- Email: [email protected]
An integrated system for change management in construction
Automation in Construction 16 (2007) 368 – 377 www.elsevier.com/locate/autcon
An integrated system for change management in construction I.A. Motawa a,⁎, C.J. Anumba b , S. Lee c , F. Peña-Mora d a
c
Department of Structural Engineering, Faculty of Engineering, Mansoura University, Egypt b Department of Civil and Building Engineering, Loughborough University, UK Intelligent Engineering Systems Laboratory, Civil and Environmental Engineering Department, MIT, Cambridge, MA 02139, USA d Civil and Environmental Engineering Department, University of Illinois at Urbana-Champaign, IL 61801, USA Accepted 25 July 2006
Abstract Change management in construction is an important aspect of project management, as changes constitute a major cause of delay and disruption, and it is widely accepted by both owners and constructors that change effects are difficult to quantify and frequently lead to disputes. The development of change management systems should consider many elements of the project processes and address all internal and external factors that influence project changes. This paper presents an integrated change management system developed to represent the key decisions required to implement changes and to simulate the iterative cycles of concurrent design and construction resulting from unanticipated changes and their subsequent impacts. The system integrates a fuzzy logic-based change prediction model with the system dynamics model of the Dynamic Planning and control Methodology (DPM), which has been developed to evaluate the negative impacts of changes on construction performance. The developed system can be used in managing change scenarios on projects and also in evaluating change effects depending on the available information at the early stages of projects. © 2006 Elsevier B.V. All rights reserved. Keywords: Change management; Fuzzy logic; Dynamic planning
1. Introduction Changes in construction projects are common and likely to occur from different sources, by various causes, at any stage of a project, and may have considerable impacts. Based on time, change could be anticipated or emergent, proactive or reactive, or pre-fixity or post-fixity. Based on need, change could be elective or required, discretionary or non-discretionary, or preferential or regulatory. Based on effect, change could be beneficial, neutral or disruptive. Change management relates all the internal and external factors that influence project changes. It seeks to forecast possible changes; identify changes that have already occurred; plan preventive impacts; and coordinate changes across the entire project [1]. Inconsistent management of the change process can result in many disruptive effects.
⁎ Corresponding author. E-mail address: [email protected] (I.A. Motawa). 0926-5805/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.autcon.2006.07.005
Some of these consequences can be relatively easy to measure, while others are more difficult to quantify. Change management is considered an integral part of project management. This paper presents the development of an integrated system for improved change management. The following section reviews the work done on this topic. 1.1. Literature review Research on modelling the change process in construction has tended to focus on the identification of factors affecting the success of a change process, and resulted in guidance for best practice in change management. Examples of such guidance include: a concept for project change management [2], best practices for managing change efficiently [3], a generic procedure for issuing a change order request [4], an analysis method to reduce the overall rate of construction change orders [5], a best practice guide to present best practice recommendations for the effective management of change on projects [6], and an advanced project change management system [7].
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
Research has also been undertaken on evaluating the change effects on certain project elements. These studies dealt mainly with a single factor or a single project element such as: construction change order impacts on labour productivity at the craft level [8], effect of the size of change and its impact time on a project [9], a linear regression model that predicts the impact of change orders on labour productivity [10], the risk of changes to safety regulations and its effect on a project [11], and decision tree models to classify and quantify productivity losses caused by change order impacts [12]. Change management has been the focus of different IT systems. For example, an integrated environment for computeraided engineering was developed by Ahmed et al. [13], which is a blackboard representation that integrates a global database, several knowledge modules, and a control mechanism to systemize object changes. Peltonen et al. [14] proposed an engineering document management system for changes that incorporated document approval and release procedures. Spooner and Hardwick [15] developed a system with rules for coordinating concurrent changes and for identifying and resolving conflict modifications. Ganeshan et al. [16] developed a system to capture the history of the design process, initiate backtracking, and determine the decisions that might be affected when changes are made in the spatial design of residential buildings. Krishnamurthy and Law [17] presented an interesting change management model that supports multidisciplinary collaborative design environments. Another change management system was proposed by Mokhtar et al. [18] for managing design change in a collaborative environment. The model is capable of propagating design changes and tracking past changes. Soh and Wang [19] proposed a constraint methodology based on a parametric technique to coordinate design consistency between different geometric models and to facilitate managing design changes. Hegazy et al. [20] introduced an information model to facilitate design coordination and management of design changes. Important dependencies between building components were represented by this model to help identify the ripple effect of changes between components. Also, a reporting system was used to view the history of all changes made by all disciplines. A more generic ITsystem was presented by Karim and Adeli [21] which is an object-oriented (OO) information model for construction scheduling, cost optimization, and change order management. Charoenngam et al. [22] developed a Web-based change order management system that supports documentation practice, communication and integration between different team members in the change order workflow. The above literature on change management and evaluation mainly focused either on the identification of the change process, best practice recommendations for managing change during the project life cycle, or on the evaluation of the change effects on a single project parameter. Much of the discussion is presented in categorical ways with little attention being paid to modelling the dependent data or simulating the iterative cycles of concurrent design and construction that result from unanticipated changes and their subsequent impacts on project performance. Therefore, there is a clear need in the construction industry for research work to focus on modelling this depen-
369
dency, especially for multi-disciplinary causes and effects, and planning such iterative tasks. Some of the IT systems developed for change management are integrated systems that represent design information, record design rationale, facilitate design co-ordination and changes, and notify users of file changes. These systems were developed mainly to deal with reactive changes, particularly design changes. The research presented in this paper focuses on both proactive and reactive changes. Reactive changes represent the events when a change occurs and the project team starts to take actions to remedy the consequences of this change. While the proactive changes represent the events when a change is likely to occur in a later stage and the project team plans to minimize the disruptive effect of these changes. Because the effect of changes always concern practitioners, the concept of change effects in the proposed system is presented in the following section. 1.2. Change effect on project life cycle Studying the life cycle of changes and their impact on construction performance during actual execution shows that construction projects are inherently complex, dynamic, and involve multiple feedback processes [23]. The uncertainty and complexity of design and construction projects are usually driven by these feedback processes. There are only two types of feedback processes: reinforcing (that generates other changes or makes errors) and balancing (that resolves such changes and errors). Dynamics in a system arise from the interaction of these two types of feedback processes among the components of the system, not from the complexity of the components themselves [24]. Fig. 1 shows the basic simultaneity of the reinforcing and balancing feedback in design and construction projects. The control actions to address changes can have the intended effect of resolving the issues that initiate the control actions, if the decision is correct and well implemented. At the same time, they can produce a side effect that may create some unintended problems, if the decision is incorrect, not well implemented, exceeds the time frame of its effectiveness or if a project manager does not realize the impact of the control actions on other related activities. In addition, the reconciliation of the gap between the initial work scope and the actual work scope can also result in these feedback processes. After the project starts, the actual work scope may be increased, since additional work is often added to the project scope in order to deal with changes. Moreover, these unintended effects become more detrimental when concurrent engineering techniques are applied. This is because the decision to take control actions against unanticipated additional work has to be made within the complex inter-relationships of activities, even with a lack of complete information about predecessor activities [25]. Based on this concept, the research presented in this paper integrates the work done on the stage of “Change identification”, shown in Fig. 1, and the work done on simulating the uncertainty resulting from feedback processes caused by change. A generic change process model was developed to give full definition for the change over the time of its occurrence and to
370
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
Fig. 1. Multiple feedback processes caused by changes.
