure A.2). It was found in average, team members spent around 20%
of their time in “talking”, which is defined as nonproductive in VOICE.
240

16:50

16:30

16:00

15:30

15:00

14:30

14:00

13:30

13:00

12:30

12:00

11:30

11:00

10:30

10:00

9:30

9:00

8:30

8:00

180
160
140
120
100
80
60
40
20
0

Figure A.2 Work sampling result of the face-to-face communication in a month
A3.2 Model parameters
Based on data instruments, the model parameters used in case study 2 were collected.
Since Monte Carlo simulation was used in case study 2, PDFs were fitted against the
collected data. These PDFs were used as the inputs of the simulations. The following
figures demonstrate some of the fitted PDFs.

0

40

2

6
0

Task duration

Task dependence
Figure A.3 Fitted PDFs for input data

241

20
Years of experience

40

Figure A.3 Cont’d

40%
20%
0%
1 2 3 4 5 6 7
0

2

4

6

8

10 12 14
0.6 0.8

Exception handling time
100%
50%
0%
1

2

1.0 1.2

Quality threshold
50%
40%
30%
20%
10%
0%

1-contact responsible person;
2-seek help from others in
counterpart team; 3-seek help
from my team; 4-contact
supervisor for coordination;
5-try to figure out a solution
myself; 6-schdule meeting; 7others
Reactions to incomplete info

40% 40%

10% 10%

1-accept
the
info/task
directly; 2-check quality
before acceptance

Quality monitoring
preference

Reactions to unquality task

A3.3 Model development
The following figure shows the organizational chart of the studied company. Detailed
team structure is not shown in this document.

242

Z

ZENG

Z HQ

Enterprise Operation
Developme- section
nt

Support
Section
Disciplines
Project
planning
Information

Execution
Planning

Project
Controls

Estimating

Disciplines

Figure A.4 Organizational Chart of company Z
The work flow of proposal development has also been summarized (Not shown). In
simulations, this flowchart was converted to the possibility of a task to be assigned to a
particular actor.

A3.4 Outputs
3300 simulations were conducted to examine the influence of task dependence and goal
congruence on the performance of the project team. The following figure demonstrates the
results.

243

0.1

GC

Efficiency
5
4
3
2
1
0

0.2

0.5

1

5
4
3
2
1
0

15%
10%
5%
0%
0

0.5

1

100%
80%
60%
40%
20%
0%
0

0.3

Quality

100%
80%
60%
40%
20%
0%
0

0.5

1

5
4
3
2
1
0
0.5

5
4
3
2
1
0
0.5

1

1

0

0.5

1

0

0.5

1

0

0.5

1

5%
0%
0.5

1
15%
10%
5%
0%

0

0.5

1

100%
80%
60%
40%
20%
0%
0

0.5

10%

0

1

0
15%

100%
80%
60%
40%
20%
0%
0

0.4

Effectiveness

15%
10%
5%
0%
0

0.5

1

Figure A.5 Influences of task dependence and goal congruence on performance of
proposal development
244

0.5

Figure A.5 Cont’d

5
4
3
2
1
0

100%
80%
60%
40%
20%
0%

0.6

0

0.7

1

5
4
3
2
1
0

0

0.5

1

100%
80%
60%
40%
20%
0%
0

0.5

1

5
4
3
2
1
0
0.5

5
4
3
2
1
0
0.5

1

1

0

0.5

1

0

0.5

1

0

0.5

1

5%
0%
0.5

1
15%
10%
5%
0%

0

0.5

1

100%
80%
60%
40%
20%
0%
0

0.5

10%

0

1

0
15%

100%
80%
60%
40%
20%
0%
0

0.8

0.5

100%
80%
60%
40%
20%
0%

15%
10%
5%
0%
0

245

0.5

1

0.9

Figure A.5 Cont’d
5
4
3
2
1
0

100%
80%
60%
40%
20%
0%

1.0

0

1

5
4
3
2
1
0

10%
5%
0%
0

0.5

1

100%
80%
60%
40%
20%
0%
0

All

0.5

15%

0.5

1

5
4
3
2
1
0
0.5

1

0.5

1

0

0.5

1

0

0.5

1

15%
10%
5%
0%
0

0.5

1

100%
80%
60%
40%
20%
0%
0

0

15%
10%
5%
0%
0

0.5

1

The following figures shows the change of efficiency when task dependence ranges from
0 to 1. Result shows task dependence amplifies the effects of goal congruence (F ratio is
increasing).

246

5
4
3
2
1
0

5
4
3
2
1
0
0

0.5

1

0

Task dependence=0.0
F ratio=34.49
5
4
3
2
1
0

5
4
3
2
1
0
0.5

1

Task dependence=0.1
F ratio=87.32
5
4
3
2
1
0

0

0.5

1

0

0.5

0.5

0

0.5

0.5

1

Task dependence=0.9
F ratio=115.73

1

Task dependence=0.5
F ratio=85.66

1

0

Task dependence=0.7
F ratio=141.32
5
4
3
2
1
0

0

0.5

5
4
3
2
1
0

Task dependence=0.6
F ratio=129.28
5
4
3
2
1
0

0

Task dependence=0.4
F ratio=120.54

1

1

Task dependence=0.2
F ratio=140.80

1

5
4
3
2
1
0
0

0.5

5
4
3
2
1
0

Task dependence=0.3
F ratio=129.33
5
4
3
2
1
0

0

0.5

1

Task dependence=0.8
F ratio=153.90
5
4
3
2
1
0

0

0.5

1

Task dependence=1.0
F ratio=259.49

0

0.5
Combined

Figure A.6 Additional influences of task dependence on the effects of goal
247

1

The following figure shows the influence of micro-management and task dependence on
team performance based on 5280 simulations.
Quality

0.06
0.04
0.02
15
15

10

5

0.15
0.1
0.05

0

5

10

15

0

5

10

15

15

10

5

0

0.15
0.1
0.05

15

10

5

0
0

15

10

0

15

10

5

0

0

15

10

5
5

10

0.05

1
0.8
0.6
0.4
0.2
0
0

5
4
3
2
1
0

5

0.1

0

15

10

0.15

1
0.8
0.6
0.4
0.2
0
0

5
4
3
2
1
0

0

15

10

5

0

1
0.8
0.6
0.4
0.2
0
5

5
4
3
2
1
0

0.08

0

15

10

5

1
0.8
0.6
0.4
0.2
0

0

0.1
0.2
0.3

Effectiveness

5
4
3
2
1
0
0

0

Efficiency

Figure A.7 Influences of micro-management preference on team performance under
levels of task dependence
248

0.1
0.05

15
15

10

5

0.15
0.1
0.05

0

5

10

15

0

5

10

15

15

10

5

0

0.15
0.1
0.05

249

15

10

5

0
0

15

10

0

15

10

5

0

0

15

10

5
5

10

0.05

1
0.8
0.6
0.4
0.2
0
0

5
4
3
2
1
0

5

0.1

0

15

10

0.15

1
0.8
0.6
0.4
0.2
0
0

5
4
3
2
1
0

0

15

10

5

0
0

15

0.15

1
0.8
0.6
0.4
0.2
0
5

5
4
3
2
1
0
0

0.5
0.6
0.7

10

1
0.8
0.6
0.4
0.2
0
5

5
4
3
2
1
0
0

0.4

Figure A.7 Cont’d

0.1
0.05

15

10

0.05

0

5

10

15

0

5

10

15

15

10

5

0

0.15
0.1
0.05

15

10

5

0
0

15

10

5

5

0.1

0

15

10

0.15

1
0.8
0.6
0.4
0.2
0
0

5
4
3
2
1
0

0

15

10

5

0
0

15

0.15

1
0.8
0.6
0.4
0.2
0
5

5
4
3
2
1
0
0

0.9
1.0

10

1
0.8
0.6
0.4
0.2
0
5

5
4
3
2
1
0
0

0.8

Figure A.7 Cont’d

The following figure shows the influence of goal congruence and micro-management on
team performance based on 4800 simulations.

250

Quality

0.1
0.05

15
15

10

5

0.15
0.1
0.05

0

5

10

15

0

5

10

15

15

10

5

0

0.15
0.1
0.05

15

10

5

0
0

15

10

0

15

10

5

0

0

15

10

5
5

10

0.05

1
0.8
0.6
0.4
0.2
0
0

5
4
3
2
1
0

5

0.1

0

15

10

0.15

1
0.8
0.6
0.4
0.2
0
0

5
4
3
2
1
0

0

15

10

5

0

1
0.8
0.6
0.4
0.2
0
5

5
4
3
2
1
0

0.15

0

15

10

5

1
0.8
0.6
0.4
0.2
0

0

0.2
0.3
0.4

Effectiveness

5
4
3
2
1
0
0

0.1

Efficiency

Figure A.8 Influences of micro-management preference on team performance under
levels of task dependence

251

0.1
0.05

15
15

10

5

0.15
0.1
0.05

0

5

10

15

0

5

10

15

15

10

5

0

0.15
0.1
0.05

252

15

10

5

0
0

15

10

0

15

10

5

0

0

15

10

5
5

10

0.05

1
0.8
0.6
0.4
0.2
0
0

5
4
3
2
1
0

5

0.1

0

15

10

0.15

1
0.8
0.6
0.4
0.2
0
0

5
4
3
2
1
0

0

15

10

5

0
0

15

0.15

1
0.8
0.6
0.4
0.2
0
5

5
4
3
2
1
0
0

0.6
0.7
0.8

10

1
0.8
0.6
0.4
0.2
0
5

5
4
3
2
1
0
0

0.5

Figure A.8 Cont’d

0.1
0.05

253

0

5

10

15

5

10

15

15

10

5

0.1
0.08
0.06
0.04
0.02
0
0

15

10

0

15

10

5

0
0

15

0.15

1
0.8
0.6
0.4
0.2
0
5

5
4
3
2
1
0
0

1.0

10

1
0.8
0.6
0.4
0.2
0
5

5
4
3
2
1
0
0

0.9

Figure A.8 Cont’d

APPENDIX 3: SAMPLE CODES
A3.1 Project
// The majority of codes of signal project scenario
public class Project

{

@Parameter (displayName = "size", usageName = "size")
public double getSize() {
return size
}
// other parameters are not shown in this document
}
public double roundnum = 0
private static final long serialVersionUID = 1L
protected static long agentIDCounter = 1
protected String agentID = "Project " +
(agentIDCounter++)
@ScheduledMethod(
start = 1d,
interval = 1d,
shuffle = true
)
public def generate() {
def returnValue
def time = GetTickCountInTimeUnits()
if (size>0) {
Context context = FindContext("VOICE")
Iterator list =
context.getObjects(Task.class).iterator()
Projection grid =
(Grid)FindProjection("VOICE/Grid")
int flag=1
while (list.hasNext() && flag==1) {
Task atask = list.next()
setEffective(effective+atask.process)
setIneffective(ineffective+atask.routine+atask.com+atask.mee
ting)
setErrorpercent(perror/(1000-left))
setEffectiveratio(effective/(effective+ineffective))
254

if (atask.dependency==1) {
flag=0
} else {
}
}
if (flag==0) {
} else {
Task task=new Task ()
AddAgentToContext("VOICE", task)
task.Initialize()
Network network =
(Network)FindProjection("VOICE/Work")
Iterator actors = new
NetworkSuccessor(network, this).query().iterator()
Object startActor = new Object()
while (actors.hasNext()) {
Object actor = actors.next()
if (actor.fit==task.fit) {
startActor = actor
} else {
}
}
int x = grid.getLocation(startActor).getX()
int y = grid.getLocation(startActor).getY()
MoveAgent("VOICE/Grid", task, x, y)
setSize(size-task.amount)
}
} else {
Context context = FindContext("VOICE")
Iterator list =
context.getObjects(Task.class).iterator()
if (list.hasNext()) {
} else {
setProjecttick(GetTickCount())
EndSimulationRun()
}
}
return returnValue
}
@ScheduledMethod(
start = 0d,
shuffle = true
255

