Policy analysis for improving performance of a construction project by system dynamics modeling

The first part of the research is to gain insight into problems and success factors pertinent to large construction projects in Vietnam.. Chapter Title Page 8.2 Implications of the Base

POLICY ANALYSIS FOR IMPROVING PERFORMANCE OF A CONSTRUCTION PROJECT BY SYSTEM DYNAMICS MODELING

by

Nguyen Duy Long

A thesis submitted in partial fulfillment of the requirements for the degree of

Previous Degree Bachelor of Engineering in Civil Engineering

Hochiminh City University of Technology Hochiminh City, Vietnam

Scholarship Donor Asian Development Bank – Japan Scholarship

Program (ADB-JSP)

Asian Institute of Technology School of Civil Engineering

Thailand April 2003

ACKNOWLEDGMENTS

Many institutions and persons have contributed to the work that culminates in this thesis

My graduate study seemed impossible if I had not received a scholarship from donors Therefore, I would firstly like to thank Asian Institute of Technology (AIT) and Asian Development Bank (ADB) for granting me the scholarship

To all the faculties: Professor Ogunlana, Dr Chotchai, Dr Suthi, Dr Hadikusumo, and Professor Brockman, I would like to greatly appreciate their lectures and their respectable profundities from which I have luckily learned during my study life in AIT My special thanks extended to Dr Xu Honggang who has been working for Chongshan University (China) and used to be an AIT’s adjunct faculty and taught me sound fundamentals of system dynamics

My work on my thesis and the research which preceded it benefited from the excellent tutelage and guidance of Professor Ogunlana and the valuable advice and extensive helps

of the other professors in my examination committee, Dr Truong Quang and Dr Hadikusumo My special thanks also go to project managers and the rest of the Hai Van Pass Tunnel project team and practitioners who helped my data collection for the research

Next, I would like to thank my professors, senior lecturers and colleagues in Hochiminh City University of Technology for their admirable suggestions and continuous supports I would like to thank all my friends in Construction Management 2001 batch and in AIT for the wonderful time and unforgettable moments that we had been studying and sharing together In addition, I am very grateful for extensive efforts and timely helps of the staffs

in the School of Civil Engineering and AIT

I owe thanks to those who have helped me maintain a life within which to work during my time at AIT Primary among them are my family who created and kept a home for me to come home to Finally, thanks are due to my sweetheart whom I am separated from only by space

ABSTRACT

Although many studies have been undertaken on factors influencing delays, cost overruns, productivity and safety performance, etc and problems in specific types of projects, these studies seldom discuss common and general problems of construction projects Thus, comprehensive studies on these problems and respective solutions are essential The first part of the research is to gain insight into problems and success factors pertinent to large construction projects in Vietnam The research identifies components mainly responsible for the problems as (1) incompetent designers/contractors, (2) poor estimation and change management, (3) social and technological issues, (4) poor site issues, and (5) improper techniques and tools The components for the success of the large construction projects are (1) comfort, (2) competence, (3) commitment, and (4) communication

The second and major part of the research is concerned with understanding and use of project dynamics in order to achieve successful management of large construction projects

in today’s ever-evolving business environment This research investigates the impacts of dynamic project structure on performance of a road tunnel project with a focus on the influences of resource reallocation in terms of resource quantity and characteristics, overtime reduction and material incentives An applicable simulation model of a large construction project was built using system dynamics methodology The model was calibrated to a road tunnel project to investigate its dynamic problems and to formulate and evaluate practical policies to improve its performance In order to considerably improve performance and to be able to adapt to different scenarios, six integrated policies combining the individual policies having the most potentials were evaluated The result reveals that reasonable increases of manpower and equipment, material incentives and/or one or two of them combined with no-overtime are the most effective policies for the case study project

TABLE OF CONTENTS

Title Page i

Acknowledgments ii

Abstract iii

Table of Contents iv

List of Figures ix

List of Tables xi

1 INTRODUCTION 1 1.1 Background 1

1.2 Problem Statement 2 1.3 Objectives 3 1.4 Scope of Work 3 1.5 Expected Contribution 4 2 LITERATURE REVIEW 5 2.1 General 5 2.2 Characteristics of Construction Industry 5

2.3 Construction Management 5 2.3.1 Important Elements of Construction Management 6 2.3.2 Multiple Project Objectives and Their Trade-Off 6 2.3.3 Characteristics of Traditional Approaches 7 2.3.4 Factors Affecting Project Performance 8 2.3.5 Causes and Costs of Rework 9 2.3.6 Problems of Construction Projects 10