represent the key decisions required to implement changes. A simulation system was also developed to address the uncertainty resulting from changes. The Dynamic Planning and control Methodology (DPM) has been adopted to assist in the preparation of robust construction plans and to provide policy guidelines to handle changes [25,26]. The system provides improved co-ordination and control over changes, thus helping to increase the consistency and productivity of the overall project process. The following sections present a generic change process model and a fuzzy-based model for predicting the likelihood and impact of changes. 2. Generic change process model The developed model is based on a synthesis of the change process models reviewed in the literature and the process models adopted on a number of case studies undertaken during the research project. The process model, shown in Fig. 2, is a generic change process model that can be applied to different change categories, such as pre- or post-fixity changes. The process model has four main sections, as shown in Fig. 2. These are: Start up:
Identify and evaluate:
At the “Start Up” stage, the generic process defines a set of proactive requirements that are essential for effective change management. These requirements enable the project team to respond readily to change, to manage change effectively, and to facilitate contingency plans for any unanticipated change. Details of such requirements are presented elsewhere [27]. Change identification should cover change causes, types and effects. It
should also define the relevant project processes and departments affected by the change or involved in the change decision. The change management process model requires project teams to keep records of all relevant information on change cases to build a case base for future use. Analysis of change options is required for decision-making – whether to go ahead with any of the change options or to undertake further investigations. The criteria required to carry out this analysis may include tangible and intangible criteria. The evaluation steps include options evaluation, implications assessment and optimum selection of change options. Different models and decision support systems can be used to help decision-makers select an optimum solution and evaluate the change options using quantitative and qualitative criteria. Approval and propagation: Client approval is an important step in the process while different outputs are expected, as shown in Fig. 2. The client needs to review potential changes against the project baseline using tangible and intangible criteria. In many cases, clients need to use decision-making techniques for evaluation and comparison
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
Fig. 2. Generic change process model.
371
372
Post change:
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
in order to decide on a change option. The rest of the process stages in Fig. 2 involve integration between documentation and communication facilities. When dispute resolution is applicable, it requires the investigation of direct and indirect causes of change. In this situation, the analysis of the effect of multiple change causes could be prepared.
This process model has been studied further. The study paid much attention to the link between project implementation and the role of change. From this study the term ‘stability’ has been defined. Change depends on the stability of the given initial scope of an activity. Stability indicates the degree to which the given work scope would be performed without a request for change. High stability means that only a small number of changes would be expected during the execution of a particular activity, while low stability represents the possibility that a great number of changes would be requested. The internal iterative sub-processes entailed in the implementation of the generic change process, as shown by the bold links in Fig. 2, increase the effects of change on the project life cycle as denoted above for the feedback processes in Fig. 1. Identified changes are usually approved based on their feasibility in the project through the claim and change management process. For instance, if the identified change is perceived to require a significant investment or there are other options to replace it, it can be rejected. If a change is approved, the corresponding additional work scope is introduced to that particular activity. A rejected change may either become a permanently rejected change or it can be designated as a latent change in terms of its potential for reconsideration later in the process, as shown by the bold links in Fig. 2. Based on the above change concept, an IT system has been developed to help in managing the change process effectively. This paper presents two main functions of the system. The first aims to predict the level of stability, defined above, by predicting the likelihood of change occurrence. The second function aims to simulate the potential iterations that may occur during change implementation. The level of stability, determined by the first function, is among the basic data required to run the simulation system, developed to implement the second function. Fig. 3 illustrates the application logic of the system, which shows how the system components, presented in this paper, work together. The following sections describe how the system operates conceptually. 3. Change prediction system The main purpose of this component is to help in estimating the level of stability defined above. The estimation is based on investigating the information available at the early stages of projects and using such information to predict change events. Aggregating such events will help in defining the level of stability
Fig. 3. Application logic of the system.
for a project. This prediction will also help in taking appropriate actions to minimize the disruptive effects of changes. To develop this component, it was important to identify what project characteristics lead to change causes and what these causes are, and then to understand how these causes are related to effects. Some of the key questions were: How does one factor relate to another? What are the internal mechanisms by which a particular factor causes a change in another factor? Fig. 4 gives an example of how the change prediction system relates these factors to one another for a change case. Multiple causes can lead to a change case, and the effects of this change case cannot be added linearly (i.e. different causes of change may be responsible for a certain effect or a set of effects). In other words, the effect of a change cause C1 and C2, occurring together, may in general result in more or less impact than the sum of the effect of change cause C1 occurring on its own and the effect of change cause C2 occurring on its own. An analysis is required to find out the possibility of occurrence of each cause of change (Cm) with respect to each project characteristic (Fn). Multiple causes of change are a complex issue to determine the corresponding impacts (El) on projects. Lists of Fn, Cm, and El, were identified through case studies, details can be found elsewhere [28]. This helps project teams check their project against these lists. Fuzzy systems were found more suitable to describe the relationship between the above variables for the following reasons: • The information, if available, for the change cases is always vague. When little or no information is available, fuzzy systems are appropriate to represent the human perception on a specific issue based on a variety of assumptions which may be numerically or linguistically formatted. • Change consequences might lead to dispute, therefore, detailed analyses and investigation of the change causes and effects are required. This needs close observation for the simulation route of cause–effect relationships.