)
public def Initialize() {
def returnValue
def time = GetTickCountInTimeUnits()
setDependprob(GetParameter("dependprob"))
setCongruence(GetParameter("congruence"))
setRoundnum(GetParameter("roundnum"))
return returnValue
}
@ProbeID()
public String toString() {
def returnValue
def time = GetTickCountInTimeUnits()
returnValue = this.agentID
return returnValue
}
}
// The majority of codes of multi-project scenario
public class Project {
@Parameter (displayName = "Project size", usageName =
"size")
public double getSize() {
return size
}
public void setSize(double newValue) {
size = newValue
}
public double size = 0
// other parameters are not shown in this document
private static final long serialVersionUID = 1L
protected static long agentIDCounter = 1
protected String agentID = "Project " +
(agentIDCounter++)
@ScheduledMethod(
start = 1d,
interval = 1d,
shuffle = true
256

)
public def generate() {
def returnValue
def time = GetTickCountInTimeUnits()
if (size>0) {
Projection grid =
(Grid)FindProjection("VOICE/Grid")
Iterator list = new GridWithin(grid, this,
40).query().iterator()
int flag=1
while (list.hasNext() && flag==1) {
Object o = list.next()
if (o instanceof Task) {
Task otask=(Task)o
if (otask.idnumber == idnumber) {
setEffective(effective+otask.process)
setIneffective(ineffective+otask.routine+otask.com+otask.mee
ting)
setEffectiveratio(effective/(effective+ineffective))
if (otask.dependency==1) {
flag=0
} else {
}
} else {
}
} else {
}
}
if (flag==0) {
} else {
double[][] AmountArray=new double[3][20]
AmountArray=[[5,5,10,15,15,10,20,20,25,10,5,5,10,15,20,15,10
,10,5,5],[5,9,10,10,15,12,20,10,25,10,10,10,10,15,10,15,10,1
0,5,10],[15,15,10,15,15,10,20,20,30,20,5,5,10,15,20,15,10,10
,5,8]]
double[][] DifficultyArray=new double[3][20]
DifficultyArray=[[1,1,1,1.1,1.1,1.2,1.2,1.5,1.2,1,1,1,1.2,1,
0.5,1,1,1,1,1],[1.1,1,1.1,1.1,1.1,1,1.2,1.2,1.2,1,1,1,1.2,1,
0.5,1,1,1,1,1],[1,1.2,1,1.1,1.1,1.2,1.2,1.2,1.2,1,1.1,1,1,1,
0.8,1,1,1.1,1,1.1]]
double[][] PriorityArray=new double[3][20]
257

PriorityArray=[[1,1,1,1.1,1.1,1,1,1,1.2,1,1,1,1.5,1,1,2,1,2,
2,1.5],[1.1,1,1.1,1.1,1.1,1,1.2,1.2,1.2,1,1,1,1.2,1,1,1,1,2,
2,2],[1,1,1,1.2,1.1,1,1,1.2,1.1,1,1.1,1,1,1,1,1,1,2,1.8,2]]
int[][] DependencyArray=new int[3][20]
DependencyArray=[[1,1,1,1,1,0,0,0,1,1,1,0,1,0,0,0,1,1,1,1],[
1,1,0,0,1,0,0,0,1,1,1,0,1,0,0,0,1,1,1,1],[1,1,1,1,1,0,1,0,1,
1,1,0,1,0,0,0,1,1,1,1]]
int[][] NeedinfoArray=new int[3][20]
NeedinfoArray=[[1,1,0,0,0,1,1,1,1,0,0,0,0,0,1,1,1,0,0,0],[1,
1,0,0,0,1,1,1,1,0,0,0,0,0,1,1,1,0,0,0],[1,1,0,0,0,1,1,1,1,0,
0,0,0,0,1,1,1,0,0,0]]
int[][] AuthorityArray=new int[3][20]
AuthorityArray=[[1,1,1,2,2,2,1,1,1,1,1,3,3,1,1,1,1,1,1,2],[1
,2,2,2,2,2,1,1,1,1,1,3,3,1,1,1,1,1,1,2],[1,1,1,1,1,2,1,2,2,1
,1,3,3,1,1,1,1,1,1,1]]
int[][] FitArray=new int[3][20]
FitArray=[[1,1,1,2,2,2,1,1,1,1,1,3,3,1,1,1,1,1,1,2],[1,2,2,2
,2,2,1,3,1,3,1,3,3,1,1,1,1,1,1,2],[1,1,1,1,1,2,1,2,2,1,1,3,3
,1,1,1,1,3,1,1]]
int[][] ConcurArray=new int[3][20]
ConcurArray=[[1,2,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,
2,1,1,2,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,2,1,2,1,1,1,1,1,
1,1,1,1,1,1,1,1,1]]
Task task=new Task ()
AddAgentToContext("VOICE", task)
task.fit = FitArray[idnumber-1][taskid]
Network network =
(Network)FindProjection("VOICE/Work")
Iterator actors = new
NetworkSuccessor(network, this).query().iterator()
Object startActor = new Object()
while (actors.hasNext()) {
Object actor = actors.next()
if (actor.fit==task.fit) {
startActor = actor
} else {
}
}
int x = grid.getLocation(startActor).getX()
int y = grid.getLocation(startActor).getY()
MoveAgent("VOICE/Grid", task, x, y)
task.idnumber=idnumber
258

task.type=type
task.difficulty=complexity*DifficultyArray[idnumber1][taskid]
task.priority=urgent*PriorityArray[idnumber1][taskid]
task.authority = AuthorityArray[idnumber1][taskid]
task.needinfo = NeedinfoArray[idnumber1][taskid]
task.dependency = DependencyArray[idnumber1][taskid]
task.concur = ConcurArray[idnumber-1][taskid]
task.amount = AmountArray[idnumber-1][taskid]
task.left=task.amount
if (size<task.amount) {
task.amount=size
} else {
}
setSize(size-task.amount)
setTaskid(taskid+1)
if (task.concur>1) {
Task task2=new Task ()
AddAgentToContext("VOICE", task2)
task2.fit = FitArray[idnumber-1][taskid]
Iterator actors2 = new
NetworkSuccessor(network, this).query().iterator()
Object startActor2 = new Object()
while (actors2.hasNext()) {
Object actor2 = actors2.next()
if (actor2.fit==task2.fit) {
startActor2 = actor2
} else {
}
}
int x2 =
grid.getLocation(startActor2).getX()
int y2 =
grid.getLocation(startActor2).getY()
MoveAgent("VOICE/Grid", task2, x2, y2)
task2.idnumber=idnumber
task2.type=type
259

task2.difficulty=complexity*DifficultyArray[idnumber1][taskid]
task2.priority=urgent*PriorityArray[idnumber-1][taskid]
task2.authority =
AuthorityArray[idnumber-1][taskid]
task2.needinfo = NeedinfoArray[idnumber1][taskid]
task2.dependency =
DependencyArray[idnumber-1][taskid]
task2.amount = AmountArray[idnumber1][taskid]
task2.left=task2.amount
if (size<task2.amount) {
task2.amount=size
} else {
}
setSize(size-task2.amount)
setTaskid(taskid+1)
} else {
}
}
} else {
Projection grid =
(Grid)FindProjection("VOICE/Grid")
Iterator list = new GridWithin(grid, this,
40).query().iterator()
int flag=1
while (list.hasNext() && flag==1) {
Object o = list.next()
if (o instanceof Task) {
Task otask=(Task)o
setEffective(effective+otask.process)
setIneffective(ineffective+otask.routine)
setEffectiveratio(effective/(effective+ineffective))
if (otask.idnumber == idnumber) {
flag=0
} else {
}
} else {
260

}
}
if (flag==0) {
} else {
Context context =
RemoveAgentFromContext("Grid", this)
}
}
return returnValue
}
@ProbeID()
public String toString() {
def returnValue
def time = GetTickCountInTimeUnits()
returnValue = this.agentID
return returnValue
}
}

A3.2 Task
// The majority of codes
public class Task {
@Parameter (displayName = "amount", usageName = "amount")
public double getAmount() {
return amount
}
public void setAmount(double newValue) {
amount = newValue
}
public double amount = 0
// other parameters are not shown in this document
private static final long serialVersionUID = 1L
protected static long agentIDCounter = 1
protected String agentID = "Task " + (agentIDCounter++)
public def Initialize() {
def returnValue
def time = GetTickCountInTimeUnits()
261

RandomHelper.createNormal(24, 4)
setAmount(Math.floor(0.5*RandomHelper.getNormal().nextDouble
()))
setFit((int)Math.ceil(RandomDraw(0, 3)))
setAuthority(RandomDraw(0, 3))
setLeft(amount)
setPriority(RandomDraw(0, 3))
setDifficulty(1)
RandomHelper.createNormal(0.3, 0.15)
setInfo(RandomHelper.getNormal().nextDouble())
double dependprob= GetParameter("dependprob")
if (RandomDraw(0, 1)>dependprob) {
setDependency(0)
} else {
setDependency(1)
}
return returnValue
}
@ProbeID()
public String toString() {
def returnValue
def time = GetTickCountInTimeUnits()
returnValue = this.agentID
return returnValue
}
}

A3.3 President/VP
// The majority of codes
public class Vp {
@Parameter (displayName = "meeting", usageName =
"meeting")
public int getMeeting() {
return meeting
}
public void setMeeting(int newValue) {
meeting = newValue
}
262

public int meeting = 0
@Parameter (displayName = "urgentTask", usageName =
"urgentTask")
private static voice.Task getUrgentTask() {
return urgentTask
}
private static void setUrgentTask(voice.Task newValue) {
urgentTask = newValue
}
private static voice.Task urgentTask = new
voice.Task()
// other parameters are not shown in this document
private static final long serialVersionUID = 1L
protected static long agentIDCounter = 1
protected String agentID = "Vp " + (agentIDCounter++)
@ScheduledMethod(
start = 1d,
interval = 1d,
shuffle = true
)
public def vp() {
def returnValue
def time = GetTickCountInTimeUnits()
if (meeting>0) {
setMeeting(meeting-1)
} else {
Network network =
(Network)FindProjection("VOICE/Report")
Projection grid =
(Grid)FindProjection("VOICE/Grid")
this.meeting()
this.prioritize()
if (urgentTask.request==1) {
this.approve()
} else {
if (urgentTask.conflict==1) {
this.coordinate()
} else {
}
263

}
}
return returnValue
}
@ScheduledMethod(
start = 0d,
shuffle = true
)
public def initialize() {
def returnValue
def time = GetTickCountInTimeUnits()
setExceptionthred((int)Math.floor(PDF1(a, b)))
setProcesstime(Math.floor(PDF2(c, d)))
setProcesstimeb(processtime)
return returnValue
}
@ProbeID()
public String toString() {
def returnValue
def time = GetTickCountInTimeUnits()
returnValue = this.agentID
return returnValue
}
}

A3.4 Coordinator/Manager
// The majority of codes
public class Coordinator {
@Parameter (displayName = "authoritythres", usageName =
"authoritythres")
public double getAuthoritythres() {
return authoritythres
}
public void setAuthoritythres(double newValue) {
authoritythres = newValue
}