2.3.7 Success Factors for Construction Projects 12

2.4 Applications of System Dynamics in Management 12

2.4.1 The Roles of System Dynamics 12

2.4.2 New Paradigms for Complex Projects 13

2.4.3 System Dynamics in Project Management 14

2.4.4 System Dynamics in Construction Management 16

2.5 Summary 18

3 METHODOLOGY 19

3.1 Introduction 19

3.2 Principles of System Dynamics Modeling 19

3.3 System Dynamics Modeling Process 19

3.4 Study Methodology 20

3.4.1 Questionnaire Survey 21

3.4.2 System Dynamics Modeling 22

3.4.2.1 Problem Articulation 22

3.4.2.2 Feedback Structures as Dynamic Hypotheses 22

3.4.2.3 Formulating a Simulation Model 23

3.4.2.4 Model Testing 23

3.4.2.5 Policy Design and Evaluation 24

3.5 The Major Elements of System Dynamics 24

3.5.1 Feedback 24

Chapter Title Page

3.5.2 Time Delays 24

3.5.3 Stocks and Flows 25

3.5.4 Nonlinearity 25

3.6 Tools for the Research 25

4 PROBLEMS AND SUCCESS FACTORS 26

4.1 Introduction 26

4.2 Respondent’s Characteristics 26

4.3 Achievements of Project Objectives 27

4.4 Problems of Large Construction Projects 27

4.4.1 Rankings of Problems’ Occurrence and Influence 27

4.4.2 Factor Analysis of Problems’ Occurrence 29

4.5 Success Factors for Large Construction Projects 32

4.5.1 Analysis and Ranking of Success Factors 33

4.5.2 Factor Analysis of the Success Factors 34

4.5.3 Discussion of Factor Analysis for the Success Factors 35

4.6 Summary of Problems and Success Factors 36

5 PROBLEM ARTICULATION AND FEEDBACK STRUCTURES 38

5.1 Introduction 38

5.2 Hai Van Pass Tunnel Project 38

5.2.1 Project Description 38

5.2.2 Tunnel Construction Methods 39

5.2.3 Problems Encountered 40

5.3 Reference Modes 40

5.3.1 Project Schedule 40

5.3.2 Budget Status 41

5.3.3 Quality goal 41

5.4 Key Feedback Structures of Construction Projects 42

5.4.1 The Structure of Labor 42

5.4.2 The Structure of Equipment 43

5.4.3 The Structure of Material 43

5.4.4 The Structure of Labor-Equipment Interaction 44

5.4.5 The Structure of Schedule 44

5.4.6 The Structure of Rework 45

5.4.7 The Structure of Safety 46

5.4.8 The Structure of Quality 46

5.5 Summary 47

6 MODEL DESCRIPTION 48

6.1 Introduction 48

6.2 Model Boundary 48

6.3 Model Structure 49

6.3.1 Model Subsystems 49

6.3.2 Scope Subsystem 50

6.3.3 Progress and Rework Subsystem 51

6.3.4 Resources Subsystem 52

6.3.4.1 Manpower Sector 52

Chapter Title Page

6.3.4.2 Major Equipment Sector 54

6.3.4.3 Supportive Equipment Sector 55

6.3.4.4 Material Sector 56

6.3.4.5 Skills sector 57

6.3.5 Performance Subsystem 58

6.3.5.1 Labor Productivity Sector 58

6.3.5.2 Equipment Productivity Sector 59

6.3.5.3 Experience Sector 60

6.3.5.4 The Safety Sector 61

6.3.5.5 The Quality of Practice Sector 62

6.3.5.6 Work-month Sector 62

6.3.5.7 The Supervision Sector 63

6.3.6 Cost Breakdown Subsystem 64

6.3.6.1 Material Cost Sector 64

6.3.6.2 Labor Cost Sector 65

6.3.6.3 Equipment Cost Sector 65

6.3.7 Objectives Control Subsystem 66

6.3.7.1 Schedule Control Sector 66

6.3.7.2 Cost Control Sector 67

6.3.7.3 Quality Control Sector 68

6.4 Model References 69

6.5 Model Description Summary 70

7 MODEL TESTING 71

7.1 General 71

7.2 Model Behavior 71

7.2.1 Introduction 71

7.2.2 Work and Rework 71

7.2.3 Resource Allocation and Progress 72

7.2.4 Labor Productivity 74

7.2.5 Experience and Safety 75

7.2.6 Scheduling and Work-month 77

7.2.7 Project Expenses 78

7.2.8 Summary of Model Behavior 79

7.3 Model Testing 79

7.3.1 Introduction 79

7.3.2 Boundary Adequacy Tests 79

7.3.3 Structure Assessment Tests 79

7.3.4 Dimensional Consistency 80

7.3.5 Parameter Assessment 80

7.3.6 Extreme Condition Tests 80

7.3.7 Integration Error Tests 80

7.3.8 Behavior Reproduction Tests 82

7.3.9 Sensitivity Analysis Tests 83

7.3.10 Model testing summary 84

8 POLICY ANAYSIS 85

8.1 Introduction 85

Chapter Title Page

8.2 Implications of the Base Run and the Project Practice 85

8.2.1 Simulated and Actual Progresses 85

8.2.2 The Tunneling Method 85

8.2.3 The Project Control Mechanism 85

8.2.4 Employment and Workers’ Morale 86

8.2.5 Equipment Investment 86

8.3 Policy Alternatives 86

8.3.1 Adjustments of the Proportion of Skilled and Unskilled 86

Workers 8.3.2 Adjustments of the Proportion of Management Team and 87

Workforce 8.3.3 Reallocation of the Resources 87

8.3.4 Reduction of Overtime and Worker’s Fatigue 89

8.3.5 Pay Increases as Material Incentives 89

8.3.6 Summary of the Policy Alternatives 90

8.4 Policy Evaluation 91

8.4.1 Evaluation of Separate Policies 91

8.4.1.1 Skills Adjustment Policies 91

8.4.1.2 Management Team Adjustment Policies 92

8.4.1.3 Resource Reallocation Policies 93

8.4.1.4 Overtime Reduction and Material Incentives 94

Policies 8.4.2 Evaluation for Integrated Policies 95

8.5 Policy Analysis Summary 96

9 CONCLUSIONS AND RECOMMENDATIONS 97

9.1 General Conclusions 97

9.2 Problems of Large Construction Projects 97

9.3 Success Factors of Large Construction Projects 97

9.4 Importance of the Dynamics of Construction Projects 98

9.5 Major Findings and Discussion about ASM 99

9.5.1 The Trustworthiness of ASM 99

9.5.2 The Contribution of Equipment to Project Progress 99

9.5.3 The Dynamics of Manpower and Equipment Interaction 99

9.5.4 Feedback, Delays and Nonlinear Relationships in 100

Construction Projects 9.5.5 Policies for Improving Performance of the 100

Case Study Project 9.6 Contributions 101

9.6.1 Lessons Learned for Vietnam Construction Industry 101

9.6.2 A New Promising Tool for Construction Practitioners 101

9.7 Limitations of the Research 102

9.7.1 Limitations of the Survey 102

9.7.2 Limitations of the Dynamic Simulation Model 102

9.8 Recommendations for Future Research 102

References 104

Appendix A Questionnaire 110

Chapter Title Page

Appendix B Structured Interview 114

Appendix C Ranking of Problems’ Occurrence 125

Appendix D Ranking of Problems’ Influence 127

Appendix E Model Equations 129

Appendix F Sensitivity Analysis Results of Selected Parameters 137

Appendix G Sensitivity Analysis of Selected Nonlinear Relationships 141 Appendix H Selected Plots of Sensitivity Analysis 143

Appendix I Project Schedule and Progress 145

Appendix K Model Variables 146

LIST OF FIGURES

2.2 Framework for determining Critical Success/Failure factors 9

2.3 Fishbone diagram for identifying causes of rework 10

2.4 Project complexity 13

2.5 Overview of the SYDPIM process logic 15

3.1 Iterative cycle of modeling process 19

3.2 Flowchart of research methodology 21

3.3 Sixteen steps involved in the construction of reference mode 23

3.4 Positive feedback versus negative feedback 24

3.5 Stock and flow diagramming notation 25

4.1 Scree plot of factor analysis of problems’ occurrence 30

4.2 Scree plot of factor analysis of success factors 34

5.1 Drill and blast method 39

5.2 Hydraulic hammering method 39

5.3 The dynamics of the project showing schedule delays 40

5.4 The dynamics of the project showing cost overruns 41

5.5 The dynamics of the project showing quality erosion 42

5.6 The labor structure 42

5.7 The equipment structure 43

5.8 The material structure 44

5.9 The labor-equipment interaction structure 44

5.10 The schedule structure 45

5.11 The rework structure 45

5.12 The safety structure 46

5.13 The quality structure 47

6.1 Model breakdown structure 50

6.2 Scope and scope change sectors 50

6.3 Progress and rework sectors 51

6.4 Manpower sector 52

6.5 Major equipment sector 54

6.6 Supportive equipment sector 55

6.7 Supportive equipment sector 56

6.8 Skills sector 57

6.9 Labor productivity sector 58

6.10 Equipment productivity sector 59

6.11 Experience sector 60

6.12 Safety sector 61

6.13 Quality of practice sector 62

6.14 Work-month sector 62

6.15 Supervision sector 64

6.16 Material cost sector 64

6.17 Labor cost sector 65

6.18 Equipment cost sector 66

6.19 Schedule control sector 66

6.20 Cost control sector 68

Figure No Title Page

6.21 Quality control sector 69

7.1 Work and rework 72

7.2 Work quality 72

7.3 Project progress 73

7.4 Manpower input 73

7.5 Equipment input 74

7.6 Material input 74

7.7 Material delivery 75

7.8 Labor productivity 75

7.9 Cumulative experience and experience index 76

7.10 Accident frequency and accident index 76

7.11 Scheduling and schedule pressure 77

7.12 Average work-month and overtime index 77

7.13 Project cash flow 78

7.14 Project costing and budget status 78

7.15 Response to no purchase 81

7.16 Response to no employment 81

7.17 Sensitivity to time step 82

7.18 Work accomplished – simulated output and historical data 82

7.19 Unskilled workers – simulated output and historical data 82

7.20 Skilled workers – simulated output and historical data 83

7.21 Management team – simulated output and historical data 83

8.1 Key feedbacks in response to skills adjustment 92

8.2 Key feedbacks in response to management team adjustment 93

8.3 Progress under policy P3b 94

8.4 Progress under policy P3c 94

H.1 Experience effect on labor productivity 143

H.2 Experience effect on equipment productivity 143

H.3 Experience effect on rework 143

H.4 Schedule pressure effect on work-month 143

H.5 Schedule pressure effect on labor productivity 143

H.6 Schedule pressure effect on equipment productivity 143

H.7 Schedule pressure effect on rework 143

H.8 Fatigue effect on labor productivity 143

H.9 Fatigue effect on equipment productivity 144

H.10 Fatigue effect on rework 144

H.11 Skills effect on rework 144

H.12 Skills effect on labor productivity 144

H.13 Supervision effect on rework 144

H.14 Supervision effect on productivity 144

H.15 Accidents effect on labor productivity 144

H.15 Accidents effect on equipment productivity 144

LIST OF TABLES

2.1 Evolution of models of project management 8

2.2 Characteristics of the traditional and SD approaches 15

4.1 Questionnaire return rate 26

4.2 Respondents’ position 26

4.3 Construction experience 26

4.4 Construction sector 27

4.5 Project types 27

4.6 Failure to meet project objectives 27

4.7 Problems as high occurrence 28

4.8 Problems as high influence 28

4.9 Ranking of degree of occurrence of problem categories 29

4.10 Ranking of degree of influence of problem categories 29

4.11 KMO and Bartlett's Test 30

4.12 Rotated component matrix (factor loading) for problems’ occurrence 30

4.13 Factor analysis grouping for problems’ occurrence 31

4.14 Ranking of success factors 33

4.15 KMO and Bartlett's Test 34

4.16 Rotated factor matrix (factor loading) for success factors 34

4.17 Factor analysis grouping for success factors 35

6.1 Model boundary chart 48

6.2 Model references 69

7.1 Parameter estimation 80

8.1 Policy alternatives 91

8.2 Project performance in response to skills adjustment 92

8.3 Project performance in response to management team adjustment 92

8.4 Project performance in response to resource reallocation 93

8.5 Project performance in response to no overtime and material incentives 94

8.6 Project performance in response to integrated policies 95

F.1 Results of sensitivity analysis at 48th-month 137

G.1 Range of values in nonlinear relationships 141

CHAPTER 1 INTRODUCTION 1.1 Background

Construction industry has been considered as one of the most important industries of every economy in the worldwide Within a certain society, it interacts nearly all fields of human endeavors, and this diversity is reflected in its projects, which are always a very substantial part of human civilization (Ogunlana et al., 2002) It has been, however, recognized that construction is a fragmented industry together with various negative factors In other words, it is usually characterized by high capital investment, dependence on developers, subcontractors and suppliers, an extensive and complex regulatory framework, high interest costs, severe competition, multi-work locations, fluctuating workforce, and large turnover of workers In addition, companies in this industry have faced various risks including availability of funds for repayment of borrowings, availability of skilled labor and fluctuation of material costs and interest rates All impede its developments and performance improvements at both construction industry and construction organizations Moreover, it is known that the productivity of construction industry has improved much slower than that of other industries (Oglesby et al., 1989)

Large-scaled construction projects are usually extremely complex in dynamic and changing environments Uncertainty is also an inherent aspect of most projects (Meyer et al., 2002), and hence construction projects are not the exception Furthermore, Sterman (1992) asserted that large-scaled construction projects belong to the class of complex dynamic systems with: (1) very complex, consisting of multiple interdependent components, (2) highly dynamic, (3) multiple feedback processes, (4) nonlinear relationships, and (5) involving both “hard” and “soft” data These make construction project management very challenging and extremely sophisticated

ever-Construction project management is often one of the most crucial elements of construction companies It usually requires multidiscipline, a variety of technical and managerial skills

as well as many management tools to deal with fast-moving and high-intensive workflow, interactive and interdependent components, and uncertain environment Unfortunately, delays, cost overruns and low level of quality are very common in construction since its large-scaled projects are consistently mismanaged endeavors in today‟s modern society (Sterman, 2000)