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
373
the values for “Fn” will be given a weight on this scale to represent its contribution, then the Rij will have another weight on a 0–10 scale to represent the sensitivity of the change cause to variations in the Fn and so for the Rjk. The system enables the user to estimate the level of stability of the studied project, i.e. the degree to which the given work scope would be performed with changes. Fig. 5 shows an example of the system output. 3.1. Simulating the iterative cycles caused by changes
Fig. 4. A dependency diagram for a change case.
Two different fuzzy approaches were investigated; namely building a fuzzy rule-based model to simulate the cause and effect relationships of a change event or establishing fuzzy relations between the prediction elements of change. The first approach is part of any general fuzzy logic system which, in addition to the fuzzy rule base, includes fuzzy inference process, fuzzification process, and defuzzification process. Developing a fuzzy rule-based system requires identification of different rules that link the prediction elements in IF–THEN format. The main difficulty in this approach is about the complexity of the problem in hand, as the list of prediction elements collected for the proposed model initially includes: 30-elements of project characteristics, 20-elements of change causes, and 29-elements of change effects. A further approximation may be used to simplify building such rule-based system. This is by assuming predefined fuzzy-sets to fire the IF–THEN rules. This is also a controversial solution and the justifications for such assumptions are always questioned. Therefore, the proposed system adopted the fuzzy relation approach considering the most appropriate methods for improving the model accuracy. This approach gives flexibility for the practitioners to express their belief about the problem with limited calculation assumptions. The relationship between the elements are formed and combined to produce the cumulative relation, as shown by Eq. (1): ! j¼m Xk¼l i¼n; X j¼m; f Fi ; Rij ; Rjk ¼ Fi 4Rij ο Rjk ð1Þ j; k¼1
Based on the level of stability obtained, the impact of iterative cycles caused by changes on construction performance is estimated by the Dynamic Planning and control Methodology (DPM) [26,29]. DPM aims to provide policy guidelines for unexpected events by supplementing network-based tools with mechanisms to represent the dynamics of a project. To identify the impact of iterative change cycles on construction performance, the DPM incorporates the system dynamics based design and construction process model [29], as seen in Fig. 6. It aims to address the iterative change cycles, their settlement through Requests For Information (RFI), the change decision process, and the corresponding work amount increased. Managing changes in design and construction usually have two major components, the Scope Management (SM) process and the Claim and Change Management (CCM) process. The SM process is the review process and the CCM process is the decision making process for change adoption o rejection. Specifically, before the execution of a task, the SM process is applied to WorkToDo (WTD) stock, as denoted by A in Fig. 7. The SM process aims to make sure that the given scope of work and the corresponding work setting are the same, as planned. Thus, if this WTD stock differs from the originally planned WTD stock, those tasks in WTD stock are sent to the stock of WorkAwaitingCCMGDecision (WACCMGD – B in Fig. 7), which needs the decision or analysis of Claim and Change Management (CCM) group. However, some potential changes may not be identified (i.e. latent changes) during the SM process. In this sense, Scope Management THoroughness (SMTH) is defined as the degree to
i; j¼1
where: n m l Rij Rjk
no. of the project characteristics no. of the change causes no. of the change effects the sensitivity of a change cause ( j) occurrence to variations in the project characteristics (i); the sensitivity of the change impact (k) to variations in the change causes ( j).
All units used to quantify the variables in Eq. (1) are based on a 0–10 scale. It is assumed that all variables have the value of “0” when no change occurs. In case of predicting a change case,
Fig. 5. Output of the change prediction system.
374 I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
Fig. 6. Design and construction process model [based on Pugh–Roberts Associates; Ford and Sterman, 1998].
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
375
Fig. 7. Modelling change management with system dynamics.
which the potential changes have been identified during the SM process. Based on SMTH, changes could be identified (i.e. identified change) or not (i.e. latent change). In the model, tasks that flow from WTD stock to WACCMGD stock are identical to identified changes (C in Fig. 7). Then, these identified changes can be approved (i.e. approved change) or rejected (i.e. rejected change), based on the decision of the CCM group. In the model, if a change is rejected, tasks that have been suspended would flow back to WTD and will be performed (D in Fig. 7). If a change is approved, tasks would also flow back to WTD,
however, in this case, with additional work generated by this approved change (E in Fig. 7). The detailed model structure is presented elsewhere [29]. An example of the simulation results generated by the DPM is presented in Fig. 8. The typical relationships between activities, such as the finish-to-start (FS) relationship between final design activity and the shop drawing submittal activity in Fig. 8, do not make the following construction activities start until the design activities are completed. While in concurrent design and construction processes, the design work may be
Fig. 8. Simulation results generated by the DPM.