264

@Parameter (displayName = "urgentTask", usageName =
"urgentTask")
private static voice.Task getUrgentTask() {
return urgentTask
}
private static void setUrgentTask(voice.Task newValue) {
urgentTask = newValue
}
private static voice.Task urgentTask = new
voice.Task()
// other parameters are not shown in this document
public double authoritythres = 0
private static final long serialVersionUID = 1L
protected static long agentIDCounter = 1
protected String agentID = "Coordinator " +
(agentIDCounter++)
@ScheduledMethod(
start = 1d,
interval = 1d,
shuffle = true
)
public def coor() {
def returnValue
def time = GetTickCountInTimeUnits()
if (meeting>0) {
setMeeting(meeting-1)
} else {
Projection grid =
(Grid)FindProjection("VOICE/Grid")
GridPoint point = grid.getLocation(this)
setThisX(point.getX())
setThisY(point.getY())
this.prioritize()
if (urgentTask.approval==0 &&
urgentTask.authority > authoritythres) {
this.request()
} else {
if (urgentTask.request==1) {
this.approve()
265

} else {
if (urgentTask.complain==1) {
this.reassign()
} else {
if (urgentTask.conflict==1) {
this.coordinate()
} else {
if (urgentTask.report==1) {
grid.moveTo(urgentTask, urgentTask.staffX, urgentTask.staffY)
urgentTask.report=0
urgentTask.priority=urgentTask.priority-3
} else {
if (urgentTask.submit==1) {
this.monitor()
} else {
this.assign()
}
}
}
}
}
}
}
return returnValue
}
@ProbeID()
public String toString() {
def returnValue
def time = GetTickCountInTimeUnits()
returnValue = this.agentID
return returnValue
}
}
A3.5 Staff
// The majority of codes
public class Staff

{

266

@Parameter (displayName = "competence", usageName =
"competence")
public double getCompetence() {
return competence
}
public void setCompetence(double newValue) {
competence = newValue
}
public double competence = 0
@Parameter (displayName = "urgentTask", usageName =
"urgentTask")
private static voice.Task getUrgentTask() {
return urgentTask
}
private static void setUrgentTask(voice.Task newValue) {
urgentTask = newValue
}
private static voice.Task urgentTask = new voice.Task()
// other parameters are not shown in this document
private static final long serialVersionUID = 1L
protected static long agentIDCounter = 1
protected String agentID = "Staff " + (agentIDCounter++)
@ScheduledMethod(
start = 1d,
interval = 1d,
shuffle = true
)
public def staff() {
def returnValue
def time = GetTickCountInTimeUnits()
if (meeting>0) {
setMeeting(meeting-1)
} else {
Projection grid =
(Grid)FindProjection("VOICE/Grid")
GridPoint point = grid.getLocation(this)
setThisX(point.getX())
setThisY(point.getY())
this.prioritize()
267

setAttt(attt+ttt)
double tick = GetTickCount()
setAveattt(attt/tick)
if (ttt>capacity && urgentTask.must==0) {
this.complain()
} else {
if (urgentTask.approval==0 &&
urgentTask.authority > authoritythres ) {
this.request()
} else {
if (urgentTask.info<infothres &&
urgentTask.round <=
(int)Math.ceil(roundthres/Math.pow(10,timepressure)) &&
urgentTask.infogot==0) {
this.infoneed()
} else {
if (urgentTask.infogot==1 &&
urgentTask.round <= (int)Math.ceil(roundthres)) {
this.monitor()
} else {
if
(urgentTask.round>(int)Math.ceil(roundthres))) {
this.report()
} else {
if (urgentTask.redo==1) {
this.redo()
} else {
this.process()
}
}
}
}
}
}
}
return returnValue
}

A3.6 Engineer
268

// The majority of codes
public class Engineer {
@Parameter (displayName = "meeting", usageName =
"meeting")
public def getMeeting() {
return meeting
}
public void setMeeting(def newValue) {
meeting = newValue
}
public def meeting = 0
@Parameter (displayName = "urgentTask", usageName =
"urgentTask")
private static voice.Task getUrgentTask() {
return urgentTask
}
private static void setUrgentTask(voice.Task newValue) {
urgentTask = newValue
}
private static voice.Task urgentTask = new voice.Task()
// other parameters are not shown in this document
private static final long serialVersionUID = 1L
protected static long agentIDCounter = 1
protected String agentID = "Engineer " +
(agentIDCounter++)
public def infofeed() {
def returnValue
def time = GetTickCountInTimeUnits()
Projection grid = (Grid)FindProjection("VOICE/Grid")
urgentTask.com=1
urgentTask.process=0
urgentTask.routine=0
urgentTask.meeting=0
if (infotime>0) {
setInfotime(infotime-1)
} else {
urgentTask.info=urgentTask.info+(1urgentTask.info)*info
269

urgentTask.infogot=1
urgentTask.engX=thisX
urgentTask.engY=thisY
grid.moveTo(urgentTask, urgentTask.staffX,
urgentTask.staffY)
setInfotime(infotimeb)
}
return returnValue
}
@ProbeID()
public String toString() {
def returnValue
def time = GetTickCountInTimeUnits()
returnValue = this.agentID
return returnValue
}
}

A3.7 Behaviors
Prioritizing

public def prioritize() {
def returnValue
def time = GetTickCountInTimeUnits()
setTtt(0)
setUrgentTask(new Task())
Projection grid = (Grid)FindProjection("VOICE/Grid")
Iterator list = new GridWithin(grid, this,
0).query().iterator()
while (list.hasNext()) {
Task task = list.next()
setTtt(ttt+task.left*task.difficulty/competence)
if (task.priority >= urgentTask.priority) {
setUrgentTask(task)
} else {
}
}
if (urgentTask.amount>0) {
} else {
270

}
return returnValue
}

Processing/Submission

public def process() {
def returnValue
def time = GetTickCountInTimeUnits()
Projection grid = (Grid)FindProjection("VOICE/Grid")
if (urgentTask.left <= 0) {
urgentTask.terror=urgentTask.terror+processerror
urgentTask.staffX=thisX
urgentTask.staffY=thisY
grid.moveTo(urgentTask, upperX, upperY)
urgentTask.submit=1
urgentTask.left=urgentTask.amount
} else {
setFsa(fsa+competence)
setFaa(faa+competence/urgentTask.difficulty)
urgentTask.left=urgentTask.leftcompetence/urgentTask.difficulty
urgentTask.process=1
urgentTask.routine=0
urgentTask.com=0
urgentTask.meeting=0
}
return returnValue
}

Assigning

public def assign() {
def returnValue
271

def time = GetTickCountInTimeUnits()
Projection grid = (Grid)FindProjection("VOICE/Grid")
Network network =
(Network)FindProjection("VOICE/Report")
urgentTask.routine=1
if (mgrpreference==1) {
Iterator lowers = new NetworkPredecessor(network,
this).query().iterator()
Staff bestLower = new Staff()
while (lowers.hasNext()) {
Staff lower = lowers.next()
if (lower.competence >= bestLower.competence)
{
bestLower = lower
} else {
}
}
int loX = grid.getLocation(bestLower).getX()
int loY = grid.getLocation(bestLower).getY()
grid.moveTo(urgentTask, loX, loY);
} else {
if (mgrpreference==2) {
Iterator lowers = new
NetworkPredecessor(network, this).query().iterator()
Staff bestLower = new Staff()
while (lowers.hasNext()) {
Staff lower = lowers.next()
if (lower.processerror <
bestLower.processerror) {
bestLower = lower
} else {
}
}
int loX = grid.getLocation(bestLower).getX()
int loY = grid.getLocation(bestLower).getY()
grid.moveTo(urgentTask, loX, loY);
} else {
Iterator lowers = new
NetworkPredecessor(network, this).query().iterator()
Staff bestLower = new Staff()
while (lowers.hasNext()) {
Staff lower = lowers.next()
272

if (lower.ttt < bestLower.ttt) {
bestLower = lower
} else {
}
}
int loX = grid.getLocation(bestLower).getX()
int loY = grid.getLocation(bestLower).getY()
grid.moveTo(urgentTask, loX, loY);
}
}
return returnValue
}

Requesting

public def request() {
def returnValue
def time = GetTickCountInTimeUnits()
Projection grid = (Grid)FindProjection("VOICE/Grid")
urgentTask.staffX=thisX
urgentTask.staffY=thisY
grid.moveTo(urgentTask, upperX, upperY);
urgentTask.request=1
urgentTask.priority=urgentTask.priority+3
urgentTask.routine=1
urgentTask.process=0
urgentTask.com=0
urgentTask.meeting=1
return returnValue
}

Approving

public def monitor() {
def returnValue
273

def time = GetTickCountInTimeUnits()
urgentTask.process=1
urgentTask.routine=0
urgentTask.com=0
urgentTask.meeting=0
if (processtime<=0) {
setProcesstime(processtimeb)
if
(urgentTask.terror<mgrerrorthred*Math.pow(10,timepressure))
{
Context context = FindContext("VOICE")
Iterator list =
context.getObjects(Project.class).iterator()
Project prjt = list.next()
prjt.left=prjt.left-urgentTask.amount
prjt.perror=prjt.perror+urgentTask.terror*urgentTask.amount
Context context2 =
RemoveAgentFromContext("VOICE", urgentTask)
} else {
Projection grid =
(Grid)FindProjection("VOICE/Grid")
grid.moveTo(urgentTask, urgentTask.staffX,
urgentTask.staffY);
urgentTask.redo=1
urgentTask.priority=urgentTask.priority+3
setException(exception+1)
}
} else {
setProcesstime(processtime-1)
}
return returnValue
}

Information exchange

public def infoneed() {
def returnValue
def time = GetTickCountInTimeUnits()
274

Projection grid = (Grid)FindProjection("VOICE/Grid")
Network infonetwork =
(Network)FindProjection("VOICE/Info")
Iterator infosources = new
NetworkPredecessor(infonetwork, this).query().iterator()
Engineer bestInfosource=new Engineer()
while (infosources.hasNext()) {
Engineer infosource = infosources.next()
bestInfosource = infosource
}
Iterator infojudge = new
NetworkPredecessor(infonetwork, this).query().iterator()
if (infojudge.hasNext()) {
int infoX =
grid.getLocation(bestInfosource).getX()
int infoY =
grid.getLocation(bestInfosource).getY()
urgentTask.staffX=thisX
urgentTask.staffY=thisY
urgentTask.priority=urgentTask.priority+1
urgentTask.round=urgentTask.round+1
grid.moveTo(urgentTask, infoX, infoY);
} else {
urgentTask.info=infothres
urgentTask.terror=urgentTask.terror+0.01
urgentTask.amount=urgentTask.amount+2
}
return returnValue
}
public def infofeed() {
def returnValue
def time = GetTickCountInTimeUnits()
Projection grid = (Grid)FindProjection("VOICE/Grid")
urgentTask.com=1
urgentTask.process=0
urgentTask.routine=0
urgentTask.meeting=0
if (infotime>0) {
setInfotime(infotime-1)
} else {
urgentTask.info=urgentTask.info+(1urgentTask.info)*info
275

urgentTask.infogot=1
urgentTask.engX=thisX
urgentTask.engY=thisY
grid.moveTo(urgentTask, urgentTask.staffX,
urgentTask.staffY)
setInfotime(infotimeb)
}
return returnValue
}

Meeting

public def meeting() {
def returnValue
def time = GetTickCountInTimeUnits()
Network network =
(Network)FindProjection("VOICE/Report")
Projection grid = (Grid)FindProjection("VOICE/Grid")
Iterator lowers = new NetworkPredecessor(network,
this).query().iterator()
while (lowers.hasNext()) {
Coordinator coor= lowers.next()
setException(exception+coor.exception)
}
if (exception>exceptionthred) {
setMeeting((int)Math.ceil(5*exception/exceptionthred))
setException(0)
Iterator objs = new GridWithin(grid, this,
40).query().iterator()
while (objs.hasNext()) {
Object o = objs.next()
if (o instanceof Staff) {
Staff staff=(Staff)o
staff.meeting=meeting+1
} else {
if (o instanceof Coordinator) {
276