Depending on different management processes – strategic or operational levels and the degree of the project complexity and uncertainty, different analytic tools incorporated with experience, continuous learning and intuition are developed and deployed to ensure high project performance and participant satisfaction Simultaneously, to cope with managing such uncertainty, project managers should find the balance between planning and learning Planning provides discipline and a concrete set of activities while learning permits adapting to unforeseen or chaotic events (Meyer et al., 2002) Figure 1.1 illustrates management style changes from different uncertainty types – variation, foreseen uncertainty, unforeseen uncertainty, and chaos The more uncertain the project environment, the more learning is As such, system dynamics methodology can facilitate this learning via emphasizing on systems thinking and systems simulation More importantly, system dynamics model helps overcome many limitations of our mental

models (Sterman, 1992 and Sterman, 2000), especially in complex and fast-changing project environments

Figure 1.1 Management style vs uncertainty profile (Source: Modified from Meyer et al., 2002) 1.2 Problem Statement

A typical construction project can be described as a complex dynamic system of large number of interrelated and interconnected elements, various organizational units and a wide variety of workforce (Ogunlana et al., 2002) In addition, the crucial difficulty with such project is that the requirements is not fixed, and uncertainty and changes in some requirements mean that interfacing elements also need to change, and hence incurring cross-impacts, rework, feedback loops leads to increasing such complexity (Williams, 1999) Construction projects often seem to be going smoothly until near the end, which errors incurred earlier are discovered, requiring costly rework, expediting, overtime, hiring, schedule slippage, or reduction in scope of work or quality (Sterman, 1992) More seriously, project management is simultaneously one of the most important and most poorly understood domains of management and then delays, cost overruns and quality problems are the norms rather than the exception in construction and other areas (Sterman,

1992, and Sterman, 2000) Also, project management suffers numerous problems of costing and scheduling (Sterman, 1992) All of these problems result in poor profitability, loss of market share, higher costs, and, all very often, divisive and costly litigation between owners and contractors over responsibility for overruns and delays (Sterman, 1992)

To manage a large-scaled construction project efficiently and effectively, the project team increasingly demands to create and/or deploy various systematic and analytical techniques

 Learning and response

HOW MANAGEMENT STYLE CHANGES

 Planning and anticipation

 Exchange of structured information along defined interfaces

 Exchange of unstructured information along emerging interfaces

 Fulfillment of targets  Flexible use of methods to

reap upside rewards

 Tracking progress  Tracking assumptions and

unknowns

PROJECTS vs DEGREES OF UNCERTAINTY

IMPROVING SYSTEMS THINKING FACILITATES LEARNING

and tools for both strategic and operational levels In responding to such requirements, many techniques and tools as well as different project management styles have been introduced, developed and implemented for more than 40 years To some extent, Work Breakdown Structure (WBS) and network-based techniques (e.g CPM, PERT) have been playing vital roles in many areas of construction project management, from planning to monitoring and controlling “The larger, the more complex the project, the more galaxies

of project variables will strain project management Redressing of such galaxies into hierarchically trees will then be necessary for project management to remain on top WBS

is a prominent tool for project integration” (Lamers, 2002)

Unfortunately, such techniques and tools have exposed their deficiencies, especially in strategic decision-making domain These traditional approaches are prone to assume that if each element of the project can be understood then that the whole project may be controlled (Rodrigues and Bowers, 1996) Furthermore, the traditional scheduling techniques (PERT, CPM) evolved from a need for the systematic evaluation of the relationship between activities and events (Partington, 1996) Senge (1990) (cited in Partington, 1996), however, stated that “fixation on events” is one of the „seven learning disabilities‟ that prevent the existence of learning

Better construction project management requires proactive and appropriate policies to be identified and implemented properly in coping with ever-increasing project complexity and uncertainty in dynamic and ever-changing environment nowadays System dynamics approach must be a good option, which helps the project manager and project team possess

a holistic overview whereas the traditional approaches are inadequate Additionally, experiments and then learning via systems simulation may be one of the most critical powers of system dynamics Thus, alternative policies can be evaluated to select the bests for managing construction projects successfully Finally, system dynamics methodology is employed as an analytical tool to complement for the traditional project management methodologies at a strategic level

Vietnam is currently among countries having the highest gross domestic product (GDP) growth rate Thus, construction investment in Vietnam has been increasing to meet social needs Management of construction projects in Vietnam, however, has faced various difficulties due to many controllable and uncontrollable causes Gaining insight into such

difficulties and respective solutions is obviously necessary

1.3 Objectives

The major objectives of the study are:

a) To identify problems and success factors pertinent to large-scaled construction projects

in Vietnam

b) To develop an applicable simulation model for a large-scaled construction project, using Hai Van Pass Tunnel Project (Vietnam) as a case study so as to:

 Capture dynamic problems in the project

 Propose practical policies to improve project performance

1.4 Scope of Work

This study aims to define problems of and success factors for large construction projects in Vietnam Building, industrial and road construction projects are focused The research

defines that large construction projects are projects with an estimated total budget of greater than $1 million Especially, the study also aims to construct a comprehensively analytical tool in order to improve a road tunnel project performance As mentioned earlier, this tool should be complementary, instead of replacement, for traditional techniques and tools such as WBS, network-based techniques (e.g CPM, PERT), and Trade-off Analysis

1.5 Expected Contribution

Since system dynamics has be considered as a useful tool for systems thinking, which is very essential in dynamic and uncertain environments Applications of system dynamics into construction project management system are promising in order to handle construction projects successfully The model developed in this research is expected as one of facilitators for top management in construction organizations, project managers, and project team in their decision-making domain, where traditional management methodologies are inadequate Besides, an investigation of problems and success factors pertinent to construction projects is supposed to be a comprehensive guidance for concerned project parties in Vietnam construction business environment

CHAPTER 2 LITERATURE REVIEW 2.1 General

This literature review consists of various aspects regarding to construction industry, construction management, and applications of system dynamics, especially in project management Critical characteristics and other factors of construction industry as well as construction project management will be briefly identified so as to explain why system dynamics must be needed in construction management while traditional approaches have exposed inadequately Moreover, previous studies and their applications of system dynamics will be outlined to advocate system dynamics as a promising approach for managing construction projects at strategic level

2.2 Characteristics of Construction Industry

It has been known that the construction industry is usually characterized by its complexities, reluctance to change and resistance to innovations (Oglesby et al., 1989, and Palaneeswaran and Kumaraswamy, 2000) Construction is inherently risky; its projects are generally unique and prototype (Wantanakorn et al., 1999 and Kale and Arditi, 1999) Oglesby et al (1989) pointed out some constraints peculiar to construction They include: a) Construction operates differently from other industries Most construction projects are unique, fast-moving

b) The contractual structure is seldom conductive to cooperation among participants c) Traditional hierarchical management structure within each organization blocks free discussion and exchanges of ideas

d) The usual attitude of construction people is to get on with the job

In addition, Kale and Arditi (1999) summarized several unique characteristics of the construction industry: (1) fragmented industry structure, (2) fragmented organization of the construction process, (3) easy entry to the construction business, (4) post-demand production, (5) uniqueness of projects, (6) high uncertainty and risk involved, (7) high capital required for constructed facilities, and (8) temporary nature of the relationships between parties They all must hamper new philosophies for performance improvements in construction organizations It therefore requires more commitment of time, effort, talent and money (Oglesby et al., 1989)

2.3 Construction Management

Practices and philosophies such as benchmarking, concurrent engineering, driven, supply-chain management, integrated information systems, integrated performance measurement, just-in-time, lean production, reengineering, total quality management, and six sigma have played vital roles in manufacturing and business sector (Palaneeswaran and Kumaraswamy, 2000) To some extent, they have been also applied in construction at all levels: business, project or construction process However, as mentioned earlier, due to the unique characteristics of construction industry, construction management must bear its own particular traits to cope with performance requirements and improvements

customer-2.3.1 Important Elements of Construction Management

Wang (1987) (cited in Chang, 1990) proposed important elements of construction management They consist of (1) planning, (2) organizing, (3) motivating, (4) directing, (5) communicating, (6) controlling, (8) coordinating and (9) forecasting Thanks to the elements, construction management transform input including materials, equipment, manpower and finance into the facilities in such a way as to meet stakeholders‟ satisfaction Therefore, project performance evaluation has often included the satisfaction

of parties involved as a criterion for measuring the success of a project (Ashley et al., 1987; Pinto and Slevin, 1988, cited in Liu, 1999)

Figure 2.1 illustrates the Factor-Resource Model that provides a framework to explore the relationship between variability and performance (Thomas et al., 1990, Sanders and Thomas, 1991, cited in Thomas et al., 2002)

Figure 2.1 Factor-resource model

(Source: Thomas el al, 2002) Thomas et al (2002) identified in previous research that the availability of resources is a significant determinant of a good crew or craft performance or labor productivity and one

of the most common problems in construction is unable to deliver materials at the right time and at the right place Subsequently, workers slow their pace when an adequate supply of materials is suffered Equipment and tools must be indispensable resources, which should be maintained and utilized properly Information is another flow that needs

in an accurate, complete, and timely manner for decision-making (Charoenngam and Kazi,

1997, and Thomas et al., 2002)

2.3.2 Multiple Project Objectives and Their Trade-Off

It is known that the three commonly critical project objectives are time, cost, and quality in accordance with certain expectations Project managers use budgets to fulfil multiple objectives, e.g cost control, short duration, and high quality (Ford, 2002) It is obvious that

a construction project can be regarded as successful if the project is finished on time,

Conversion Technology (Work Method)

within budget, without any accidents (Chan and Kumaraswamy, 2002), in accordance with the specified quality level, and the participant satisfaction (Ashley et al., 1987, Liu, 1999 and, Chan and Kumaraswamy, 2002)