376
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
forced to proceed without the establishment of a detailed work scope by the owner. Thus, requests for design clarification and/ or changes are often made by the contractor. Design omissions and errors may also be committed by the design group of the team. These are the main reasons for the delay. The simulation results exactly capture the aforesaid reality. For the studied example, a large coordinating work amount, caused by waiting for a RFI reply and the CCM decision, was observed (column A in Fig. 8). In addition, a significant work amount was newly introduced to deal with changes in the final design (column B in Fig. 8) and also contributed to delay. Furthermore, a large fraction of the changes was not identified and became latent changes (column B in Fig. 8). This situation also required much effort in coordinating with the predecessor activity through RFIs. Identifying such detrimental impacts caused by changes, DPM explicitly captures the negative impact of iterative change cycles. More results about the studied example can be found elsewhere [29]. 4. Discussion and conclusion An integrated change management system has been presented in this paper. It was developed to enable AEC professionals manage changes more effectively. The system covers the life cycle of changes within construction projects. A change process model was developed first to represent the key decision points required to implement changes. In support of this model, an integrated system was developed. Two components of the system were presented in the paper; a Dynamic Planning and control Methodology (DPM) and a change prediction system. DPM has been developed to overcome the uncertainties and complexities resulting from changes in concurrent design and construction by focusing on iterative cycles caused by changes and their impacts on construction performance. Among the main data required to simulate such iterative cycles, there is the level of project stability. The stability of a project can be predicted by studying the available information of the project at the early stages. Therefore, the change prediction system aimed to determine the likelihood of change occurrence, which is a measure of the project stability. The integrated system can help in: • providing robust planning and control actions, • identifying various dynamic impacts of construction feedbacks and iterative cycles, • planning projects with allowances for the potential changes, • exploring the cause-and-effect relationships of change events, and • examining the impact of changes on different project parameters. Although the current system has established the potential for proactive change management, there is a need for significant effort to ensure its realistic implementation. Thus, further simulation and experimentation are required to establish robust mathematical relationships between other variables of the change process. Additional case studies would be required in order to validate the effectiveness of the developed system. In this way, the
system could enhance its practical simulation capability and provide more reliable and realistic implementation. References [1] V. Voropajev, Change management – a key integrative function of PM in transition economies, Int. J. Proj. Manag. 16 (1) (1998) 15–19. [2] Construction Industry Institute (CII), Project Change Management Research Team, Special Publication, vol. 43-1, The University of Texas at Austin, US, 1994. [3] Construction Industry Institute (CII) conference, Project Change Management – Implementation Feedback Report, The University of Texas at Austin, US, 1996. [4] I.D. Cox, J.P. Morris, J.H. Rogerson, G.E. Jared, A quantitative study of post contract award design changes in construction, J. Constr. Manag. Econ. 17 (1999) 427–439. [5] S.N. Stocks, A. Singh, Studies on the impact of functional analysis concept design on reduction in change orders, J. Constr. Manag. Econ. 17 (1999) 251–267. [6] Construction Industry Research and Information Association (CIRIA), Managing Project Change – a Best Practice Guide, CIRIA, UK, 2001. [7] C.W. Ibbs, C.K. Wong, Y.H. Kwak, Project change management system, J. Manage. Eng. ASCE 17 (3) (2001) 159–165. [8] W.T. Hester, J.A. Kuprenas, T.C. Chang, Construction changes and change orders: their magnitude and impact, Construction Industry Institute (CII), Source Document, vol. 66, CII, Austin, Tex, 1991. [9] C.W. Ibbs, Quantitative impacts of project change: size issues, J. Constr. Eng. Manag., ASCE 123 (3) (1997) 308–311. [10] A.S. Hanna, J.S. Russell, P.J. Vandenberg, The impact of change orders on mechanical construction labour efficiency, J. Constr. Manag. Econ. 17 (1999) 721–730. [11] T.M. Williams, Safety regulation changes during projects: the use of system dynamics to quantify the effects of change, Int. J. Proj. Manag. 18 (1) (2000) 23–31. [12] M. Lee, A.S. Hanna, W. Loh, Decision tree approach to classify and quantify cumulative impact of change orders on productivity, J. Comput. Civ. Eng., ASCE 18 (2) (2004) 132–144. [13] S. Ahmed, D. Sriram, R. Logcher, Transaction-management issues in collaborative engineering, J. Comput. Civ. Eng., ASCE 6 (1) (1992) 85–105. [14] H. Peltonen, T. Mannisto, K. Alho, R. Sulonen, An engineering document management system, Proc. ASME Winter Annu. Meeting, 1993. [15] D. Spooner, M. Hardwick, Using persistent object technology to support concurrent engineering systems, Concurrent Engineering: Methodology and Applications, Elsevier Science, Amsterdam, 1993, pp. 205–234. [16] R. Ganeshan, J. Garrett, S. Finger, A framework for representing design intent, Des. Stud. 15 (1) (1994) 59–84. [17] K. Krishnamurthy, K. Law, A data management model for design change control, Concurr. Eng. Res. Appl. 3 (4) (1995) 329–343. [18] A. Mokhtar, C. Bedard, P. Fazio, Information model for managing design changes in a collaborative environment, J. Comput. Civ. Eng., ASCE 12 (2) (1998) 82–92. [19] C. Soh, Z. Wang, Parametric coordinator for engineering design, J. Comput. Civ. Eng., ASCE 14 (4) (2000) 233–240. [20] T. Hegazy, E. Zaneldin, D. Grierson, Improving design coordination for building projects: I. Information model, J. Constr. Eng. Manag., ASCE 127 (4) (2001) 322–329. [21] A. Karim, H. Adeli, CONSCOM: an OO construction scheduling and change management system, J. Constr. Eng. Manag., ASCE 125 (5) (1999) 368–376. [22] C. Charoenngam, S.T. Coquince, B.H.W. Hadikusumo, Web-based application for managing change orders in construction projects, J. Constr. Innov. 3 (2003) 197–215. [23] J. Sterman, System Dynamics Modeling for Project Management, Sloan School of Management, MIT, On-line Publication, 1992 bhttp://web.mit. edu/jsterman/www/N, Last Visit: October, 2002. [24] J. Sterman, Business Dynamics: System Thinking and Modeling for a Complex World, McGraw-Hill Companies, New York, NY (2000) pp. 55–61, 469–511, 563–567, and 587–595.
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377 [25] S. Lee, F. Peña-Mora, M. Park, Dynamic quality and change management framework for concurrent design and construction, ASCE Construction Research Congress, March 19–21, 2003, Honolulu, Hawaii, ASCE, Reston, 2003, pp. NA (Electronic Book). [26] M. Park, F. Peña-Mora, Dynamic change management for construction: introducing change cycle into model-based project management, System Dynamics Review, vol. 19, No. 3, System Dynamics Society, Albany, NY, 2003, pp. 213–242. [27] I.A. Motawa, C.J. Anumba, A. El-Hamalawi, P.W.H. Chung, M. Yeoh, Modelling change processes within construction projects, Proceedings of
377
the Second International Conference on Structural and Construction Engineering (ISEC-02), Rome, Italy, Sep. 2003, pp. 2185–2190. [28] I.A. Motawa, C.J. Anumba, A. El-Hamalawi, P.W.H. Chung, M. Yeoh, An innovative approach to the assessment of change implementation in construction projects, Proceedings of the Second International Conference on Innovation in Architecture, Engineering and Construction, Loughborough University, UK, June 2003, pp. 729–740. [29] S. Lee, Dynamic quality and change management for large scale concurrent design and construction projects, MIT Master Thesis, 2003.