Coordinator coor=(Coordinator)o
coor.meeting=meeting+1
coor.exception=0
} else {
if (o instanceof Engineer) {
Engineer eng=(Engineer)o
eng.meeting=meeting+1
} else {
if (o instanceof Task) {
Task task=(Task)o
task.approval=1
task.terror=task.terror*0.1
task.info=task.info*1.5
task.difficulty=task.difficulty*0.9
} else {
}
}
}
}
}
} else {
}
return returnValue
}

Monitoring

public def monitor() {
def returnValue
def time = GetTickCountInTimeUnits()
urgentTask.process=1
urgentTask.routine=0
urgentTask.com=0
urgentTask.meeting=0
if (processtime<=0) {
setProcesstime(processtimeb)
277

if
(urgentTask.terror<mgrerrorthred*Math.pow(10,timepressure))
{
Context context = FindContext("VOICE")
Iterator list =
context.getObjects(Project.class).iterator()
Project prjt = list.next()
prjt.left=prjt.left-urgentTask.amount
prjt.perror=prjt.perror+urgentTask.terror*urgentTask.amount
Context context2 =
RemoveAgentFromContext("VOICE", urgentTask)
} else {
Projection grid =
(Grid)FindProjection("VOICE/Grid")
grid.moveTo(urgentTask, urgentTask.staffX,
urgentTask.staffY);
urgentTask.redo=1
urgentTask.priority=urgentTask.priority+3
setException(exception+1)
}
} else {
setProcesstime(processtime-1)
}
return returnValue
}

Correction/rework

public def reassign() {
def returnValue
def time = GetTickCountInTimeUnits()
Projection grid = (Grid)FindProjection("VOICE/Grid")
Network network =
(Network)FindProjection("VOICE/Report")
urgentTask.routine=1
urgentTask.process=0
urgentTask.com=0
278

setException(exception+1)
urgentTask.meeting=0
if (processtime<=0) {
Iterator lowers = new NetworkPredecessor(network,
this).query().iterator()
Staff bestLower = new Staff()
bestLower.ttt=10000
setProcesstime(processtimeb)
while (lowers.hasNext()) {
Staff lower = lowers.next()
if (lower.ttt <= bestLower.ttt) {
bestLower = lower
} else {
}
}
int loX = grid.getLocation(bestLower).getX()
int loY = grid.getLocation(bestLower).getY()
urgentTask.must=1
urgentTask.coorX=thisX
urgentTask.coorY=thisY
urgentTask.complain=0
grid.moveTo(urgentTask, loX, loY);
} else {
setProcesstime(processtime-1)
}
return returnValue
}
public def redo() {
def returnValue
def time = GetTickCountInTimeUnits()
Projection grid = (Grid)FindProjection("VOICE/Grid")
urgentTask.submit=0
if (urgentTask.left <= 0) {
urgentTask.terror=urgentTask.terror*0.85
urgentTask.staffX=thisX
urgentTask.staffY=thisY
grid.moveTo(urgentTask, upperX, upperY)
urgentTask.submit=1
urgentTask.left=urgentTask.amount
urbanTask.redo=0
} else {
279

setFsa(fsa+competence)
setFaa(faa+competence/urgentTask.difficulty)
urgentTask.left=urgentTask.leftcompetence/urgentTask.difficulty
urgentTask.process=1
urgentTask.routine=0
urgentTask.com=0
}
return returnValue
}

Overburden

public def complain() {
def returnValue
def time = GetTickCountInTimeUnits()
Projection grid = (Grid)FindProjection("VOICE/Grid")
urgentTask.staffX=thisX
urgentTask.staffY=thisY
grid.moveTo(urgentTask, upperX, upperY);
urgentTask.complain=1
return returnValue
}

Conflict management

public def report() {
def returnValue
def time = GetTickCountInTimeUnits()
Projection grid = (Grid)FindProjection("VOICE/Grid")
urgentTask.staffX=thisX
urgentTask.staffY=thisY
grid.moveTo(urgentTask, upperX, upperY);
urgentTask.conflict=1
280

urgentTask.priority=urgentTask.priority+3
urgentTask.routine=1
urgentTask.round=0
return returnValue
}
public def coordinate() {
def returnValue
def time = GetTickCountInTimeUnits()
Projection grid = (Grid)FindProjection("VOICE/Grid")
urgentTask.process=0
urgentTask.routine=1
urgentTask.com=0
urgentTask.meeting=0
if (processtime<=0) {
setProcesstime(processtimeb)
double coordination= RandomDraw(0, 10)
if (coordination>=coorthres) {
grid.moveTo(urgentTask, urgentTask.staffX,
urgentTask.staffY);
urgentTask.conflict=0
urgentTask.info=urgentTask.info+(1urgentTask.info)*congruence
urgentTask.priority=urgentTask.priority-3
} else {
urgentTask.coorX=thisX
urgentTask.coorY=thisY
grid.moveTo(urgentTask, upperX, upperY);
setException(exception+1)
urgentTask.routine=1
}
} else {
setProcesstime(processtime-0.2)
}
return returnValue
}

281

REFERENCES

282

REFERENCES
Adams, B. (2004). "Public meetings and the democratic process." Public Administration
Review 64(1): 43-54.
Ahmad, I. (1999). "Managing, Processing and Communicating Information: What A/E/C
Organizations Should Know." Journal of Management in Engineering 15(4): 3336.
Ahuja, H. N., Dozzi, S. and Abourizk, S. (1994). Project management: techniques in
planning and controlling construction projects, Wiley.
Albanese, R. (1994). "Team Building Process: Key to Better Project Results." Journal of
Management in Engineering 10: 36.
Allen, S. G. (1985). "Why construction industry productivity is declining." The Review of
Economics and Statistics 67(4): 661-669.
Allen, T. J. (1984). "Managing the flow of technology: Technology transfer and the
dissemination of technological information within the R&D organization." MIT
Press Books 1.
Alshawi, M. and Faraj, I. (2002). "Integrated construction environments: technology and
implementation." Construction Innovation: Information, Process, Management
2(1): 33-51.
Amason, A. C. (1996). "Distinguishing the effects of functional and dysfunctional conflict
on strategic decision making: Resolving a paradox for top management teams."
Academy of Management Journal: 123-148.
Ancona, D. G. and Caldwell, D. F. (1992). "Demography and design: Predictors of new
product team performance." Organization Science 3(3): 321-341.
Antony, J. (1976). "How to run a meeting." Harvard Business Review, May-April, pgs:
43-58.
Anvuur, A. (2008). Cooperation in construction projects: concept, antecedents and
strategies. Hong Kong, University of Hong Kong. PhD Thesis.
Anvuur, A. and Kumaraswamy, M. (2007). "Conceptual model of partnering and
alliancing." Journal of Construction Engineering and Management 133: 225.
Arditi, D., Koksal, A. and Kale, S. (2000). "Business failures in the construction
industry." Engineering Construction and Architectural Management 7(2): 120-132.
283

Axelrod, R. and Hamilton, W. (1981). "The evolution of cooperation." Science 211(4489):
1390.
Azhar, S., Hein, M. and Sketo, B. (2008). "Building information modeling (BIM):
Benefits, risks and challenges".
Baccarini, D. (1996). "The concept of project complexity--a review." International
Journal of Project Management 14(4): 201-204.
Baiden, B., Price, A. and Dainty, A. (2006). "The extent of team integration within
construction projects." International Journal of Project Management 24(1): 13-23.
Ballard, G. and Howell, G. (1998). "Shielding production: essential step in production
control." Journal of Construction Engineering and Management 124(1): 11-17.
Balliet, D. (2010). "Communication and cooperation in social dilemmas: A meta-analytic
review." Journal of Conflict Resolution 54(1): 39.
Banks, J., Carson, J., Nelson, B. and Nicol, D. (2005). "Discrete-Event System
Simulation–Prentice Hall." New Jersey.
Barlow, J., Cohen, M., Jashapara, A. and Simpson, Y. (1997). "Towards positive
partnering." Policy, Bristol.
Barrick, M., Stewart, G., Neubert, M. and Mount, M. (1998). "Relating member ability
and personality to work-team processes and team effectiveness." Journal of
Applied Psychology 83: 377-391.
Barthorpe, S., Chien, H. J. and Shih, J. K. C. (2004). "A survey of the potential for
enterprise resource planning (ERP) in improving the effectiveness of construction
management in the UK construction industry." International Journal of computer
applications in technology 20(1): 120-128.
Bayliss, R., Cheung, S., Suen, H. and Wong, S. (2004). "Effective partnering tools in
construction: a case study on MTRC TKE contract 604 in Hong Kong."
International Journal of Project Management 22(3): 253-263.
Becerik, B. (2004). "A review on past, present and future of web based project
management & collaboration tools and their adoption by the US AEC industry."
International Journal of IT in Architecture Engineering and Construction 2: 233248.
Belassi, W. and Tukel, O. (1996). "A new framework for determining critical
success/failure factors in projects." International Journal of Project Management
14(3): 141-151.
284

Bennett, J. (2000). Construction, the third way: managing cooperation and competition in
construction, Butterworth-Heinemann.
Bennett, J. and Jayes, S. (1995). Trusting the team: the best practice guide to partnering
in construction, Thomas Telford.
Bentley, M. (1981). Quality control on building sites. Building Research Station Current
Paper 7/81. Building Research Establishment. Watford, UK.
Bertelsen, S. (2003a). "Complexity–Construction in a new Perspective." IGLC-11,
Blacksburg, Virginia.
Bertelsen, S. (2003b). "Construction as a complex system".
Bewsey, K. and McCord, D. (2006). Psychopathy and Cooperation: Personality Traits as
Predictors of Cooperation and Success in a Prisoner's Dilemma, Western Carolina
University.
Bijlsma, K. and Costa, A. C. (2003). "Risks at work: Trust, monitoring and cooperation in
temporary project teams".
Bishop, k. (2008). Is Integrated Project Delivery A Revolution or Gimmick? . Engieering
News-Record, McGraw Hill Construction.
Boden, D. (1995). "Agendas and arrangements: Everyday negotiations in meetings." The
discourse of negotiation: Studies of language in the workplace: 83-99.
Bodwell, D. J. (2006). "A Historical Perspective of High Performing Teams." from
http://highperformanceteams.org/hpt_history.htm.
Boetti, G. and Leandro, V. (2008). Use of Social Network Analysis as a tool to evaluate
success on construction projects and team interactions, UNIVERSITY OF
COLORADO AT BOULDER.
Bonabeau, E. (2002). "Agent-based modeling: Methods and techniques for simulating
human systems." Proceedings of the National Academy of Sciences of the United
States of America 99(Suppl 3): 7280.
Bonoma, T. (1976). "Conflict, cooperation and trust in three power systems." Behavioral
Science 21(6): 499-514.
Bordat, C., McCullouch, B. and Sinha, K. (2004). "An Analysis of Cost Overruns and
Time Delays of INDOT Projects." Joint Transportation Research Program: 11.