Unfortunately, there has been generally a trade-off between time and cost to complete a task – the less expensive the resources, the longer it takes (Burns et al 1996), and the less quality may be Increasingly, project managers today are under pressure to complete complex projects under conditions of uncertainty in less time, without sacrifice to cost and quality and without leaving concerned parties dissatisfied (Laufer et al., 1996) Moreover, the bounded rationality of human judgment means that the best-intentioned mental model analysis of a problem as complex as a large-scale construction projects cannot expect to account accurately for a multitude of interactions which jointly determine the project outcome (Sterman, 1992) More awesomely, under such pressure, management errors such

as in decision making are likely more occurrence (Wantankorn et al., 1999)

2.3.3 Characteristics of Traditional Approaches

A variety of techniques and tools have been developed to facilitate project management Construction projects have been recognized as uniqueness, complexity, and uncertainty (Oglesby et al 1989, Sterman, 1992, Gidado, 1996, Baccarini, 1996, Williams, 1999, Ford,

2002, and Meyer et al., 2002) To deal with such complexity, it is natural that a project can

be manageable if its project work is systematically decomposed into simple components or elements The process is known as Work Breakdown Structure (WBS) WBS has been proving as an effective tool in project management To manage project schedules and costs, Gantt chart and network-based techniques (e.g CPM and PERT) have also proven their indispensable roles in managing a project The logic of the project gives the basis for reconstructing the project from its components or elements and calculating the duration, cost and resource requirements of the whole project from those of its elements (Rodrigues and Bowers, 1996)

Breaking large project down into work packages known as WBS has been recognized one

of the most important tasks in project management (Clarke, 1999, Charoenngam and Sriprasert, 2001, and Lamers, 2002) The WBS is a graphical description of a project, detailing it in a level-by-level fashion down to the degree of detail needed for effective planning and control (Liu and Horowitz, 1989, cited in Barros et al., 2000) The Critical

Path Method (CPM) model (Wiest and Levy, 1977, cited in Barros et al., 2000) emphasizes the precedence relationships among activities These relationships form a directed acyclic graph, from where the earliest start date and the latest completion date of each activity can be inferred, based on strictly defined activity duration Additionally, the CPM model concentrates

on the coordination of sequential and parallel activities in control of performance backed up by information technology (Laufer et al., 1996) It has been recognized that CPM provides a practical tool for planning and controlling construction projects CPM techniques can help the overall project cost be reduced by using less expensive resources for non-critical activities without impacting the duration (Burns et al., 1996) That process is called resource leveling

To some extent, CPM has proved a relatively effective tool for claims, especially in extension

of time

Traditional approaches have been, however, recognized as inadequate methodologies in today ever-changing environments that are inherent in construction projects (Sterman, 1992, Laufer

et al., 1996, Rodrigues and Bowers, 1996, Williams, 1999, Love et al., 2002, Mayer et al.,

2002, and Ford, 2002) The traditional techniques have been described as linear or as “static and closed” (Rodrigues and Bowers, 1996) For example, Mayer et al (2002) stated that “a

Gantt chart is more a reflection of what happened last week, and what someone hopes will happen next week” Some of assumptions in traditional analysis of projects are likely causes for such inadequacy Rodrigues and Bowers, 1996 listed those assumptions: (1) progresses of project are well defined, predictable stages in completions, (2) all the information is available

at the start of the project Saad (2002) commented that the common techniques of network planning (CPM, PERT…) attend to minimize project duration by assuming no limit on the availability of the resources to be employed to complete all activities on schedule This unrealistic assumption potentially leads to ineffective resource usage and project delays (Saad, 2002) Love et al (2002) added that traditional control mechanisms (such as WBS, Gantt Charts, PERT/CPM networks, Trade-off Analysis, etc.) are not entirely sufficient for managing complex projects Moreover, traditional methods capture only „hard‟ quantitative data while

„soft‟ ones cannot be captured (Williams, 1999)

With many problems mentioned above, evolving project management style is axiomatic Laufer et al (1996) indicated four distinct generations of the styles, each combining the principles of the preceding one: scheduling (control), teamwork (integration), reducing uncertainty (flexibility), and simultaneous management (dynamism) Table 2.1 displays such evolution of project management styles

Table 2.1 Evolution of models of project management

Central

concept

Era of model

Project characteristics

Process facilitation, definition

of roles Reducing

uncertainty

1980s Complex,

uncertain

Making stable decisions

Search for information, selective redundancy

Experience, responsiveness and adaptability

(Source: Adapted from Laufer et al., 1996) Furthermore, Mayer et al (2002) recognized that managers move progressively from traditional approaches that are based on a fixed sequence of tasks to approaches that allow for the vision to change, even in the middle of the project and learn to adapt to unforeseen

or chaotic events

2.3.4 Factors Affecting Project Performance

Time, cost, quality and participant satisfaction are usually main criteria for project success (Dissanayaka and Kumaraswamy, 1999) The three parameters needed to be defined and developed in specific theoretical context to explain the performance evaluation process are project performance, project success and participation satisfaction (Liu, 1999) Many previous studies (Morris and Hough, 1987, Ashley et al., 1987, Pinto and Slevin, 1987 and Belassi and Tukel, 1996) have identified and analyzed critical success factors (CSFs) in construction project management Belassi and Tukel (1996) proposed a framework for determining critical success/failure factors in projects (Figure 2.2) They divided into four areas for those factors related to: the project, the project manager and the team members, the organization, and the external environment

Furthermore, Liu (1999) pointed out that project success often depends on: (1) the right policy environment, and (2) the changes in policy forced on a project by an unfavorable external environment He also pointed out that a set of CSFs may not be able to transfer to another project due to the differences in environmental variables, the type and complexity

of the project, the type and complexity of the related parties‟ organization, and the prioritization of project goals Thus, using CSFs for better project management is likely ineffective and problematic, especially in dynamic and uncertain environments

Figure 2.2 Framework for determining Critical Success/Failure factors

(Source: Adapted from Belassi and Tukel, 1996)

2.3.5 Causes and Costs of Rework

Rework has become an inherent feature of the procurement process in construction that inevitably leads to time and cost overruns in projects (Josephson et al., 2002) Love et al (1998) (cited in Josephson et al., 2002) defined that rework cost is the total cost derived from problems occurring before and after a product or service is delivered In a recent case study of two six-story residential apartment blocks in Australia (Love et al., 2002), overall cost of variation and rework was found as equal as 10.5% of original contract value in which overall rework cost was 3.15% (Love and Li, 2000) In addition, poor motivation levels of architects and engineers were the cause for 50% of the rework costs in the case study project Hence, in order to improve the project performance, it is essential to identify the causes and costs of construction rework (Love and Li, 2000 and, Josephson et al., 2002) It is obvious that causes and their impacts on project costs are very different from this project to the next However, identifying commonly main causes is needed as

- Size & value

- Uniqueness of project activities

- Life cycle

- Urgency

Organization

- Top management support

- Project organizational structure

- Functional managers‟ support

- Project champion

Client consultation & acceptance

Project Manager’ s performance

on job

Effective:

- Planning & scheduling

- Coordination & communication

- Use of managerial skills

- Monitoring & control

- Use of technology

Project preliminary estimates

Availability of resources (Human, financial, raw material &

facilities)

External environment

guidelines for the future projects Figure 2.3 illustrates the significant causes based on reviewing previous studies (Love and Li, 2000, Love et al., 2002, and Josephson et al., 2002)

Figure 2.3 Fishbone diagram for identifying causes of rework

2.3.6 Problems of Construction Projects

The major problems that contractors handling projects in developing countries face have been classified as: (1) problems imposed by the industry‟s infrastructure, (2) problems of inaccurate information and frequent changes in instructions and failure to meet obligations

on the part of clients and consultants, and (3) problems imposed by their on shortcomings (Ogunlana and Olomolaiye, 1989) Research into the delays experienced in high-rise building construction projects in Thailand (Ogunlana et al., 1996) supported this classification More explicitly, Ogunlana et al (1996) confirmed that construction industry problems in developing economies could be nested in three layers: problems of shortages

or inadequacies in industry infrastructure, problems caused by clients and consultants, and problems caused by contractor‟s incompetence/ inadequacies Thus, if the problems are not solved swiftly, they can cause delays and cost overruns in projects, harm cooperative relationships, reduce efficiency, lead to claims and disputes, and probably invoke litigation proceedings (Cheung et al, 2000)

Causes and factors pertinent to time and cost overruns and other project performance indicators in previous studies are generally termed “adverse factors” in this research A set

of the adverse factors is a comprehensive foothold for identifying common problems of large construction projects By probing underlying relationships, the adverse factors can be classified into groups of problems responsible by: financiers, owners, contractors, consultants (including designers), project attributes, coordination and environment The problems or adverse factors are listed in the respective groups as follows:

Financier-related problems are problems in the financing domain High cost of financing

and difficulty in getting loans are adverse factors identified in previous work (Baldwin and

Faulty material handling

Erroneous workmanship

Insufficient cleaning Faulty manufacturing

Material hard to work with Late deliveries

Machine breakdown

Manufacturing defects of machines Machine delivered

Wrong setting up

Delivery with wrong type

Manthei, 1971 and Arditi et al., 1985) Furthermore, interference in owner‟s decisions and funding shortage are other problems attributed to financiers

Owner-related problems are the problems for which clients or employers are responsible

Finance and payments for completed work, excessive change orders, slow owner‟s decision-making process, owner interference, and ill-defined duties and responsibilities are adverse factors frequently cited in previous studies

Contractor-related problems concern problems or adverse factors caused by contractors