An integrated system for change management in construction I.A. Motawa a,⁎, C.J. Anumba b , S. Lee c , F. Peña-Mora d a
c
Department of Structural Engineering, Faculty of Engineering, Mansoura University, Egypt b Department of Civil and Building Engineering, Loughborough University, UK Intelligent Engineering Systems Laboratory, Civil and Environmental Engineering Department, MIT, Cambridge, MA 02139, USA d Civil and Environmental Engineering Department, University of Illinois at Urbana-Champaign, IL 61801, USA Accepted 25 July 2006
Abstract Change management in construction is an important aspect of project management, as changes constitute a major cause of delay and disruption, and it is widely accepted by both owners and constructors that change effects are difficult to quantify and frequently lead to disputes. The development of change management systems should consider many elements of the project processes and address all internal and external factors that influence project changes. This paper presents an integrated change management system developed to represent the key decisions required to implement changes and to simulate the iterative cycles of concurrent design and construction resulting from unanticipated changes and their subsequent impacts. The system integrates a fuzzy logic-based change prediction model with the system dynamics model of the Dynamic Planning and control Methodology (DPM), which has been developed to evaluate the negative impacts of changes on construction performance. The developed system can be used in managing change scenarios on projects and also in evaluating change effects depending on the available information at the early stages of projects. © 2006 Elsevier B.V. All rights reserved. Keywords: Change management; Fuzzy logic; Dynamic planning
1. Introduction Changes in construction projects are common and likely to occur from different sources, by various causes, at any stage of a project, and may have considerable impacts. Based on time, change could be anticipated or emergent, proactive or reactive, or pre-fixity or post-fixity. Based on need, change could be elective or required, discretionary or non-discretionary, or preferential or regulatory. Based on effect, change could be beneficial, neutral or disruptive. Change management relates all the internal and external factors that influence project changes. It seeks to forecast possible changes; identify changes that have already occurred; plan preventive impacts; and coordinate changes across the entire project [1]. Inconsistent management of the change process can result in many disruptive effects.
⁎ Corresponding author. E-mail address: [email protected] (I.A. Motawa). 0926-5805/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.autcon.2006.07.005
Some of these consequences can be relatively easy to measure, while others are more difficult to quantify. Change management is considered an integral part of project management. This paper presents the development of an integrated system for improved change management. The following section reviews the work done on this topic. 1.1. Literature review Research on modelling the change process in construction has tended to focus on the identification of factors affecting the success of a change process, and resulted in guidance for best practice in change management. Examples of such guidance include: a concept for project change management [2], best practices for managing change efficiently [3], a generic procedure for issuing a change order request [4], an analysis method to reduce the overall rate of construction change orders [5], a best practice guide to present best practice recommendations for the effective management of change on projects [6], and an advanced project change management system [7].
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
Research has also been undertaken on evaluating the change effects on certain project elements. These studies dealt mainly with a single factor or a single project element such as: construction change order impacts on labour productivity at the craft level [8], effect of the size of change and its impact time on a project [9], a linear regression model that predicts the impact of change orders on labour productivity [10], the risk of changes to safety regulations and its effect on a project [11], and decision tree models to classify and quantify productivity losses caused by change order impacts [12]. Change management has been the focus of different IT systems. For example, an integrated environment for computeraided engineering was developed by Ahmed et al. [13], which is a blackboard representation that integrates a global database, several knowledge modules, and a control mechanism to systemize object changes. Peltonen et al. [14] proposed an engineering document management system for changes that incorporated document approval and release procedures. Spooner and Hardwick [15] developed a system with rules for coordinating concurrent changes and for identifying and resolving conflict modifications. Ganeshan et al. [16] developed a system to capture the history of the design process, initiate backtracking, and determine the decisions that might be affected when changes are made in the spatial design of residential buildings. Krishnamurthy and Law [17] presented an interesting change management model that supports multidisciplinary collaborative design environments. Another change management system was proposed by Mokhtar et al. [18] for managing design change in a collaborative environment. The model is capable of propagating design changes and tracking past changes. Soh and Wang [19] proposed a constraint methodology based on a parametric technique to coordinate design consistency between different geometric models and to facilitate managing design changes. Hegazy et al. [20] introduced an information model to facilitate design coordination and management of design changes. Important dependencies between building components were represented by this model to help identify the ripple effect of changes between components. Also, a reporting system was used to view the history of all changes made by all disciplines. A more generic ITsystem was presented by Karim and Adeli [21] which is an object-oriented (OO) information model for construction scheduling, cost optimization, and change order management. Charoenngam et al. [22] developed a Web-based change order management system that supports documentation practice, communication and integration between different team members in the change order workflow. The above literature on change management and evaluation mainly focused either on the identification of the change process, best practice recommendations for managing change during the project life cycle, or on the evaluation of the change effects on a single project parameter. Much of the discussion is presented in categorical ways with little attention being paid to modelling the dependent data or simulating the iterative cycles of concurrent design and construction that result from unanticipated changes and their subsequent impacts on project performance. Therefore, there is a clear need in the construction industry for research work to focus on modelling this depen-
369
dency, especially for multi-disciplinary causes and effects, and planning such iterative tasks. Some of the IT systems developed for change management are integrated systems that represent design information, record design rationale, facilitate design co-ordination and changes, and notify users of file changes. These systems were developed mainly to deal with reactive changes, particularly design changes. The research presented in this paper focuses on both proactive and reactive changes. Reactive changes represent the events when a change occurs and the project team starts to take actions to remedy the consequences of this change. While the proactive changes represent the events when a change is likely to occur in a later stage and the project team plans to minimize the disruptive effect of these changes. Because the effect of changes always concern practitioners, the concept of change effects in the proposed system is presented in the following section. 1.2. Change effect on project life cycle Studying the life cycle of changes and their impact on construction performance during actual execution shows that construction projects are inherently complex, dynamic, and involve multiple feedback processes [23]. The uncertainty and complexity of design and construction projects are usually driven by these feedback processes. There are only two types of feedback processes: reinforcing (that generates other changes or makes errors) and balancing (that resolves such changes and errors). Dynamics in a system arise from the interaction of these two types of feedback processes among the components of the system, not from the complexity of the components themselves [24]. Fig. 1 shows the basic simultaneity of the reinforcing and balancing feedback in design and construction projects. The control actions to address changes can have the intended effect of resolving the issues that initiate the control actions, if the decision is correct and well implemented. At the same time, they can produce a side effect that may create some unintended problems, if the decision is incorrect, not well implemented, exceeds the time frame of its effectiveness or if a project manager does not realize the impact of the control actions on other related activities. In addition, the reconciliation of the gap between the initial work scope and the actual work scope can also result in these feedback processes. After the project starts, the actual work scope may be increased, since additional work is often added to the project scope in order to deal with changes. Moreover, these unintended effects become more detrimental when concurrent engineering techniques are applied. This is because the decision to take control actions against unanticipated additional work has to be made within the complex inter-relationships of activities, even with a lack of complete information about predecessor activities [25]. Based on this concept, the research presented in this paper integrates the work done on the stage of “Change identification”, shown in Fig. 1, and the work done on simulating the uncertainty resulting from feedback processes caused by change. A generic change process model was developed to give full definition for the change over the time of its occurrence and to
370
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
Fig. 1. Multiple feedback processes caused by changes.