285

Bosch-Sijtsema, P., Postma, T., Bosch-Sijtsema, P. and Postma, T. (2006). "Knowledge
transfer in project-based environments: a study on innovation projects in the
construction industry."
Bouchlaghem, D., Kimmance, A. and Anumba, C. (2004). "Integrating product and
process information in the construction sector." Industrial Management and Data
Systems 104(3): 218-233.
Bourn, J. (2001). "Modernising Construction." UK: National Audit Office.
Bowden, S., Dorr, A., Thorpe, T. and Anumba, C. (2006). "Mobile ICT support for
construction process improvement." Automation in Construction 15(5): 664-676.
Brannick, M. T. and Prince, C. (1997). "An overview of team performance measurement."
Team performance assessment and measurement: Theory, methods, and
applications: 3-16.
Bresnen, M., Goussevskaia, A. and Swan, J. (2005). "Organizational routines, situated
learning and processes of change in project-based organizations." Project
Management Journal 36(3): 27-41.
Bresnen, M. and Marshall, N. (2000a). "Building partnerships: case studies of client–
contractor collaboration in the UK construction industry." Construction
Management and Economics 18(7): 819-832.
Bresnen, M. and Marshall, N. (2000b). "Partnering in construction: a critical review of
issues, problems and dilemmas." Construction Management and Economics 18(2):
229-237.
Bromiley, P. and Harris, J. (2006). "Trust, transaction cost economics, and mechanisms."
Handbook of trust research: 124.
Bryant, P. T. (2006). "Decline of the engineering class: Effects of global outsourcing of
engineering services." Leadership and Management in Engineering 6: 59.
Burton, R. M., Obel, B., Hunter, S., Søndergaard, M. and Døjbak, D. (1998). Strategic
organizational diagnosis and design: Developing theory for application, Kluwer
Academic Pub.
BusinessDictionary. (2012). "Organizational Structure." 2012, from
http://www.businessdictionary.com/definition/organizational-structure.html.
Cederman, L. E. (2003). "Modeling the size of wars: from billiard balls to sandpiles."
American Political Science Review 97(01): 135-150.
286

Chan, A., Chan, D. and Ho, K. (2003). "An empirical study of the benefits of construction
partnering in Hong Kong." Construction Management and Economics 21(5): 523533.
Chan, A. P. C., Chan, D. W. M., Chiang, Y., Tang, B., Chan, E. H. W. and Ho, K. S. K.
(2004). "Exploring critical success factors for partnering in construction projects."
Journal of Construction Engineering and Management 130: 188.
Chan, A. P. C., Chan, D. W. M. and Yeung, J. F. (2009). Relationship Contracting for
Construction Excellence: Principles, Practices and Case Studies, Taylor &
Francis Group.
Chatman, J. and Barsade, S. (1995). "Personality, organizational culture, and cooperation:
Evidence from a business simulation." Administrative Science Quarterly 40(3).
Chen, C., Chen, X. and Meindl, J. (1998). "How can cooperation be fostered? The cultural
effects of individualism-collectivism." Academy of Management Review 23(2):
285-304.
Chen, H. M. and Tien, H. C. (2007). "Application of peer-to-peer network for real-time
online collaborative computer-aided design." Journal of Computing in Civil
Engineering 21: 112.
Chen, W., Jin, R. and Sudjianto, A. (2005). "Analytical variance-based global sensitivity
analysis in simulation-based design under uncertainty." Journal of mechanical
design 127: 875.
Cheng, E., Li, H., Love, P. and Irani, Z. (2001). "An e-business model to support supply
chain activities in construction." Logistics Information Management 14(1/2): 6878.
Cheng, M. and Tsai, M. (2003). "Reengineering of construction management process."
Journal of Construction Engineering and Management 129: 105.
Cheung, S., Ng, T., Wong, S. and Suen, H. (2003). "Behavioral aspects in construction
partnering." International Journal of Project Management 21(5): 333-343.
Cheung, S. O., Yiu, T. W. Y. and Yeung, S. F. (2006). "A study of styles and outcomes in
construction dispute negotiation." Journal of Construction Engineering and
Management 132: 805.
Chinowsky, P., Diekmann, J. and Galotti, V. (2008). "Social network model of
construction." Journal of Construction Engineering and Management 134: 804.

287

Chinowsky, P., Taylor, J. E. and Di Marco, M. (2009). "Project Network Interdependency
Alignment: A New Approach to Assessing Project Effectiveness".
Chinowsky, P. S., Diekmann, J. and O’Brien, J. (2010). "Project Organizations as Social
Networks." Journal of Construction Engineering and Management 136: 452.
Chinowsky, P. S. and Rojas, E. M. (2004). "Virtual teams: a guide to successful
implementation", ASCE.
Christiansen, J. H. and Altaweel, M. (2004). "Simulation of Natural and Social Process
Interactions in Bronze Age Mesopotamian Settlement Systems." 69th Annual
Proceedings of the Society for American Archeology.
Christodoulou, S. (2010). "Scheduling Resource-Constrained Projects with Ant Colony
Optimization Artificial Agents." Journal of Computing in Civil Engineering 24: 45.
Chua, D., Kog, Y. and Loh, P. (1999). "Critical success factors for different project
objectives." Journal of Construction Engineering and Management 125(3): 142150.
Cicmil, S. and Marshall, D. (2005). "Insights into collaboration at the project level:
complexity, social interaction and procurement mechanisms." Building Research
& Information 33(6): 523-535.
Clemmer, T., Spuhler, V., Berwick, D. and Nolan, T. (1998). "Cooperation: the
foundation of improvement." Annals of Internal medicine 128(12 Part 1): 1004.
Cohen, S. G. and Bailey, D. E. (1997). "What makes teams work: Group effectiveness
research from the shop floor to the executive suite." Journal of management 23(3):
239.
Collier, N. (2003). "Repast: An extensible framework for agent simulation." The
University of Chicago’s Social Science Research 36.
Consoli, A., Tweedale, J. and Jain, L. (2007). "The link between agent coordination and
cooperation." Intelligent Information Processing III: 11-19.
Cornick, T. and Mather, J. (1999). Construction project teams: making them work
profitably, Thomas Telford Services Ltd.
Cowan, C. (1992). "Partnering—A concept for success".
Cox, T., Griffiths, A. and Rial, E. (2010). "Work Related Stress." Occupational Health
Psychology.
288

Curtis, B. (1989). "Modeling coordination from field experiments".
Cyert, R. and March, J. (2005). A behavioral theory of the firm, Blackwell.
Dabke, S., Salem, O., Genaidy, A. and Daraiseh, N. (2008). "Job satisfaction of women in
construction trades." Journal of Construction Engineering and Management 134:
205.
Daft, R. (2009). Organization theory and design, South-Western Pub.
Davis, K. A. and Songer, A. D. (2009). "Resistance to IT Change in the AEC Industry:
Are the Stereotypes True?" Journal of Construction Engineering and Management
135: 1324.
Davis, T. R. V. (1985). "The influence of the physical environment in offices." Journal of
Library Administration 5(4): 91-111.
De Saram, D. and Ahmed, S. (2001a). "Construction coordination activities: what is
important and what consumes time." Journal of Management in Engineering 17(4):
202-213.
De Saram, D. D. and Ahmed, S. M. (2001b). "Construction coordination activities: what
is important and what consumes time." Journal of Management in Engineering
17(4): 202-213.
Deshpande, R. and Webster Jr, F. E. (1989). "Organizational culture and marketing:
defining the research agenda." The Journal of Marketing 53(1): 3-15.
Deutsch, M. (1949). "A theory of co-operation and competition." Human Relations 2(2):
129.
Deutsch, M. (1973). "The resolution of conflict." American Behavioral Scientist 17(2):
248.
Deutsch, M. (1980). "Fifty years of conflict." Retrospections on social psychology: 46-77.
Deutsch, M. and Krauss, R. M. (1960). "The effect of threat upon interpersonal
bargaining." Journal of Abnormal and Social Psychology 61(2): 181-189.
Devine, D., Clayton, L., Philips, J., Dunford, B. and Melner, S. (1999). "Teams in
organizations." Small Group Research 30(6): 678.
Diallo, A. and Thuillier, D. (2005). "The success of international development projects,
trust and communication: an African perspective." International Journal of
Project Management 23(3): 237-252.
289

Dikmen, I. and Birgönül, M. T. (2003). "Strategic perspective of Turkish construction
companies." Journal of Management in Engineering 19: 33.
Dorée, A. and Holmen, E. (2004). "Achieving the unlikely: innovating in the loosely
coupled construction system." Construction Management and Economics 22(8):
827-838.
Du, J. and El-Gafy, M. (2012). "Virtual Organizational Imitation for Construction
Enterprises (VOICE): An Agent Based Simulation Framework for Exploring
Human and Organizational Implications in Construction Management." Journal of
Computing in Civil Engineering 26:282.
Dubois, A. and Gadde, L. (2002). "The construction industry as a loosely coupled system:
implications for productivity and innovation." Construction Management and
Economics 20(7): 621-631.
Ducrot, R., Le Page, C., Bommel, P. and Kuper, M. (2004). "Articulating land and water
dynamics with urbanization: an attempt to model natural resources management at
the urban edge." Computers, Environment and Urban Systems 28(1-2): 85-106.
Dukerich, J., Golden, B. and Shortell, S. (2002). "Beauty is in the eye of the beholder: the
impact of organizational identification, identity, and image on the cooperative
behaviors of physicians." Administrative Science Quarterly 47(3): 507-537.
Dulaimi, M., Ling, F. and Bajracharya, A. (2003). "Organizational motivation and interorganizational interaction in construction innovation in Singapore." Construction
Management and Economics 21(3): 307-318.
Dulaimi, M., Y. Ling, F., Ofori, G. and De Silva, N. (2002). "Enhancing integration and
innovation in construction." Building Research & Information 30(4): 237-247.
Eastman, C. M., Eastman, C., Teicholz, P. and Sacks, R. (2008). BIM Handbook: A guide
to building information modeling for owners, managers, designers, engineers, and
contractors, John Wiley & Sons Inc.
Eckblad, S., Ashcraft, H., Audsley, P., Blieman, D., Bedrick, J., Brewis, C., Hartung, R.,
Onuma, K., Rubel, Z. and Stephens, N. (2007). "Integrated Project Delivery-A
Working Definition." AIA California Council, Sacramento, CA.
Eclipse. (2012). "Official website of Eclipse." 2012, from http://www.eclipse.org/.
Egan, J. (1998a). "Rethinking construction." Department of the Environment, Transport
and the Regions, London.

290

Egan, J. (1998b). "Rethinking construction: The report of the construction task force."
DETR, London.
Egan, J. (2002). "Accelerating change".
Epstein, J. and Axtell, R. (1996). Growing artificial societies: social science from the
bottom up, The MIT Press.
Epstein, J. M. (1999). "Agent-based computational models and generative social science."
Generative Social Science: Studies in Agent Based Computational Modeling: 4–46.
Eriksson, P. E. and Laan, A. (2007). "Procurement effects on trust and control in clientcontractor relationships." Engineering, Construction and Architectural
Management 14(4): 387-399.
Eriksson, P. E. and Pesämaa, O. (2007). "Modelling procurement effects on cooperation."
Construction Management and Economics 25(8): 893-901.
Fan, S., Shen, Q. and Lin, G. (2007). "Comparative Study of Idea Generation between
Traditional Value Management Workshops and GDSS-Supported Workshops."
Journal of Construction Engineering and Management 133: 816.
Fayol, H., Gray, I., Electrical, I. o. and Engineers, E. (1929). General and industrial
management, Pitman London.
Feinberg, T. (2009) "The Team Performance Curve." Ezine Articles.
Fisher, N. and Green, S. (2001). "Partnering and the UK construction industry: The first
ten years—A review of the literature." Modernising Construction: Report by the
Comptroller and Auditor General (HC 87): 58-66.
Fleetham, C. and Griesmer, S. K. (2006). "Leveraging Personality for Business Success."
Leadership and Management in Engineering 6: 160.
Flyvbjerg, B., Bruzelius, N. and Rothengatter, W. (2003). Megaprojects and risk: An
anatomy of ambition, Cambridge Univ Pr.
Folcik, V. and Orosz, C. (2006). "An agent-based model demonstrates that the immune
system behaves like a complex system and a scale-free network." SwarmFest 2006.
Fong, P. S. W. and Lung, B. W. C. (2007). "Interorganizational teamwork in the
construction industry." Journal of Construction Engineering and Management 133:
157.