They include inadequate experience, construction errors, poor site management and supervision, equipment failures or allocation problems, inadequate labor skills, site manager lacking authority, improper planning and scheduling, inaccurate estimation, and poor contract management

Consultant-related problems are problems or adverse factors attributable to

designers/consultants Preparation and approval of drawings, design errors, delays in work approval, uncompromising attitude are common problems for which consultants are held responsible in literature

Project attributes-related problems are problems or adverse factors that derive from the

characteristics of the project and/or are difficult to classify into other problem categories when the project delivery system and other project information are not taken into account For example, in some research (Odeh and Battaineh, 2002), an adverse factor, namely improper quality assurance/control was considered to be the purview of consultants Except for traditional procurement, this classification may be inappropriate in other project delivery systems In the other project delivery systems such as design and build, contractors have to establish systematic quality assurance/control or face the consequences

of low quality otherwise This research therefore considers improper quality assurance/control as a project attributes-related problem Unforeseen site conditions, confined site, problems with neighbors such as pollution, unrealistic imposed contract duration, inaccuracy of project information are other adverse factors in this category

Coordination-related problems are problems or adverse factors such as poor

communication, excessive use of subcontractors and nominated suppliers, excessive bureaucracy, fraudulent practices and kickbacks, misalignment of client‟s expectations, and jurisdictional disputes

Environmental problems are external problems or adverse factors They may be caused by

natural conditions, for example inclement weather or socioeconomic conditions such as material shortage or late delivery, labor shortage, price fluctuations In addition, problems

or adverse factors perceived to be responsible by external project stakeholders are also

grouped into this category as inconsistent policies and slow government permits

Obviously, the previous research revealed that most of the problems are human and management problems These problems or adverse factors, furthermore, are included in a list of problems that were investigated to ascertain the extent of their occurrence in and influence on large construction projects in Vietnam

2.3.7 Success Factors for Construction Projects

In a business context, a success factor is defined as any knowledge, skill, trait, motive, attitude, value or other personal characteristics that is essential to perform the job or role and that differentiates solid from superior performance Rockart (1979) defined critical success factors (CSFs) as those few key areas of activity in which favorable results are absolutely necessary for a particular manager to reach his or her goals Boynton and Zmud (1984) defined CSFs as those few things that must go well to ensure success for a manager and an organization, and therefore, they represent those managerial or enterprise areas that must be given special and continual attention to bring about high performance

Several research on factors affecting construction project success has proposed either general factors (Sanvido et al., 1992) or specific factors (Chua et al., 1999) In building construction, Sanvido et al (1992) found four CSFs: (1) a well-organized and cohesive facility team, (2) a series of contracts allowing to encourage the various specialists to behave as a team without conflicts and to allocate risk and reward correctly, (3) experience

in various aspects of similar facilities, and (4) timely, valuable optimization information from related parties in the planning and design phases Chua et al (1999) identified different sets of CSFs for different project objectives Undoubtedly, these CSFs contribute different facets of project success as well as measures of project success, yet their applications must be limited A possible reason is that the trade-off and/or optimization of the project objectives is perceived to be very complicated, the balance of these CSFs considering different project objectives is much more complicated or even impossible The distinction between “success criteria” and “success factors” is also important (Cooke-Davies, 2002) Success criteria are the measures by which success or failure of a project or business will be judged whereas success factors are those inputs to the management system that lead directly or indirectly to the success of the project or business In construction projects, Ashley (1986) (cited in de Wit, 1988) indicated seven success factors and six success criteria These success factors are planning effort (construction), planning effort (design), project manager goal commitment, project team motivation, project manager technical capabilities, scope and work definition and control systems The six success criteria are budget performance, schedule performance, client satisfaction, functionality,

contractor satisfaction and project manager/team satisfaction

2.4 Applications of System Dynamics in Management

2.4.1 The Roles of System Dynamics

Accelerating economic, technological, social and environmental change has been challenging managers and policy makers to learn at increasing rates, while simultaneously the complexity of the systems in which we live is growing (Sterman, 2000) Several scholars (Senge, 1990 and Sterman, 2000) warned that many problems have risen as unanticipated side effects of our own actions It has been inferred as policy resistance Senge (1990) stated that “today‟s problems come from yesterday‟s „solution‟” According

to Sterman (2000), effective decision-making and learning in a world of growing dynamic complexity require systems thinking Systems thinking is a paradigm, method, language and set of tools for constructing better mental models, simulating them more reliably, and communicating them more effectively

Fortunately, system dynamics has proved as an effective approach to facilitate systems thinking since it was introduced by Forrester (1961) Sterman (2000) indicated that system

dynamics is a perspective and set of conceptual tools that enable us to understand the structure and dynamics of complex system Thus, tools and methods for system dynamics modeling, a variety of successful applications as well as insights into the effective use of the tools with executives and organizations are all growing and expanding rapidly (Sterman, 2000 and Santos et al., 2001) More interestingly, system dynamics has been used to analyze various social, economic and environmental systems (Wolstenholme,

1990, cited in Rodrigues and Bowers, 1996) Nesan and Holt (cited in Love et al., 2002) pointed out that system dynamics model is able to incorporate technical, organizational, human and environmental factors

System dynamics has been applied practically and/or academically to various issues such

as corporate strategy (Sterman, 2000 and Liehr et al., 2001), performance measurement systems and evaluation (Santos et al., 2001 and Tang, 2001) and different aspects in project management (Richardson and Pugh, 1981, Chang, 1990, Ogunlana et al., 1995, Li

et al., 2000, Eden et al., 2000, Howick and Eden, 2001, Rodrigues and Williams, 1996a, Rodrigues and Williams, 1996b, Rodrigues, 2001, Chritamara et al., 2001, Ford, 2002 and Love et al., 2002) That is, system dynamics can be applied to any dynamic system, with any time and spatial scale (Sterman, 2000) Sterman (1992) reported that system dynamics has repeatedly been demonstrated to be an effective analytical tool in a wide variety of situations, both academic and practical, and is currently used by a number of corporations, including Fortune 500 firms in the worldwide

2.4.2 New Paradigms for Complex Projects

Williams (1999) realized that new paradigms have been needed to cope with the increasing complexity of projects In construction industry, Ogunlana et al (2002) stated that the rapid technology improvements have created an increasing number of construction projects of very complex nature Baccarini (1996) defined that project complexity is

ever-“consisting of many varied interrelated part‟s and can be operationalized in terms of differentiation and interdependency” Alternatively, project complexity can be defined as the measure of the difficulty of implementing a planned production workflow in relation to any one or a number of quantifiable managerial objectives (Gidado, 1996) Furthermore, the complexity consists of structural complexity and uncertainty (figure 2.4)

Figure 2.4 Project complexity

(Source: Williams, 1999) Meyer et al (2002) stated that uncertainty is an inevitable aspect of most projects, and even the most proficient managers have difficulty managing it They found out that failing to recognize different types of uncertainty and each of which requiring a different management approach have been root causes They also thought that an uncertainty profile comprises four uncertainty types: variation, foreseen uncertainty, unforeseen uncertainty, and chaos Therefore, project management must move progressively from traditional

Project complexity

Structural complexity

Uncertainty

Uncertainty in methods Uncertainty in goals Interdependence of elements Number of elements

approaches that are based on a fixed sequence of tasks to approaches that allow the vision

to change from variation to chaos (Mayer et al., 2002) In addition, Love et al (2002) indicated that the dynamics of construction projects are derived from two fundamental sources: planned activities and uncertainties Planned activities initiate the progress of construction works and their dynamics is called attended dynamics whereas unattended dynamics is seen as an emphasis on factors beyond the control of project

Moreover, Williams (1999) recommended that analysts must be able to model complex projects to facilitate the management function, especially in particular planning, forecasting, monitoring and control He also persuaded that system dynamics must be the most successful approach for top-down holistic models at a strategic overview In addition, flexibility and learning are keys to moving projects beyond the vague assumptions characteristic of unforeseen uncertainty and chaos (Meyer et al., 2002)

2.4.3 System Dynamics in Project Management

The traditional project management methodologies such as network-based techniques have provided useful support (Rodrigues and Bowers, 1996), powerful and credible tools for assessing the effect of a small number of discrete impacts on a project (Williams, 2001) Network models based on the WBS have been used widely to support planning and monitoring (Rodrigues and Williams, 1996a) However, the inadequacies of traditional project management techniques have been highlighted for most projects failing to meet their objectives (Rodrigues and Williams, 1996a) Therefore, system dynamics has been recognized to offer important benefits to analysis of project management by providing a strategic overview

Rodrigues and Bowers (1996) listed various factors to have motivated the applications of system dynamics to project management They include: (1) a concern to consider the whole project rather than a sum of individual elements, (2) the need to examine major non-linear aspects typically described by balancing or reinforcing feedback loops, (3) a need for

a flexible project model offering a laboratory for experiments with management‟s options, and (4) the failure of traditional analytic tools to solve all project management problems and the desire to experiment with something new

Applications of system dynamics into project management have been found since 1964 in research and development (R&D) project (Rodrigues and Bowers, 1996 and Rodrigues and Williams, 1996b) Rodrigues and Bowers (1996) pointed out that three problem areas and the interactions among them that have been commonplace in most projects are control, rework, and human resource management They also proposed three cycles for different problem areas In the project control circle, system dynamics provides a language for expressing influences by many disruptive factors and giving numerical estimates of their effects In the rework cycle, system dynamics provides a deeper understanding of a background while it is able to include many important factors explicitly