represent the key decisions required to implement changes. A simulation system was also developed to address the uncertainty resulting from changes. The Dynamic Planning and control Methodology (DPM) has been adopted to assist in the preparation of robust construction plans and to provide policy guidelines to handle changes [25,26]. The system provides improved co-ordination and control over changes, thus helping to increase the consistency and productivity of the overall project process. The following sections present a generic change process model and a fuzzy-based model for predicting the likelihood and impact of changes. 2. Generic change process model The developed model is based on a synthesis of the change process models reviewed in the literature and the process models adopted on a number of case studies undertaken during the research project. The process model, shown in Fig. 2, is a generic change process model that can be applied to different change categories, such as pre- or post-fixity changes. The process model has four main sections, as shown in Fig. 2. These are: Start up:
Identify and evaluate:
At the “Start Up” stage, the generic process defines a set of proactive requirements that are essential for effective change management. These requirements enable the project team to respond readily to change, to manage change effectively, and to facilitate contingency plans for any unanticipated change. Details of such requirements are presented elsewhere [27]. Change identification should cover change causes, types and effects. It
should also define the relevant project processes and departments affected by the change or involved in the change decision. The change management process model requires project teams to keep records of all relevant information on change cases to build a case base for future use. Analysis of change options is required for decision-making – whether to go ahead with any of the change options or to undertake further investigations. The criteria required to carry out this analysis may include tangible and intangible criteria. The evaluation steps include options evaluation, implications assessment and optimum selection of change options. Different models and decision support systems can be used to help decision-makers select an optimum solution and evaluate the change options using quantitative and qualitative criteria. Approval and propagation: Client approval is an important step in the process while different outputs are expected, as shown in Fig. 2. The client needs to review potential changes against the project baseline using tangible and intangible criteria. In many cases, clients need to use decision-making techniques for evaluation and comparison
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
Fig. 2. Generic change process model.
371
372
Post change:
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
in order to decide on a change option. The rest of the process stages in Fig. 2 involve integration between documentation and communication facilities. When dispute resolution is applicable, it requires the investigation of direct and indirect causes of change. In this situation, the analysis of the effect of multiple change causes could be prepared.
This process model has been studied further. The study paid much attention to the link between project implementation and the role of change. From this study the term ‘stability’ has been defined. Change depends on the stability of the given initial scope of an activity. Stability indicates the degree to which the given work scope would be performed without a request for change. High stability means that only a small number of changes would be expected during the execution of a particular activity, while low stability represents the possibility that a great number of changes would be requested. The internal iterative sub-processes entailed in the implementation of the generic change process, as shown by the bold links in Fig. 2, increase the effects of change on the project life cycle as denoted above for the feedback processes in Fig. 1. Identified changes are usually approved based on their feasibility in the project through the claim and change management process. For instance, if the identified change is perceived to require a significant investment or there are other options to replace it, it can be rejected. If a change is approved, the corresponding additional work scope is introduced to that particular activity. A rejected change may either become a permanently rejected change or it can be designated as a latent change in terms of its potential for reconsideration later in the process, as shown by the bold links in Fig. 2. Based on the above change concept, an IT system has been developed to help in managing the change process effectively. This paper presents two main functions of the system. The first aims to predict the level of stability, defined above, by predicting the likelihood of change occurrence. The second function aims to simulate the potential iterations that may occur during change implementation. The level of stability, determined by the first function, is among the basic data required to run the simulation system, developed to implement the second function. Fig. 3 illustrates the application logic of the system, which shows how the system components, presented in this paper, work together. The following sections describe how the system operates conceptually. 3. Change prediction system The main purpose of this component is to help in estimating the level of stability defined above. The estimation is based on investigating the information available at the early stages of projects and using such information to predict change events. Aggregating such events will help in defining the level of stability
Fig. 3. Application logic of the system.
for a project. This prediction will also help in taking appropriate actions to minimize the disruptive effects of changes. To develop this component, it was important to identify what project characteristics lead to change causes and what these causes are, and then to understand how these causes are related to effects. Some of the key questions were: How does one factor relate to another? What are the internal mechanisms by which a particular factor causes a change in another factor? Fig. 4 gives an example of how the change prediction system relates these factors to one another for a change case. Multiple causes can lead to a change case, and the effects of this change case cannot be added linearly (i.e. different causes of change may be responsible for a certain effect or a set of effects). In other words, the effect of a change cause C1 and C2, occurring together, may in general result in more or less impact than the sum of the effect of change cause C1 occurring on its own and the effect of change cause C2 occurring on its own. An analysis is required to find out the possibility of occurrence of each cause of change (Cm) with respect to each project characteristic (Fn). Multiple causes of change are a complex issue to determine the corresponding impacts (El) on projects. Lists of Fn, Cm, and El, were identified through case studies, details can be found elsewhere [28]. This helps project teams check their project against these lists. Fuzzy systems were found more suitable to describe the relationship between the above variables for the following reasons: • The information, if available, for the change cases is always vague. When little or no information is available, fuzzy systems are appropriate to represent the human perception on a specific issue based on a variety of assumptions which may be numerically or linguistically formatted. • Change consequences might lead to dispute, therefore, detailed analyses and investigation of the change causes and effects are required. This needs close observation for the simulation route of cause–effect relationships.