291

Fortune, J. and White, D. (2006). "Framing of project critical success factors by a systems
model." International Journal of Project Management 24(1): 53-65.
Galbraith, J. (1973). Designing complex organizations, Addison-Wesley Reading, MA.
Galbraith, J. R. and Nathanson, D. A. (1982). Strategy implementation: The role of
structure and process, West St Paul, MN.
George, J. M. (1992). "Extrinsic and intrinsic origins of perceived social loafing in
organizations." Academy of Management Journal 35(1): 191-202.
Giritli, H. and Civan, I. (2008). "Personality Study of Construction Professionals in the
Turkish Construction Industry." Journal of Construction Engineering and
Management 134: 630.
Girmscheid, G. and Brockmann, C. (2010). "Inter-and Intraorganizational Trust in
International Construction Joint Ventures." Journal of Construction Engineering
and Management 136: 353.
Gladstein, D. L. (1984). "Groups in context: A model of task group effectiveness."
Administrative Science Quarterly: 499-517.
Gouldner, A. W. (1954). Patterns of industrial bureaucracy, Free Press New York.
Graham, R. (1989). Project management as if people mattered. Bala Cynwyd, PA,
Primavera Press.
Grant, A. (2007). "Relational job design and the motivation to make a prosocial
difference." Academy of Management Review 32(2): 393.
Gratch, J. and Marsella, S. (2004). "A domain-independent framework for modeling
emotion." Cognitive Systems Research 5(4): 269-306.
Gudgel, J. (2008). "Interoperability in the Construction Industry." Smart Market Report.
Mc Graw Hill Construction.
Hackman, J. (1990). Groups that work (and those that don't): Creating conditions for
effective teamwork, Jossey-Bass San Francisco, CA.
Hall, D. S. (2004). "Work-related stress of registered nurses in a hospital setting." Journal
for Nurses in Staff Development 20(1): 6.
Hamel, G., Doz, Y. and Prahalad, C. (1989). "Collaborate with your competitors and
win." Harvard Business Review 67(1): 133-139.

292

Hechter, M. (1988). Principles of group solidarity, Univ of California Pr on Demand.
Hise, R., O'Neal, L., Parasuraman, A. and McNeal, J. (1990). "Marketing/R&D
interaction in new product development: implications for new product success
rates." Journal of Product Innovation Management 7(2): 142-155.
Hoegl, M., Weinkauf, K. and Gemuenden, H. (2004). "Interteam coordination, project
commitment, and teamwork in multiteam R&D projects: A longitudinal study."
Organization Science 15(1): 38-55.
Horii, T., Jin, Y. and Levitt, R. (2005a). "Modeling and analyzing cultural influences on
project team performance." Computational & Mathematical Organization Theory
10(4): 305-321.
Horii, T., Jin, Y. and Levitt, R. E. (2005b). "Modeling and analyzing cultural influences
on project team performance." Computational & Mathematical Organization
Theory 10(4): 305-321.
Humpel, N. and Caputi, P. (2001). "Exploring the relationship between work stress, years
of experience and emotional competency using a sample of Australian mental
health nurses." Journal of Psychiatric and Mental Health Nursing 8(5): 399-403.
Ioannidou, D. and Fruchter, R. (2009). "Predictive Simulation as a Decision Support
System to Manage A/E/C Global Teamwork", ASCE.
Itoh, H. (1994). "Job design, delegation and cooperation: A principal-agent analysis* 1."
European Economic Review 38(3-4): 691-700.
Jaselskis, E. and Ashley, D. (1991). "Optimal allocation of project management resources
for achieving success." Journal of Construction Engineering and Management
117(2): 321-340.
Jehn, K. A. (1995). "A multimethod examination of the benefits and detriments of
intragroup conflict." Administrative Science Quarterly: 256-282.
Jenner, M. S. (1997). Designing High-Performance Teams for Projects: A Study of 49
Project Teams in the United Kingdom Construction Industry, Open University.
PhD.
Jesus, M. and Vorster, M. C. (2002). "Relationship between personality traits and
performance for engineering and architectural professionals providing design
services." Journal of Management in Engineering 18: 158.
Jha, K. and Iyer, K. (2006). "Critical determinants of project coordination." International
Journal of Project Management 24(4): 314-322.
293

Jin, Y. and Levitt, R. (1996). "The virtual design team: A computational model of project
organizations." Computational & Mathematical Organization Theory 2(3): 171195.
John, G. and Martin, J. (1984). "Effects of organizational structure of marketing planning
on credibility and utilization of plan output." Journal of Marketing Research 21(2):
170-183.
Johnson, D. (1975). "Cooperativeness and social perspective taking." Journal of
Personality and Social Psychology 31(2): 241-244.
Johnson, H. M. and Singh, A. (1998). "Personality of civil engineers." Journal of
Management in Engineering 14(4): 45-56.
Joseph, G. (2005). Global Virtual Engineering Team Utilization in the Engineering,
Procurement, and Construction (EPC) Industry, Pennsylvania State University.
Kadefors, A. (2004). "Trust in project relationships--inside the black box." International
Journal of Project Management 22(3): 175-182.
Kaman, V. S. and McCambridge, J. A. (1992). "Conflict Management Training for
Today's Engineering Managers." Journal of Management in Engineering 8: 298.
Kanawattanachai, P. and Yoo, Y. (2002). "Dynamic nature of trust in virtual teams." The
Journal of Strategic Information Systems 11(3-4): 187-213.
Kang, J. H., Anderson, S. D. and Clayton, M. J. (2007). "Empirical study on the merit of
web-based 4D visualization in collaborative construction planning and
scheduling." Journal of Construction Engineering and Management 133: 447.
Kangari, R. (1988). "Business failure in construction industry." Journal of Construction
Engineering and Management 114(2): 172-190.
Kanter, R. (1994). "Collaborative advantage: the art of alliances." Harvard Business
Review 72: 96-96.
Katzenbach, J., Smith, D. and Bookspan, M. (1993). The wisdom of teams, Harvard
Business School Press Boston.
Katzenbach, J. R. and Smith, D. K. (1993). The wisdom of teams: Creating the highperformance organization, Harvard Business Press.
Kazanjian, R., Drazin, R. and Glynn, M. (2000). "Creativity and technological learning:
the roles of organization architecture and crisis in large-scale projects." Journal of
Engineering and Technology Management 17(3-4): 273-298.
294

Keller, R. T. and Holland, W. E. (1983). "Communicators and innovators in research and
development organizations." Academy of Management Journal 26(4): 742-749.
Khalfan, M., McDermott, P. and Swan, W. (2007). "Building trust in construction
projects." Supply Chain Management: An International Journal 12(6): 385-391.
Kim, J. and Burton, R. M. (2002). "The effect of task uncertainty and decentralization on
project team performance." Computational & Mathematical Organization Theory
8(4): 365-384.
Kim, K. (2010). "Case Study on the Evaluation of Equipment Flow at a Construction
Site." Journal of Computing in Civil Engineering 1: 28.
Klammer, T. P. (2002). Team Performance: Using Financial Measures to Evaluate the
Influence of Support Systems on Team Performance, University of North Texas.
Knaup, A. and Piazza, M. (2007). "Business Employment Dynamics data: survival and
longevity, II." Monthly Labor Review, BLS 130(9): 3-10.
Korczynski, M. (2000). "The political economy of trust." Journal of Management Studies
37(1): no-no.
Kozlowski, S. W. J. and Bell, B. S. (2003). "Work groups and teams in organizations."
Kristof-Brown, A. L. and Stevens, C. K. (2001). "Goal congruence in project teams: Does
the fit between members' personal mastery and performance goals matter?"
Journal of Applied Psychology 86(6): 1083.
Kunda, G. (2006). Engineering culture: Control and commitment in a high-tech
corporation, Temple Univ Pr.
Kunz, J. C., Levitt, R. E. and Thomsen, J. (1997). "Intervention studies using the virtual
design team", IEEE.
Kuprenas, J. A. and Nasr, E. H. B. (2000). "Personalities of Construction Project
Managers—A Link to High Performance Teams", ASCE.
Lam, K., So, A., Hu, T., Ng, T., Yuen, R., Lo, S., Cheung, S. and Yang, H. (2001). "An
integration of the fuzzy reasoning technique and the fuzzy optimization method in
construction project management decision-making." Construction Management
and Economics 19(1): 63-76.
Langfred, C. W. (2000). "Work-Group Design and Autonomy." Small Group Research
31(1): 54.
295

Latham, M. (1994). "Constructing the team Joint Review of Procurement and Contractual
Arrangements in the United Kingdom Construction Industry; final report."
Laughlin, P. (1978). "Ability and group problem solving." Journal of Research and
Development in Education 12(1): 114-120.
Lawless, A. (2001). "The Inherent Personality of Women as an Asset to the Engineering
Workplace." Leadership and Management in Engineering 1(4): 36-40.
LeBaron, B. (2002). "Building the Santa Fe artificial stock market." Physica A.
Leicht, R. M., Hunter, S. T., Saluja, C. and Messner, J. I. (2010). "Implementing
Observational Research Methods to Study Team Performance in Construction
Management." Journal of Construction Engineering and Management 136: 76.
Levitt, R. E. (2007). "The Virtual Design Team (VDT): A Computational Model of
Project Teams."
Li, H., Cheng, E., Love, P. and Irani, Z. (2001). "Co-operative benchmarking: a tool for
partnering excellence in construction." International Journal of Project
Management 19(3): 171-179.
Li, H., Cheng, E. W. L. and Love, P. E. D. (2000). "Partnering research in construction."
Engineering, Construction and Architectural Management 7(1): 76-92.
Ligmann-Zielinska, A. and Jankowski, P. (2007). "Agent-based models as laboratories for
spatially explicit planning policies." Environment and Planning B: Planning and
Design 34(2): 316-335.
Liu, A. M. M. and Zhai, X. (2010). "Influences of Personality on the Adoption of Conflict
Handling Styles and Conflict Outcomes for Facility Managers." Journal of Legal
Affairs and Dispute Resolution in Engineering and Construction 1(1): 17.
Loch, C. and Terwiesch, C. (1998). "Communication and uncertainty in concurrent
engineering." Management Science 44(8): 1032-1048.
Love, P., Gunasekaran, A. and Li, H. (1998). "Concurrent engineering: a strategy for
procuring construction projects." International Journal of Project Management
16(6): 375-383.
Love, P., Irani, Z., Cheng, E. and Li, H. (2002). "A model for supporting inter
organizational relations in the supply chain." Engineering Construction and
Architectural Management 9(1): 2-15.