Many scholars (Sterman, 1992, Rodrigues and Bowers, 1996, Rodrigues and Williams, 1996a, Rodrigues and Williams, 1996b, and Williams, 2001) agreed that system dynamics approach should be considered as complementary and/or addition to traditional project management techniques and tools Thus, comparisons as well as incorporation between the two must be necessary Rodrigues and Bowers (1996) compared the characteristics of traditional approach with those of system dynamics as illustrated in Table 2.2 They also clarified that, in system dynamics approach, the project work represented by a continuous

flow of units of works is modeled at a higher level and holistic view is adopted In addition, the units of work change from the initial state to final state, as the staffs allocated

to the project do the work What work is done, when and by whom are not specifically taken into consideration (Rodrigues and Bowers, 1996)

Table 2.2 Characteristics of the traditional and SD approaches

Focus Project work and “problem” Feedback processes and the “situation” Level of detail Considerable detail in some

areas but ignores others

Little detail but attempts to capture the whole project

reflecting a desired outcome

A simulation of reality, including human and system frailties, indicating likely outcome

(Source: Rodrigues and Bowers, 1996)

Besides, as mentioned earlier, the integrated view between the traditional and system dynamics approaches should be needed to clarify the roles of system dynamics in project management Rodrigues and Williams (1996a) developed a formal integrated framework named SYDPIM in software project management to specify how system dynamics models are to be used within the traditional management process and how they exchange information with the traditional models This framework comprises two main methods: the model development method to support the development of the valid system dynamics models for a specific project and the project management method to support the use of this model embedded within the traditional project management framework (Rodrigues, 2001) Figure 2.5 shows an overview of process logic of the framework According to Rodrigues (2001), the framework places the use of a system dynamics project model at the core of this process, enhancing both planning and monitoring and thereby the general project control in which the flows within the traditional project control process are illustrated by the thicker arrows

Figure 2.5 Overview of the SYDPIM process logic

Steady Behavior

Past Behavior Metrics

Report

Progress Report

Enhanced Planning Enhanced

Monitoring

1

5 6

2.4.4 System Dynamics in Construction Management

Construction projects are dynamic and relatively unstable in both their internal and external environments (Sterman, 1992 and, Love, et al., 2002) Sterman (1992) indicated that large-scale construction projects belong to the class of complex dynamic systems The systems are:

a) Extremely complex, consisting of multiple interdependent components: Interdependencies confound analysis beyond our mental models since a change in certain part of the system may have implications in other parts Thus, system dynamics models are very suitable to representing such multiple interdependencies

b) Highly dynamic: There are various time delays in carrying out programs, in discovering and correcting errors, and in responding to changes in scope or specifications System dynamics has the most highly evolved guidelines for proper representations, analysis, and explanation of the dynamics of complex systems such as construction projects

c) Involving multiple feedback processes: “Feedback refers to the correcting or reinforcing side effects of decisions” A construction project with its tightly coupled systems contains a numbers of important feedback relationships System dynamics is a choice for such significant feedback processes

self-d) Involving nonlinear relationships: They are the norm in complex systems such as construction projects Hence system dynamics models demonstrate the wide range of nonlinear relationships

e) Involving both “hard” and “soft” data: Large scale construction projects can not be understood solely in terms of technical relations among components Many significant data needed to understand such complex systems involve managerial decision-making and other known as “soft” variables Also, system dynamics can use multiple sources

of information as well as other important managerial dimensions of the system

Since large-scale construction projects have those characteristics as previously discussed, system dynamics approach must be employed to manage such complexity properly It is therefore recognized that system dynamics has been applied practically and academically

in several aspects of the construction project management although both its applications and success in the field have been limited and fragmented System dynamics that has been applied in construction project management can be preliminarily classified as follows:

Impacts of the owner decisions on project performance: An owner can make decisions

concerning to his/her project over time during its life cycle The influence of the owner behavior is complex and subjective and its impacts on project performance are problematic (Rodrigues and Williams, 1996b) Whether the owner will ask for changes, introduce delays, impose schedule restrictions (Rodrigues and Williams, 1996b), or establish different levels of initial scope (Chritamara et al., 2001), project outcome will be affected crucially In design and build (D/B) projects, Chritamara et al., (2001) studied the effects

of different levels of initial scope establishment on project performance by system dynamics modeling They found that increasing the level of initial scope leads to shortening project duration and project cost is likely to be optimized at the 50-70 percents level of initial scope

Claim management: As a contractor in today business, the ability to deliver a project

rapidly is becoming an increasingly important factor in winning a bid (Williams, 2001) Thus, contractors seem to ensure that their claims have to be suitable for delays caused by their owners, otherwise liquidated damages suffered Critical Path Method (CPM) has played a vital role in preparation for such claims Arditi and Patel (1989) (cited in Williams, 2001) used five CPMs: as-planned schedule, as-built schedule, owner-accountable schedule, adjusted schedule and as-projected schedule However, there are a number of crucial shortcomings of CPM analysis in evaluating disruption and delay (D&D) (Eden et al., 2000, Howick and Eden, 2001, and Williams, 2001) System dynamics models have thus been developed to explicate the factors and effects and then to provide transparency in the claims process (Williams, 2001) In practice, applications of system dynamics in claim management can be exemplified such as shipbuilding project for US Navy in 1970 (Sterman, 2000) and Channel Tunnel (Williams, 1999 and Eden et al., 2000)

Risk management: The rapidly changing environments and project complexity have

increased risk exposure (Rodrigues, 2001) It is known that a project risk is an uncertain factor – positive or negative Risk management is the practice of identifying, analyzing and controlling or responding those factors to avoid or mitigate potential negative effects (Meyer et al., 2002 and Love et al., 2002) The traditional risk management cannot deal with the extreme and increasing uncertainty in modern projects (Rodrigues, 2001 and Mayer et al., 2002) Mayer et al (2002) realized that managers need to go beyond traditional risk management, adopting roles and techniques oriented less toward planning and more toward flexibility and learning Since such project risk dynamics are very difficult to understand and control, Rodrigues (2001) recommended that system dynamics provide a useful alternative view It is, however, recognized that system dynamics in construction project risk management is still not much in comparison with other techniques, such as AHP (analytical hierarchy process)

Effects of changes and rework on project management: It is understood that this aspect has

been applying relatively considerably system dynamics modeling When unforeseeable and unexpected changes occur, the planning, organizing, motivating, directing, and controlling

of construction can become difficult and problematic (Love et al., 2002) Love et al (2002) used system dynamics to determine the relationship and consequences of a change Chang (1990) developed a model with a dynamic hypothesis centering on the notation of rework

to identify appropriate policies in order to improve the construction project management However, many assumptions imposed have made the model less realistic

Evaluation of the project delivery policies: Homer et al (1993) (cited in Sterman, 2000)

developed a system dynamics model of the entire engineering, procurement, and construction process so as to reduce significantly the total cycle time without increasing costs The model was used to analyze many policies and helped identify policies that reduced project delivery times by at least 30% within a few years (Sterman, 2000) In addition, Ford (2002) constructed a system dynamics model of contingency management to test hypothesis of the effectiveness of aggressive and passive management strategies on cost, timeliness, and facility value

As was briefly reviewed above, it is understood that system dynamics must be a promising approach in construction project management System dynamics has been realized as a proven approach to project management and hence construction project management is no

an exception Moreover, there must be many rooms for applying system dynamics in construction management, especially for large-scale and complex projects in dynamic and

uncertain environments

2.5 Summary

The literature has highlighted that construction industry bears its own several unique characteristics in comparison with other areas Such uniqueness has been challenging construction management at any levels, from corporate to process levels Many methodologies, techniques, tools, and initiatives as well as management styles have been created in the industry and/or adopted from manufacturing or business sectors to improve both organizational and project performance They have played vital and indispensable roles to deal with the challenge It has been, however, recognized that the traditional approaches have exposed their inadequacies in coping with complex dynamic systems such

as large-scale construction projects and uncertain and ever-changing today business environments To solve the problem, many methodologies must have continued being created, among them, system dynamics has been generating its significant role since it can help executives, managers and academics in systems thinking and continuous learning In the project management arena, system dynamics has been applied so as to improve project performance Nevertheless, its applications have been limited and fragmented, in construction, for example Further studies must be therefore needed to advocate and prove system dynamics as a promising approach in construction management as strategic and

holistic overviews

CHAPTER 3 METHODOLOGY 3.1 Introduction

This chapter proposes a methodology for this research The system dynamics approach is highlighted as a main philosophy throughout the research However, questionnaire survey

is another methodology adopted Principles and process for successful and effective use

of system dynamics modeling are specifically outlined in a fashion that can position into this research Furthermore, major elements of dynamic complexity are briefly presented

so as to understand the approach Finally, tools for the research are also stated in the last section

3.2 Principles of System Dynamics Modeling

The success of the modeling project depends on many factors like any other management initiatives that have been applied at any levels: corporate, process or project level Linard and McLucas (2001) identified four key reasons for the failure of the modeling project in practice They include: (1) misleading or erroneous assumptions, (2) lack of senior executive champion, (3) lack of problem definition, and (4) failure to achieve a shared vision Therefore, lessons learned have been always essential so as to ensure the modeling process beneficially as expected Sterman (2000) listed principles for successful use of system dynamics as follows:

1 Develop a model to solve a particular problem, not to model the system

2 Modeling should be integrated into a project from the beginning

3 Be skeptical about the value of modeling and force the „why do we need it” discussion at the start of the project

4 System dynamics does not stand alone; use other tools and methods as appropriate

5 Focus on implementation from the start of the project

6 Modeling works best as an iterative process of joint inquiry between client and consultant

7 Avoid black box modeling

8 Validation is a continuous process of the testing and building confidence in the model

9 Get a preliminary model working as soon as possible; add detail only as necessary

10 A broad model boundary is more important than a great deal of detail

11 Use expert modelers, not novices

12 Implementation does not end with a single project

3.3 System Dynamics Modeling Process

Modeling is always creative and iterative Modeling, as a part of the learning process, is interactive, a continuous process of formulating hypotheses, testing, and revision, of both formal and mental model (Sterman, 2000) Modeling process has commonly had five steps Coyle (1996) recommended the structure of the system dynamics approach including five stages It consists of problem recognition, problem understanding and system description, qualitative analysis, simulation modeling, and policy testing and design

This research has adopted the modeling process proposed by Sterman (2000) This disciplined process involves the following activities: (1) articulating the problem to be addressed, (2) formulating a dynamic hypothesis or theory about the cause of the problem, (3) formulating a simulation model to test the dynamic hypothesis, (4) testing the model until you are satisfied it is suitable for your purpose, and (5) designing and evaluating policies for improvement Figure 3.1 illustrates the steps of modeling process

in an iterative cycle Moreover, models go through constant iteration, continuous questioning, testing, and refinement (Sterman, 2000) Modeling is a feedback process instead of a linear sequence of steps (Sterman, 2000)

Figure 3.1 Iterative cycle of modeling process

(Source: Verified from Sterman, 2000) Basically, the two processes proposed by Coyle (1996) and Sterman (2000) are similar For example, stages 1, 2 and 5 in Coyle (1996) are analogous with steps 1, 2 and 5, respectively, in Sterman (2000) However, there are several differences between these modeling processes The most significant reason why the modeling process of Sterman (2000) adopted in this research is that it must be represented in a more explicit fashion in comparison with that of Coyle (1996) Moreover, Coyle (1996) seemed to integrate model formulation and testing into one stage – stage 4 It is better to divide it into two steps – model formulation and testing – to emphasize different importance of the two functions This was considered in the modeling process proposed by Sterman (2000)

3.4 Study Methodology

As the research objectives implied, the research is to investigate problems and success factors by questionnaire survey and to develop an applicable simulation model for a large-scaled construction project through system dynamics methodology Typical and

1 Problem Articulation (Boundary Selection)

Theme selection Key variables Time horizon Reference modes

3 Formulation

Specification of structure Estimation Tests

4 Testing

Comparision to reference modes Robustness Sensitivity

5 Policy Formulation

and Evaluation

Scenario specification Policy design

Sensitivity analysis Interactions

Initial hypothesis generation Endogenous focus Mapping

2 Dynamic Hypothesis

severe problems identified in the construction industry are addressed in the problem articulation of the dynamic simulation model In addition, success factors are considered

as good guidelines and implications to formulate practicable policies in policy analysis of system dynamics modeling in this research An interactive cycle of modeling process as shown in Figure 3.1 is followed in this research The flowchart of the research methodology is displayed in Figure 3.2 The following sections describe the methodology

as an integration of system dynamics modeling process and flowchart of the research methodology as shown in Figures 3.1 and 3.2, respectively

Figure 3.2 Flowchart of research methodology

3.4.1 Questionnaire Survey

A questionnaire was developed based upon literature and current construction practice in Vietnam Relevant previous research works were deeply reviewed to derive common problems of and success factors for construction projects To fit into Vietnam construction conditions, the preliminary questionnaire pilot tested Six experienced professionals in the Vietnam construction industry were involved in the pilot tests They were a city government officer in the Department of Construction, a public owner, a designer, a contractor, and two senior lecturers actively involved in industry practice Their comments were used to revise the research questionnaire The final questionnaire was distributed to professionals involved in large projects in Vietnam Several means were employed to deliver these questionnaires and to potential respondents However, the direct (face-to-face) delivery was preferred to motivate respondents and to ensure the accuracy of answers and improve response rate

Responses to the questionnaire were then collected and analyzed As shown in Figure 3.2, the analysis included ranking the problems in terms of degree of occurrence and level of influence as well as degree of significance of the success factors The analysis also examined whether or not perceptions of different respondents‟ groups affect the rankings

Problemsof Large Construction Projects

Success Factors for

Large Construction Projects

Development of a Construction Project Model

System dynamics modeling

Software: Vensim PLE

The Spearman‟s rank correlation coefficient was carried out to test how strong associations between the rankings of respondents‟ group Finally, factor analysis was used to derive interrelationships among the problems and success factors

3.4.2 System Dynamics Modeling

3.4.2.1 Problem Articulation

Saeed (2000) stated that, in system dynamics approach, a problem must be defined as an internal behavioral tendency identified in a system and not a snap shot of existing conditions Thus, problem articulation is the most important step in modeling (Sterman, 2000) Construction projects suffering time and cost overruns and unsatisfactory quality have been commonplace in many countries Causes of the delays may be the owner‟s and/or contractor‟ s faults and uncontrollable factors such as natural calamities Moreover, it is inevitable that such delays can lead to cost overruns since the projects require overtime work and more staff The overtime work may be one of the major causes

of workers‟ burnout, low productivity and then low worker‟s morale This inherently incurs vicious circles

Establishing reference modes and explicitly setting the time horizons are the two of the most useful processes in defining problems (Sterman, 2000) A reference mode as a pattern of behaviors, unfolding overtime shows the problem dynamically That is, it

displays how the problem arose and how it might evolve in the future (Sterman, 2000)

Saeed (2000) pointed out that the development of a reference mode requires integration

of four abstract concepts, (1) delineation of a system boundary, (2) recognition of a fabric

of historical patterns within the defined system boundary, (3) recognition of past trend for policy related variables missing in historical data, and (4) projecting past trends into future to create a fabric of interrelated patterns that constitutes a reference mode The process entails sixteen steps built around four learning cycles as can be seen in Figure 3.3

The time horizon and key variables and parameters must be identified and defined in Hai Van Pass Tunnel Project The project data and structured interviews of management team were the major tasks in this research to derive reference modes Moreover, management team was expected to review these reference modes in order to capture sufficiently the dynamic problems of the project

3.4.2.2 Feedback Structures as Dynamic Hypotheses

A dynamic hypothesis is a working theory of how the problem arose in terms of the underlying feedback and stock and flow structure of the system (Sterman, 2000) Tools employed for the dynamic hypothesis of the research include model boundary diagrams, causal loop diagrams, and stock and flow maps Model boundary diagrams help identify different type of variables and concepts in a typical large-scale construction project They facilitate to build the causal loop diagrams and stock and flow maps hereafter Moreover, the variables and concepts identified in the problem articulation step can be classified into endogenous (internal), exogenous (external), and excluded groups

In addition, causal loop diagrams were employed as flexible and useful tools for diagramming the feedback structure of systems in any domain (Sterman, 2000) Furthermore, stock and flow diagrams were used to emphasize the underlying physical

structure Also, policy structure diagrams were employed to show the information inputs

to govern the rates of flow in the system (Sterman, 2000)

Figure 3.3 Sixteen steps involved in the construction of reference mode

(Source: Saeed, 2000)

3.4.2.3 Formulating a Simulation Model

In order to conduct experiments in Hai Van Pass Tunnel Project, this step helps formulate

an elaborate model from the model boundary diagram, causal loop diagrams, and stock and flow maps,… in the earlier step This formal model for Hai Van Pass Tunnel Project was completed with equations, parameters, and initial conditions Finally, tests for consistency with the purpose and boundary must be necessary to identify flaws in the proposed formulations and to improve the understandings of the construction project management system

3.4.2.4 Model Testing

There are a variety of tests in this stage Part of testing is comparisons between the simulated behavior of the model and the actual behavior of the system (Sterman, 2000) The tests ensure every variable corresponding to the real construction project management practice The tests which may be used in this research include: (1) boundary adequacy, (2) structure assessment, (3) dimensional consistency, (4) parameter assessment, (5) extreme conditions, (6) integration error, (7) behavior reproduction, (8) behavior anomaly, (9) family member, (10) surprise behavior, (11) sensitivity analysis,

2 Decomposing complex patterns into simpler parts

6 Collecting pertinent multiple patterns for modeling

10 Inferring past behavior of related policy variables

14 Inferring future behavior of system and policy variables

4 Aggregating patterns and defining system boundary

8 Recognition of a fabric of past trends for system variables

12 Recognition of a fabric of past trends for system and policy variables

16 Recognition of a fabric of past and inferred future trends of all variables as a reference mode

1 Complex time series and multiple manifestations of problem behavior

15 Graphing inferred future trends

for system and policy variables

and (12) system improvement Further understandings of these kinds of tests refer to Sterman (2000)

3.4.2.5 Policy Design and Evaluation

This stage consists of scenario specification, policy design and sensitivity analysis According to Sterman (2000), scenario specification is created to indicate what environmental conditions might arise, particularly in the study, in a specific construction project environment Policy design was creating entirely new strategies, structures, and decision rules in construction project management practice and incorporating them in the model The robustness of policies and their sensitivity to uncertainties in model parameters and structure were assessed, including their performance under a wide range

of alternative scenarios (Sterman, 2000)

The research focused on construction stage among four project life cycle stages proposed

by PMI (2000) Therefore, five main areas of control in construction project during this stage are (1) schedule control, (2) cost control, (3) quality control, (4) scope change control, and (5) risk control To some extent, the research was taken into account these controls Management practice of Hai Van Pass Tunnel Project was observed and/or collected to derive realistic policies for the project performance improvement