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
373
the values for “Fn” will be given a weight on this scale to represent its contribution, then the Rij will have another weight on a 0–10 scale to represent the sensitivity of the change cause to variations in the Fn and so for the Rjk. The system enables the user to estimate the level of stability of the studied project, i.e. the degree to which the given work scope would be performed with changes. Fig. 5 shows an example of the system output. 3.1. Simulating the iterative cycles caused by changes
Fig. 4. A dependency diagram for a change case.
Two different fuzzy approaches were investigated; namely building a fuzzy rule-based model to simulate the cause and effect relationships of a change event or establishing fuzzy relations between the prediction elements of change. The first approach is part of any general fuzzy logic system which, in addition to the fuzzy rule base, includes fuzzy inference process, fuzzification process, and defuzzification process. Developing a fuzzy rule-based system requires identification of different rules that link the prediction elements in IF–THEN format. The main difficulty in this approach is about the complexity of the problem in hand, as the list of prediction elements collected for the proposed model initially includes: 30-elements of project characteristics, 20-elements of change causes, and 29-elements of change effects. A further approximation may be used to simplify building such rule-based system. This is by assuming predefined fuzzy-sets to fire the IF–THEN rules. This is also a controversial solution and the justifications for such assumptions are always questioned. Therefore, the proposed system adopted the fuzzy relation approach considering the most appropriate methods for improving the model accuracy. This approach gives flexibility for the practitioners to express their belief about the problem with limited calculation assumptions. The relationship between the elements are formed and combined to produce the cumulative relation, as shown by Eq. (1): ! j¼m Xk¼l i¼n; X j¼m; f Fi ; Rij ; Rjk ¼ Fi 4Rij ο Rjk ð1Þ j; k¼1
Based on the level of stability obtained, the impact of iterative cycles caused by changes on construction performance is estimated by the Dynamic Planning and control Methodology (DPM) [26,29]. DPM aims to provide policy guidelines for unexpected events by supplementing network-based tools with mechanisms to represent the dynamics of a project. To identify the impact of iterative change cycles on construction performance, the DPM incorporates the system dynamics based design and construction process model [29], as seen in Fig. 6. It aims to address the iterative change cycles, their settlement through Requests For Information (RFI), the change decision process, and the corresponding work amount increased. Managing changes in design and construction usually have two major components, the Scope Management (SM) process and the Claim and Change Management (CCM) process. The SM process is the review process and the CCM process is the decision making process for change adoption o rejection. Specifically, before the execution of a task, the SM process is applied to WorkToDo (WTD) stock, as denoted by A in Fig. 7. The SM process aims to make sure that the given scope of work and the corresponding work setting are the same, as planned. Thus, if this WTD stock differs from the originally planned WTD stock, those tasks in WTD stock are sent to the stock of WorkAwaitingCCMGDecision (WACCMGD – B in Fig. 7), which needs the decision or analysis of Claim and Change Management (CCM) group. However, some potential changes may not be identified (i.e. latent changes) during the SM process. In this sense, Scope Management THoroughness (SMTH) is defined as the degree to
i; j¼1
where: n m l Rij Rjk
no. of the project characteristics no. of the change causes no. of the change effects the sensitivity of a change cause ( j) occurrence to variations in the project characteristics (i); the sensitivity of the change impact (k) to variations in the change causes ( j).
All units used to quantify the variables in Eq. (1) are based on a 0–10 scale. It is assumed that all variables have the value of “0” when no change occurs. In case of predicting a change case,
Fig. 5. Output of the change prediction system.
374 I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
Fig. 6. Design and construction process model [based on Pugh–Roberts Associates; Ford and Sterman, 1998].
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
375
Fig. 7. Modelling change management with system dynamics.
which the potential changes have been identified during the SM process. Based on SMTH, changes could be identified (i.e. identified change) or not (i.e. latent change). In the model, tasks that flow from WTD stock to WACCMGD stock are identical to identified changes (C in Fig. 7). Then, these identified changes can be approved (i.e. approved change) or rejected (i.e. rejected change), based on the decision of the CCM group. In the model, if a change is rejected, tasks that have been suspended would flow back to WTD and will be performed (D in Fig. 7). If a change is approved, tasks would also flow back to WTD,
however, in this case, with additional work generated by this approved change (E in Fig. 7). The detailed model structure is presented elsewhere [29]. An example of the simulation results generated by the DPM is presented in Fig. 8. The typical relationships between activities, such as the finish-to-start (FS) relationship between final design activity and the shop drawing submittal activity in Fig. 8, do not make the following construction activities start until the design activities are completed. While in concurrent design and construction processes, the design work may be
Fig. 8. Simulation results generated by the DPM.