296

Lu, L. and Argyle, M. (1991). "Happiness and cooperation." Personality and Individual
Differences 12(10): 1019-1030.
Lundin, R. A. and Söderholm, A. (1995). "A theory of the temporary organization."
Scandinavian Journal of management 11(4): 437-455.
Macal, C. and North, M. (2007). "Agent-based modeling and simulation: desktop ABMS",
IEEE Press Piscataway, NJ, USA.
Macal, C. M. and North, M. J. (2005). "Tutorial on agent-based modeling and simulation",
Winter Simulation Conference.
Maheswari, J., Varghese, K. and Sridharan, T. (2006). "Application of dependency
structure matrix for activity sequencing in concurrent engineering projects."
Journal of Construction Engineering and Management 132: 482.
Malcurat, O., Bignon, J. C. and Halin, G. (2000). "Improving cooperation in small scale
projects", ASCE.
Malone, T. and Crowston, K. (1990). "What is coordination theory and how can it help
design cooperative work systems?", ACM.
Malone, T. and Crowston, K. (1994). "The interdisciplinary study of coordination." ACM
Computing Surveys (CSUR) 26(1): 87-119.
March, J. G. and Simon, H. A. (1958). "Organizations New York." NY: John Wiley.
Mathieu, J., Maynard, M., Rapp, T. and Gilson, L. (2008). "Team effectiveness 19972007: A review of recent advancements and a glimpse into the future." Journal of
management 34(3): 410.
Mathieu, J. E., Heffner, T. S., Goodwin, G. F., Salas, E. and Cannon-Bowers, J. A. (2000).
"The influence of shared mental models on team process and performance."
Journal of Applied Psychology 85(2): 273-283.
Matthews, O. and Howell, G. A. (2005). "Integrated project delivery: an example of
relational contracting." Lean construction journal 2(1): 46-61.
Mayer, R. C., Davis, J. H. and Schoorman, F. D. (1995). "An integrative model of
organizational trust." The Academy of Management Review 20(3): 709-734.
Maznevski, M. L. and Chudoba, K. M. (2000). "Bridging space over time: Global virtual
team dynamics and effectiveness." Organization Science: 473-492.

297

McAllister, D. (1995). "Affect-and cognition-based trust as foundations for interpersonal
cooperation in organizations." Academy of Management Journal 38(1): 24-59.
McCabe, B., Karahalios, D. and Loughlin, C. (2005). "Attitudes in Construction Safety",
ASCE.
McCann, J., Galbraith, J. R., Nystrom, P. C. and Starbuck, W. H. (1981). Handbook of
Organizational Design, Oxford University Press New York.
McCowan, I., Bengio, S., Gatica-Perez, D., Lathoud, G., Monay, F., Moore, D., Wellner,
P. and Bourlard, H. (2003). "Modeling human interaction in meetings", IEEE.
McDonough III, E. F., Kahnb, K. B. and Barczaka, G. (2001). "An investigation of the
use of global, virtual, and colocated new product development teams." Journal of
Product Innovation Management 18(2): 110-120.
McGrath, J. (1984). Groups: Interaction and performance, Prentice-Hall Englewood
Cliffs, NJ.
Mead, M. (2003). Cooperation and competition among primitive peoples, Transaction
Pub.
Meissner, U. F., Rueppel, U., Theiss, M. and Lange, M. (2005). "An Agent Based Model
Compound for Fire Protection Engineering", ASCE.
Migliaccio, G. C. and Martinez, R. M. (2010). "Establishing a Collaborative Environment
among Project Stakeholders", ASCE.
Miles, M. B. and Laboratories, N. T. (1964). On temporary systems, National Training
Laboratories, National Education Association.
Miller, D. M., Fields, R., Ashish, K. and Ortiz, R. (2000). "Leadership and organizational
vision in managing a multiethnic and multicultural project team." Journal of
Management in Engineering 16(6): 18-22.
Mintzberg, H. (1983). Power in and around organizations, Prentice-Hall Englewood
Cliffs, NJ.
Miozzo, M. and Dewick, P. (2004). "Networks and innovation in European construction:
benefits from inter-organisational cooperation in a fragmented industry."
International Journal of Technology Management 27(1): 68-92.
Moenaert, R. K. and Souder, W. E. (1990). "An anlysis of the use of extrafunctional
information by R&D and marketing personnel: Review and model* 1." Journal of
Product Innovation Management 7(3): 213-229.
298

Mohamed, Y. and AbouRizk, S. (2005). "Framework for building intelligent simulation
models of construction operations." Journal of Computing in Civil Engineering 19:
277.
Moore, D. R. and Dainty, A. R. J. (2001). "Intra-team boundaries as inhibitors of
performance improvement in UK design and build projects: a call for change."
Construction Management and Economics 19(6): 559-562.
Moorman, C., Deshpande, R. and Zaltman, G. (1993). "Factors affecting trust in market
research relationships." The Journal of Marketing 57(1): 81-101.
Morgan, R. M. and Hunt, S. D. (1994). "The commitment-trust theory of relationship
marketing." The Journal of Marketing 58(3): 20-38.
Motowidlo, S. and Van Scotter, J. (1994). "Evidence that task performance should be
distinguished from contextual performance." Journal of Applied Psychology 79(4):
475-480.
Munns, A. (1995). "Potential influence of trust on the successful completion of a project."
International Journal of Project Management 13(1): 19-24.
Nakata, C. and Im, S. (2010). "Spurring Cross Functional Integration for Higher New
Product Performance: A Group Effectiveness Perspective*." Journal of Product
Innovation Management 27(4): 554-571.
Naoum, S. (2003). "An overview into the concept of partnering." International Journal of
Project Management 21(1): 71-76.
NASFA, COAA, APPA, AGC and AIA (2010). Integrated Project Delivery For Public
and Private Owners.
Nayak, N. V. and Taylor, J. E. (2009). "Offshore outsourcing in global design networks."
Journal of Management in Engineering 25: 177.
Newcombe, R. (1996). "Empowering the construction project team." International
Journal of Project Management 14(2): 75-80.
Nguyen, H. H. D., Le, H. and Boles, T. (2010). "Individualism–Collectivism and Co
operation: A Cross Society and Cross Level Examination." Negotiation and
Conflict Management Research 3(3): 179-204.
Nitithamyong, P. and Skibniewski, M. J. (2004). "Web-based construction project
management systems: how to make them successful?" Automation in Construction
13(4): 491-506.
299

Nonaka, I. and Konno, N. (1999). "The concept of ‘Ba’: building a foundation for
knowledge creation." The knowledge management yearbook 1999-2000: 37.
North, M. and Macal, C. (2007). Managing business complexity: discovering strategic
solutions with agent-based modeling and simulation, Oxford University Press,
USA.
Novelo, S. and Gabriel, F. (2010). "Strategic development of high performance teams in
large-scale construction projects."
Okuwoga, A. A. (1998). "Cost-time performance of public sector housing projects in
Nigeria*." Habitat International 22(4): 389-395.
Ortiz de Orue, D., Taylor, J., Chanmeka, A. and Weerasooriya, R. (2009). "Robust project
network design." Project Management Journal.
Oshinubi, O. O. (2008). The influence of project managers' leadership styles on project
team performance in the construction industry, UNIVERSITY OF PHOENIX.
Ouchi, W. G. (1980). "Markets, bureaucracies, and clans." Administrative Science
Quarterly 25(1): 129-141.
Pennings, J. M. (1975). "Interdependence and complementarity-The case of a brokerage
office." Human Relations 28(9): 825.
Peters, M. (1986). "Innovation for hospitals: an application of the product development
process." Journal of health care marketing 6(3): 52.
Pfeffer, J. (1994). Managing with power: Politics and influence in organizations, Harvard
Business Press.
Phua, F. (2004). "The antecedents of co-operative behaviour among project team
members: an alternative perspective on an old issue." Construction Management
and Economics 22(10): 1033-1045.
Phua, F. and Rowlinson, S. (2004). "How important is cooperation to construction project
success? A grounded empirical quantification." Engineering, Construction and
Architectural Management 11(1): 45-54.
Pich, M., Loch, C. and De Meyer, A. (2002a). "On uncertainty, ambiguity, and
complexity in project management." Management Science 48(8): 1008-1023.
Pich, M. T., Loch, C. H. and De Meyer, A. (2002b). "On uncertainty, ambiguity, and
complexity in project management." Management Science: 1008-1023.
300

Pinto, J. and Slevin, D. (1988). "Critical success factors across the project life cycle."
Project Management Journal 19(3): 67-75.
Pinto, J. K., Slevin, D. P. and English, B. (2009). "Trust in projects: An empirical
assessment of owner/contractor relationships." International Journal of Project
Management 27(6): 638-648.
Pinto, M. and Pinto, J. (1990). "Project team communication and cross-functional
cooperation in new program development* 1." Journal of Product Innovation
Management 7(3): 200-212.
Pinto, M., Pinto, J. and Prescott, J. (1993a). "Antecedents and consequences of project
team cross-functional cooperation." Management Science: 1281-1297.
Pinto, M., Pinto, J. and Prescott, J. (1993b). "Antecedents and consequences of project
team cross-functional cooperation." Management Science 39(10): 1281-1297.
Pocock, J. and Kim, M. (1997). "Impact of management approach on project interaction
and performance." Journal of Construction Engineering and Management 123:
411.
Podsakoff, N., Whiting, S., Podsakoff, P. and Blume, B. (2009). "Individual-and
organizational-level consequences of organizational citizenship behaviors: A
meta-analysis." Journal of Applied Psychology 94(1): 122-141.
Podsakoff, P. and MacKenzie, S. (1997). "Impact of organizational citizenship behavior
on organizational performance: A review and suggestion for future research."
Human Performance 10(2): 133-151.
Powell, A., Piccoli, G. and Ives, B. (2004). "Virtual teams: a review of current literature
and directions for future research." ACM SIGMIS Database 35(1): 6-36.
Prietula, M., Carley, K. and Gasser, L. (1998). Simulating organizations: Computational
models of institutions and groups, The MIT Press.
Project Management Institue, P. (2008). A Guide to the Project Management Body of
Knowledge: PMBOK Guide (Fouth Edition), Project Management Institute, Inc.
Project Management Institute, P. (2004). A Guide to the Project Management Body of
Knowledge:(PMBOK Guide), Project Management Institute Newtown Square PA,
USA.
Pruitt, D. G. and Drews, J. L. (1969). "The effect of time pressure, time elapsed, and the
opponent's concession rate on behavior in negotiation." Journal of Experimental
Social Psychology 5(1): 43-60.
301

Rabbie, J. (1991). Determinants of instrumental intra-group cooperation. Cooperation and
Prosocial Behavior. R. Hinde and Groebel, J. Cambridge, Cambridge University
Press.
Raboud, P. (1988). "Safety on Large Building Construction Projects." Journal of
Construction Engineering and Management 114: 286.
Ren, Z., Anumba, C. and Ugwu, O. (2001). "Construction claims management: towards
an agent-based approach." Engineering Construction and Architectural
Management 8(3): 185-197.
Ring, P. and Van de Ven, A. (1994). "Developmental processes of cooperative
interorganizational relationships." Academy of Management Review: 90-118.
Robbins, S. (2005). Organizational Behavior (11th), Prentice Hill.
Robbins, S. and Langton, N. (1998). Organizational behavior: Concepts, controversies,
and applications, Prentice Hall Englewood Cliffs, NJ.
Roberts, G. (2005). "Cooperation through interdependence." Animal Behaviour 70(4):
901-908.
Robey, D., Khoo, H. M. and Powers, C. (2000). "Situated learning in cross-functional
virtual teams." Professional Communication, IEEE Transactions on 43(1): 51-66.
Rojas-Villafane, J. A. (2010). "An Agent-based Model of Team Coordination and
Performance."
Rojas, E. and Aramvareekul, P. (2003). "Is Construction Labor Productivity Really
Declining?" Journal of Construction Engineering and Management 129: 41.
Ruekert, R. and Walker Jr, O. (1987a). "Marketing's interaction with other functional
units: a conceptual framework and empirical evidence." The Journal of Marketing
51(1): 1-19.
Ruekert, R. W. and Walker Jr, O. C. (1987b). "Interactions between marketing and R&D
departments in implementing different business strategies." Strategic Management
Journal 8(3): 233-248.
Ruikar, D., Anumba, C., Duke, A., Carrillo, P. and Bouchlaghem, N. (2007). "Using the
semantic web for project information management." Facilities 25(13): 507-524.
Sako, M., LANE, C. and BACHMAN, R. (2006). "Does trust improve business
performance?" Organizational trust: a reader: 267.
302