3.5 The Major Elements of System Dynamics

Among the elements of dynamic complexity, the most problematic elements are feedback, time delays, stocks and flows (accumulations) and nonlinearity

3.5.1 Feedback

Much of the art of system dynamics modeling is discovering and representing the feedback processes, which, together with stock and flow structures, time delays, and nonlinearities, determine the dynamics of a system (Sterman, 2000) One cause of policy resistance is the tendency to interpret experience as a series of events Such event-level explanations can be extended indefinitely The event-oriented, open-loop worldview leads to an event-oriented, reactionary approach to problem solving (Sterman, 2002)

“Real systems react to our interventions There is a feedback” Without an understanding

of the feedback process, the world is likely to be considered as unpredictable and unpleasant, and all people can do is react to events (Sterman, 2002)

It has been recognized that all dynamics arise from the interaction of just two types of feedback loops, positive (or self-reinforcing) and negative (or self-correcting) loops Positive loops tend to reinforce or amplify whatever is happening in the system, while negative loops counteract and oppose change (Sterman, 2000) Figure 3.4 illustrates both positive and negative loops in a causal loop diagram

Figure 3.4 Positive feedback versus negative feedback

(Source: Sterman, 2000)

Sales from Word

of Mouth Customer Base Customer Lossrate

In Figure 3.4, the arrows indicate the causal relationships The + signs at the arrowheads indicate the effect is positively related to the cause By contrast, the – signs indicate the effect is negatively related to the cause The loop is self-reinforcing, hence the loop polarity identifier R, while B in the center of a loop denotes a balancing feedback (or negative feedback)

3.5.2 Time Delays

Time delays between decisions and its effects on the state of the system are common and troublesome (Sterman, 2000) Delays also create instability in dynamic systems More subtly, delays reduce our ability to accumulate experience, test hypotheses, and learn (Sterman, 2002)

3.5.3 Stocks and Flows

Causal loop diagrams are wonderfully useful in many situations However, the crux of the problem with causal loop diagrams is that it make no distinction between information links and rate-to-level links (Richardson, 1986) Sterman (2000) stated that one of the most important limitations of causal loop diagrams is their inability to capture the stock and flow structure of systems Stocks and flows, together with feedback are the two central concepts of dynamic systems theory Flows can be divided into inflows and outflows Figure 3.5 displays general structure of stocks and flows

3.6 Tools for the Research

There were many tools that can be utilized in this research In system dynamics modeling, they included text and graph, causal loop diagrams, stock and flow maps, policy structure diagrams, output graphs and tables Especially, system dynamics software adopted in the research was Vensim PLE from Ventana Systems, Inc The software has been free-downloaded from the product web site (http://www.vensim.com) Vensim is used for constructing models of business, scientific, environmental, and social systems In the research, the software helps develop and/or implement causal loop diagrams, stock and flow maps, an elaborate model, testing and simulation Additionally, SPSS (Statistical Package for the Social Sciences) software was used to analyze the questionnaire‟s responses

Stock

CHAPTER 4 PROBLEMS AND SUCCESS FACTORS 4.1 Introduction

Although much research has been undertaken on factors influencing delays, cost overruns, productivity and safety performance, etc and problems in specific kinds of projects, this research seldom discusses general problems of construction projects Thus, comprehensive studies on these problems are essential Since the problems relatively depend on regional traits, these studies need to focus on a specific area, country or region This chapter presents problems of and success factors for large construction projects in Vietnam There are various perspectives regarding what and how a large construction project is In this research, the definition of large construction projects is considered in Vietnam and perhaps other developing countries That is, construction projects with an estimated total budget of greater than $1 million are considered large projects The respondent’s characteristics, general achievement of project objectives, problems and success factors are major parts Data analysis was carried out based on rankings of problems experienced and success factors followed by factor analysis to investigate underlying relationships among problems

as well as success factors

4.2 Respondent’s Characteristics

A set of questionnaire (Appendix A) was designed for all kinds of respondents The distribution of questionnaires was considered suitable proportions of concerned parties where the respondents were working The parties were classified into owners, designers/consultants and contractors/subcontractors It was difficult to separate construction managers from designers/consultants (here after called consultants or designers) in Vietnam since there were no specialized construction management firms The questionnaire rates of return from different respondents’ groups are displayed in Table 4.1 The respondents from owners, consultants and contractors were 33.0%, 24.8% and 42.2%, respectively These proportions can be acceptable

Table 4.1 Questionnaire return rate

questionnaires sent

Number of responses received

Response rate (%)

Proportion (%)

Table 4.2 Respondents’ position

Functional/project manager 31 28.4

Line manager/engineer 62 56.9

Table 4.3 Construction experience

Less than or equal to 5 years 42 38.5 Between 5 and 10 years 46 42.2

Table 4.4 Construction sector

Table 4.5 Project types

Project type Frequency Percent

industrial and road construction

4.3 Achievements of Project Objectives

To elicit the extent of the occurrence of unexpected status on project objectives, the survey respondents were asked to rate against the five-point scale, from “not significant” (1) to

“extremely significant” (5) This numerical equivalence scale was also applied for questions regarding problems and success factors as discussed in the following sections Table 4.6 displays the degree of occurrence of the five problems experienced in construction projects Clearly, project delays and cost overruns were rated as very high occurrence The considerable occurrence of labor accidents, low quality and disputes were also perceived It can be explained that large projects are being poorly managed in Vietnam Effective construction management at corporate, process, project, and activity levels… should be introduced to professionals to enhance construction industry

performance in Vietnam

Table 4.6 Failure to meet project objectives

Rank Unexpected project performance Mean Standard deviation

4.4 Problems of Large Construction Projects

From literature and the specific conditions of Vietnam construction industry, 62 problems were identified They were subjectively divided into organizational, project attribute-related, coordination-related and environmental problems Organizational problems were also subdivided into problems that were responsibilities of financier, owner, designer/consultant and contractor The following are discussions about degrees of occurrence and influence of the problems in large construction projects in Vietnam

4.4.1 Rankings of Problems’ Occurrence and Influence

The full ranking regarding the degree of occurrence and degree of influence of the two problems rated by respondents from different parties is available in Appendices C and

sixty-D, respectively Tables 4.7 and 4.8 show the top twenty problems perceived as high degrees of occurrence and influence by the whole respondents

Table 4.7 Problems as high occurrence

deviation

1 Inaccurate time estimating 3.64 1.04

2 Slow site clearance 3.47 1.43

3 Many change orders 3.42 1.15

4 Slow government permits 3.38 1.17

12 Inaccurate cost estimating 3.17 1.13

13 Improper planning and

17 Poor site management 3.02 1.13

18 Too many contractors

Table 4.8 Problems as high influence

deviation

1 Slow site clearance 3.77 1.40

2 Slow government permits 3.63 1.21

3 Inaccurate time estimating 3.56 1.07

4 Lack of capable owner representatives

3.54 1.28

5 Contractor's financial difficulties

3.28 1.29

17 Inadequate modern equipment

3.21 1.14

18 Unrealistic imposed contract duration

3.20 1.27

19 Lack of involvement through project life

3.15 1.31

20 Inadequate project management assistance

3.14 1.18

From the above rankings, many problems had high ranks of their degrees of both occurrence and influence Some of them were inaccurate time estimating, slow site clearance, slow government permits, lack of capable owner’s representatives, obsolete technology and unsatisfactory site compensation It can be said that these problems occurred under a wide range of causes: owners, designers, contractors, project attributes and environment However, several problems had high occurrence but low influence and vice versa Although severe overtime and inadequate modern equipment were identified as high occurrence, their influences were less Severe overtime is one of the common management policies to accelerate the project’s progress Some unintended side effects of severe overtime have been widely recognized in the industry, yet it does not usually have strong impact on project performance In Vietnam, since construction is still labor-intensive work, it is logical that inadequate modern equipment was not rated as very high impact Improper planning and scheduling and contractor’s financial difficulties were high ranks in degree of influence These have been agreed by previous research

To deeply investigate which parties and groups were responsible for these problems, problem categories were ranked in terms of their degree of occurrence and degree of influence as shown in Tables 4.9 and 4.10, respectively It was identified that consultants, contractors, and coordination were main parties/factors responsible for problems Regarding the degree of influence, problems associated with the consultants, owners and environmental factors highly impacted projects The owners and consultants obviously

have high influences on whether or not the projects are successful since they are involved

in the early stages of project life It is widely accepted in extensive research and practice

Table 4.9 Ranking of degree of occurrence of problem categories

Overall Owners Consultants Contractors Problem category Mean Rank Mean Rank Mean Rank Mean Rank

Table 4.10 Ranking of degree of influence of problem categories

Overall Owners Consultants Contractors Problem category Mean Rank Mean Rank Mean Rank Mean Rank

Interestingly, environmental problems were highly ranked This differs from previous research in other countries, such as Odeh and Battaineh (2002) They studied causes of construction delay in Jordan A possible explanation is that though price fluctuations, unforeseen ground conditions were not serious, yet slow government permits, unstable regulatory framework, slow site clearance and unsatisfactory site compensation were strongly influential problems These imply that legal framework for the construction industry has been problematic in Vietnam

4.4.2 Factor Analysis of Problems’ Occurrence

To capture any multivariate interrelationships existing among the problems in terms of degree of occurrence, factor analysis was applied Factor analysis addresses the problem of analyzing the structure of the correlations (interrelationships) among a large number of variables by defining a set of common underlying dimensions, known as factors or components (Hair et al., 1998) There are several tests required for the appropriateness of the factor analysis for the factor extraction They include anti-image correlation –