376
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377
forced to proceed without the establishment of a detailed work scope by the owner. Thus, requests for design clarification and/ or changes are often made by the contractor. Design omissions and errors may also be committed by the design group of the team. These are the main reasons for the delay. The simulation results exactly capture the aforesaid reality. For the studied example, a large coordinating work amount, caused by waiting for a RFI reply and the CCM decision, was observed (column A in Fig. 8). In addition, a significant work amount was newly introduced to deal with changes in the final design (column B in Fig. 8) and also contributed to delay. Furthermore, a large fraction of the changes was not identified and became latent changes (column B in Fig. 8). This situation also required much effort in coordinating with the predecessor activity through RFIs. Identifying such detrimental impacts caused by changes, DPM explicitly captures the negative impact of iterative change cycles. More results about the studied example can be found elsewhere [29]. 4. Discussion and conclusion An integrated change management system has been presented in this paper. It was developed to enable AEC professionals manage changes more effectively. The system covers the life cycle of changes within construction projects. A change process model was developed first to represent the key decision points required to implement changes. In support of this model, an integrated system was developed. Two components of the system were presented in the paper; a Dynamic Planning and control Methodology (DPM) and a change prediction system. DPM has been developed to overcome the uncertainties and complexities resulting from changes in concurrent design and construction by focusing on iterative cycles caused by changes and their impacts on construction performance. Among the main data required to simulate such iterative cycles, there is the level of project stability. The stability of a project can be predicted by studying the available information of the project at the early stages. Therefore, the change prediction system aimed to determine the likelihood of change occurrence, which is a measure of the project stability. The integrated system can help in: • providing robust planning and control actions, • identifying various dynamic impacts of construction feedbacks and iterative cycles, • planning projects with allowances for the potential changes, • exploring the cause-and-effect relationships of change events, and • examining the impact of changes on different project parameters. Although the current system has established the potential for proactive change management, there is a need for significant effort to ensure its realistic implementation. Thus, further simulation and experimentation are required to establish robust mathematical relationships between other variables of the change process. Additional case studies would be required in order to validate the effectiveness of the developed system. In this way, the
system could enhance its practical simulation capability and provide more reliable and realistic implementation. References [1] V. Voropajev, Change management – a key integrative function of PM in transition economies, Int. J. Proj. Manag. 16 (1) (1998) 15–19. [2] Construction Industry Institute (CII), Project Change Management Research Team, Special Publication, vol. 43-1, The University of Texas at Austin, US, 1994. [3] Construction Industry Institute (CII) conference, Project Change Management – Implementation Feedback Report, The University of Texas at Austin, US, 1996. [4] I.D. Cox, J.P. Morris, J.H. Rogerson, G.E. Jared, A quantitative study of post contract award design changes in construction, J. Constr. Manag. Econ. 17 (1999) 427–439. [5] S.N. Stocks, A. Singh, Studies on the impact of functional analysis concept design on reduction in change orders, J. Constr. Manag. Econ. 17 (1999) 251–267. [6] Construction Industry Research and Information Association (CIRIA), Managing Project Change – a Best Practice Guide, CIRIA, UK, 2001. [7] C.W. Ibbs, C.K. Wong, Y.H. Kwak, Project change management system, J. Manage. Eng. ASCE 17 (3) (2001) 159–165. [8] W.T. Hester, J.A. Kuprenas, T.C. Chang, Construction changes and change orders: their magnitude and impact, Construction Industry Institute (CII), Source Document, vol. 66, CII, Austin, Tex, 1991. [9] C.W. Ibbs, Quantitative impacts of project change: size issues, J. Constr. Eng. Manag., ASCE 123 (3) (1997) 308–311. [10] A.S. Hanna, J.S. Russell, P.J. Vandenberg, The impact of change orders on mechanical construction labour efficiency, J. Constr. Manag. Econ. 17 (1999) 721–730. [11] T.M. Williams, Safety regulation changes during projects: the use of system dynamics to quantify the effects of change, Int. J. Proj. Manag. 18 (1) (2000) 23–31. [12] M. Lee, A.S. Hanna, W. Loh, Decision tree approach to classify and quantify cumulative impact of change orders on productivity, J. Comput. Civ. Eng., ASCE 18 (2) (2004) 132–144. [13] S. Ahmed, D. Sriram, R. Logcher, Transaction-management issues in collaborative engineering, J. Comput. Civ. Eng., ASCE 6 (1) (1992) 85–105. [14] H. Peltonen, T. Mannisto, K. Alho, R. Sulonen, An engineering document management system, Proc. ASME Winter Annu. Meeting, 1993. [15] D. Spooner, M. Hardwick, Using persistent object technology to support concurrent engineering systems, Concurrent Engineering: Methodology and Applications, Elsevier Science, Amsterdam, 1993, pp. 205–234. [16] R. Ganeshan, J. Garrett, S. Finger, A framework for representing design intent, Des. Stud. 15 (1) (1994) 59–84. [17] K. Krishnamurthy, K. Law, A data management model for design change control, Concurr. Eng. Res. Appl. 3 (4) (1995) 329–343. [18] A. Mokhtar, C. Bedard, P. Fazio, Information model for managing design changes in a collaborative environment, J. Comput. Civ. Eng., ASCE 12 (2) (1998) 82–92. [19] C. Soh, Z. Wang, Parametric coordinator for engineering design, J. Comput. Civ. Eng., ASCE 14 (4) (2000) 233–240. [20] T. Hegazy, E. Zaneldin, D. Grierson, Improving design coordination for building projects: I. Information model, J. Constr. Eng. Manag., ASCE 127 (4) (2001) 322–329. [21] A. Karim, H. Adeli, CONSCOM: an OO construction scheduling and change management system, J. Constr. Eng. Manag., ASCE 125 (5) (1999) 368–376. [22] C. Charoenngam, S.T. Coquince, B.H.W. Hadikusumo, Web-based application for managing change orders in construction projects, J. Constr. Innov. 3 (2003) 197–215. [23] J. Sterman, System Dynamics Modeling for Project Management, Sloan School of Management, MIT, On-line Publication, 1992 bhttp://web.mit. edu/jsterman/www/N, Last Visit: October, 2002. [24] J. Sterman, Business Dynamics: System Thinking and Modeling for a Complex World, McGraw-Hill Companies, New York, NY (2000) pp. 55–61, 469–511, 563–567, and 587–595.
I.A. Motawa et al. / Automation in Construction 16 (2007) 368–377 [25] S. Lee, F. Peña-Mora, M. Park, Dynamic quality and change management framework for concurrent design and construction, ASCE Construction Research Congress, March 19–21, 2003, Honolulu, Hawaii, ASCE, Reston, 2003, pp. NA (Electronic Book). [26] M. Park, F. Peña-Mora, Dynamic change management for construction: introducing change cycle into model-based project management, System Dynamics Review, vol. 19, No. 3, System Dynamics Society, Albany, NY, 2003, pp. 213–242. [27] I.A. Motawa, C.J. Anumba, A. El-Hamalawi, P.W.H. Chung, M. Yeoh, Modelling change processes within construction projects, Proceedings of
377
the Second International Conference on Structural and Construction Engineering (ISEC-02), Rome, Italy, Sep. 2003, pp. 2185–2190. [28] I.A. Motawa, C.J. Anumba, A. El-Hamalawi, P.W.H. Chung, M. Yeoh, An innovative approach to the assessment of change implementation in construction projects, Proceedings of the Second International Conference on Innovation in Architecture, Engineering and Construction, Loughborough University, UK, June 2003, pp. 729–740. [29] S. Lee, Dynamic quality and change management for large scale concurrent design and construction projects, MIT Master Thesis, 2003.