Salas, E., Stagl, K. C., Burke, C. S. and Goodwin, G. F. (2007). "Fostering team
effectiveness in organizations: Toward an integrative theoretical framework."
Modeling complex systems 52: 185-243.
Sally, D. (1995). "Conversation and cooperation in social dilemmas." Rationality and
Society 7(1): 58.
Sandberg, J. (2000). "Understanding human competence at work: an interpretative
approach." Academy of Management Journal: 9-25.
Sargent, R. G. (2004). "Validation and verification of simulation models", IEEE.
Schalcher, H. (2009). "Complexity in construction."
Schermerhorn Jr, J. (1975). "Determinants of interorganizational cooperation." Academy
of Management Journal 18(4): 846-856.
Schieritz, N. and Milling, P. M. (2003). "Modeling the forest or modeling the trees".
Schmidt, S. M. and Kochan, T. A. (1972). "Conflict: Toward conceptual clarity."
Administrative Science Quarterly: 359-370.
Schmitt, D. (1984). "Interpersonal relations: Cooperation and competition." Journal of the
Experimental Analysis of Behavior 42(3): 377.
Schwartzman, H. (1989). The meeting: Gatherings in organizations and communities,
Springer Us.
Shan, Y. (2010). "Worker Satisfaction and Work Life Related Characteristics in the US
Construction Industry", ASCE.
Shek-Pui Wong, P. and Cheung, S. O. (2004). "Trust in construction partnering: views
from parties of the partnering dance." International Journal of Project
Management 22(6): 437-446.
Shen, F. Y. and Chang, A. S. (2010). "Exploring Coordination Goals of Construction
Projects." Journal of Management in Engineering 1: 25.
Shirazi, B., Langford, D. and Rowlinson, S. (1996). "Organizational structures in the
construction industry." Construction Management and Economics 14(3): 199-212.
Shohet, I. and Frydman, S. (2003a). "Communication patterns in construction at
construction manager level." Journal of Construction Engineering and
Management 129(5): 570-577.

303

Shohet, I. M. and Frydman, S. (2003b). "Communication patterns in construction at
construction manager level." Journal of Construction Engineering and
Management 129(5): 570-577.
Singh, A. and Eng, P. (2009). "Organizational power in perspective." Leadership and
Management in Engineering 9: 165.
Singh, A. and Jampel, G. (2010). "Leadership Flexibility Space." Journal of Management
in Engineering 26: 176.
Smith, K., Carroll, S. and Ashford, S. (1995). "Intra-and interorganizational cooperation:
Toward a research agenda." Academy of Management Journal 38(1): 7-23.
Smith, K. G., Smith, K. A., Olian, J. D., Sims Jr, H. P., O'Bannon, D. P. and Scully, J. A.
(1994). "Top management team demography and process: The role of social
integration and communication." Administrative Science Quarterly: 412-438.
Soibelman, L. and Pena-Mora, F. (2000). "Distributed multi-reasoning mechanism to
support conceptual structural design." Journal of Structural Engineering 126: 733.
Soltani, A., Tawfik, H., Goulermas, J. and Fernando, T. (2002). "Path planning in
construction sites: performance evaluation of the Dijkstra, A*, and GA search
algorithms." Advanced Engineering Informatics 16(4): 291-303.
Son, J. W. and Rojas, E. M. (2010). "The Evolution of Collaboration in Temporary
Project Teams: An Agent Based Modeling and Simulation Approach." Journal of
Construction Engineering and Management 1: 231.
Souder, W. (1988). "Managing relations between R&D and marketing in new product
development projects* 1." Journal of Product Innovation Management 5(1): 6-19.
Souder, W. E. (1981). "Disharmony between R&D and marketing* 1." Industrial
Marketing Management 10(1): 67-73.
Tajfel, H. (1982). "Social psychology of intergroup relations." Annual review of
psychology 33(1): 1-39.
Taylor, F. W. (1911). "Scientific management." New York.
Taylor, J. and Levitt, R. (2007). "Innovation alignment and project network dynamics: An
integrative model for change." Project Management Journal 38(3): 22-35.
Taylor, J. E., Levitt, R. and Villarroel, J. A. (2009). "Simulating Learning Dynamics in
Project Networks." Journal of Construction Engineering and Management 135:
1009.
304

Teicholz, P., Goodrum, P. and Haas, C. (2001). "US Construction Labor Productivity
Trends, 1970–1998." Journal of Construction Engineering and Management 127:
427.
Tellegen, A. (1991). "Personality traits: Issues of definition, evidence, and assessment."
Thinking clearly about psychology: Personality and psychopathology 2: 10–35.
Tepper, S. (2004). "Setting Agendas and Designing Alternatives: Policymaking and the
Strategic Role of Meetings1." Review of Policy Research 21(4): 523-542.
Terhune, K. (1970). "The effects of personality in cooperation and conflict." The structure
of conflict: 193-234.
Tesluk, P. E. and Mathieu, J. E. (1999). "Overcoming roadblocks to effectiveness:
Incorporating management of performance barriers into models of work group
effectiveness." Journal of Applied Psychology 84(2): 200-217.
Thiry, M. and Deguire, M. (2007). "Recent developments in project-based organisations."
International Journal of Project Management 25(7): 649-658.
Thomas, G. and Thomas, M. (2005). Construction partnering & integrated teamworking,
Wiley-Blackwell.
Thompson, J. (2003). Organizations in action: Social science bases of administrative
theory, Transaction Pub.
Thompson, J. D. (1967). Organizations in Action. New York, McGraw-Hill.
Thompson, P. and Sanders, S. (1998). "Partnering continuum." Journal of Management in
Engineering 14(5): 73-78.
Thomsen, J. (1998). The virtual team alliance (VTA): Modeling the effects of goal
incongruency in semi-routine, fast-paced project organizations, Stanford
University.
Thomsen, J., Levitt, R. E. and Nass, C. I. (2005). "The Virtual Team Alliance (VTA):
Extending Galbraith’s Information-Processing Model to Account for Goal
Incongruency." Computational & Mathematical Organization Theory 10(4): 349372.
Tjosvold, D. (1988). "Cooperative and competitive dynamics within and between
organizational units." Human Relations 41(6): 425.

305

Tommelein, I. D., Riley, D. R. and Howell, G. A. (1999). "Parade game: impact of work
flow variability on trade performance." Journal of Construction Engineering and
Management 125(5): 304-310.
Trevino, L. K., Lengel, R. H. and Daft, R. L. (1987). "Media symbolism, media richness,
and media choice in organizations." Communication Research 14(5): 553.
Tuckman, B. W. (1965). "Developmental sequence in small groups." Psychological
bulletin 63(6): 384.
Tyler, T. and Blader, S. (2000). Cooperation in groups: Procedural justice, social identity,
and behavioral engagement, Hogrefe & Huber.
Tyler, T. R. and Blader, S. L. (2003). "The group engagement model: Procedural justice,
social identity, and cooperative behavior." Personality and Social Psychology
Review 7(4): 349.
Unsal, H. and Taylor, J. (2009). "Integrating Agent Based Simulation with Game Theory
to Investigate the Hold Up Problem in Project Networks". CRC 2009, Seattle, WA,
ASCE.
Unsal, H. I. and Taylor, J. E. (2010). "Modeling Inter Firm Dependency: A Game
Theoretic Simulation to Examine the Hold Up Problem in Project Networks."
Journal of Construction Engineering and Management 1: 195.
Vaaland, T. (2004). "Improving project collaboration: start with the conflicts."
International Journal of Project Management 22(6): 447-454.
Van de Ven, A. H., Delbecq, A. L. and Koenig Jr, R. (1976). "Determinants of
coordination modes within organizations." American sociological review 41(2):
322-338.
van Marrewijk, A. (2005). "Strategies of cooperation: control and commitment in megaprojects." M@n@gement 8(4).
Varvel, T., Adams, S. G., Pridie, S. J. and Ulloa, B. C. R. (2004). "Team effectiveness and
individual Myers-Briggs personality dimensions." Journal of Management in
Engineering 20: 141.
Verville, J. and Halingten, A. (2003). "The effect of team composition and group role
definition on ERP acquisition decisions." Team Performance Management 9(5/6):
115-130.
Wagner III, J. (1995). "Studies of individualism-collectivism: Effects on cooperation in
groups." Academy of Management Journal 38(1): 152-172.
306

Walsh, K. and Sawhney, A. (2004). "Agent-Based Modeling of Worker Safety Behavior
at the Construction Workface".
Watkins, M., Mukherjee, A., Onder, N. and Mattila, K. (2009). "Using Agent-Based
Modeling to Study Construction Labor Productivity as an Emergent Property of
Individual and Crew Interactions." Journal of Construction Engineering and
Management 135: 657.
Welling, D. and Kamann, D. (2001). "Vertical cooperation in the construction industry:
Size does matter." Journal of Supply Chain Management 37(4): 28-33.
Westerveld, E. (2003). "The Project Excellence Model®: linking success criteria and
critical success factors." International Journal of Project Management 21(6): 411418.
Whitley, R. (2006). "Project-based firms: new organizational form or variations on a
theme?" Industrial and Corporate Change 15(1): 77.
Winch, G. (1989). "The construction firm and the construction project: a transaction cost
approach." Construction Management and Economics 7(4): 331-345.
Winch, G. (2003). "How innovative is construction? Comparing aggregated data on
construction innovation and other sectors–a case of apples and pears."
Construction Management and Economics 21(6): 651-654.
Witt, L. (1998). "Enhancing organizational goal congruence: A solution to organizational
politics." Journal of Applied Psychology 83(4): 666.
WL, C. (2004). "Strategic alliances: a model for establishing long-term commitment to
inter-organizational relations in construction." Building and environment 39(4):
459-468.
Wong, K., Unsal, H., Taylor, A. and Levitt, R. (2009). "The Global Dimension of Robust
Project Network Design." Journal of Construction Engineering and Management
1: 95.
Wong, P. S. P., Cheung, S. O. and Ho, P. K. M. (2005). "Contractor as Trust Initiator in
Construction Partnering—Prisoner’s Dilemma Perspective." Journal of
Construction Engineering and Management 131: 1045.
Wong, S. S. and Burton, R. M. (2000). "Virtual teams: what are their characteristics, and
impact on team performance?" Computational & Mathematical Organization
Theory 6(4): 339-360.

307

Wood, G., McDermott, P. and Swan, W. (2002). "The ethical benefits of trust-based
partnering: the example of the construction industry." Business Ethics A European
Review 11(1): 4-13.
Xue, X., Li, X., Shen, Q. and Wang, Y. (2005). "An agent-based framework for supply
chain coordination in construction." Automation in Construction 14(3): 413-430.
Xue, X., Wang, Y., Shen, Q. and Yu, X. (2007). "Coordination mechanisms for
construction supply chain management in the Internet environment." International
Journal of Project Management 25(2): 150-157.
Yilmaz, C. and Hunt, S. (2001). "Salesperson cooperation: The influence of relational,
task, organizational, and personal factors." Journal of the Academy of Marketing
Science 29(4): 335.
Yiu, T. W. and Lee, H. K. (2010). "How Do Personality Traits Affect Construction
Dispute Negotiation? A Study of Big Five Personality Model." Journal of
Construction Engineering and Management 1: 183.
Zheng, X., Zhong, T. and Liu, M. (2009). "Modeling crowd evacuation of a building
based on seven methodological approaches." Building and environment 44(3):
437-445.
Zineldin, M. and Jonsson, P. (2000). "An examination of the main factors affecting
trust/commitment in supplier-dealer relationships: an empirical study of the
Swedish wood industry." The TQM Magazine 12(4): 245-266.

308