Contract

10
ISSN 1744-9847 ISBN 1-905148-09-7
Effective Management of Contract Variations using a Knowledge Based Decision Support System
Dr Faisal Manzoor Arain, Institute of Business Administration, Pakistan Dr Low Sui Pheng, National University of Singapore
10
Dr Faisal Manzoor Arain, Institute of Business Administration, Pakistan Dr Low Sui Pheng, National University of Singapore
Editor: Chris Webster Editorial Assistant: Diane Bowden
Published by the Centre for Education in the Built Environment Cardiff University, Bute Building, King Edward VII Avenue, Cardiff CF10 3NB
The views expressed in this Working Paper are those of the author(s) and do not necessarily reflect those of the Higher Education Academy Centre for Education in the Built Environment
Abstract
Construction projects are complex because they involve many human and non-human factors and variables. They usually have a long duration, various uncertainties and complex relationships among the participants. The need to make changes in a construction project is a matter of practical reality. Even the most thoughtfully planned project may necessitate changes due to various factors. The management of variations in a project can be enhanced by the identification and analysis of potential project variations as early as possible. Learning from these variations is imperative because the professionals can improve and apply their experience in the future. This CEBE Working paper outlines how a knowledge based decision support system could be used to help students gain an understanding of how contract variations might be avoided. The paper principally outlines the results of a survey of contractors, consultants and developers working on educational building projects in order to identify the effects of unforeseen variations, and how they might be controlled. Finally, the study presents a knowledge-based decision support system (KBDSS) for effective management of variations in educational building projects.
Keywords: Building Contracts, Construction Management, Contract Administration,

Variations, Professional Practice, Knowledge Based Decision Support Systems.
CEBE Working Paper No. 10
Table of Contents
1.0 INTRODUCTION ....................................................................... 4 2.0 SIGNIFICANCE AND RATIONALE OF STUDY ............................. 6 3.0 OBJECTIVE OF THE STUDY ....................................................... 9 4.0 HOLISTIC VIEW OF VARIATION ORDERS ............................... 11
4.1 Potential Causes of Variations A. Owner Related Changes B. Consultant Related Variations C. Contractor Related Variations D. Other Variations 4.2 Potential Effects of Variations 4.3 A. B. C. Controls for Variation Orders Design Stage Controls for Variation Orders Construction Stage Controls for Variation Orders Design-Construction Interface Stage Controls for Variation Orders 11 12 14 17 19 20 24 25 26 28
5.0 SCOPE OF RESEARCH ............................................................. 31 6.0 RESEARCH METHODOLOGY .................................................... 32 7.0 BACKGROUND OF RESPONDENTS ........................................... 34 8.0 ANALYSIS OF RESULTS .......................................................... 36 9.0 DISCUSSION .......................................................................... 43
9.1 Most Significant Causes of Variation Orders 9.2 Most Frequent Effects of Variation Orders 9.3 Most Effective Controls for Variations 43 45 46
10.0 KNOWLEDGE-BASED DECISION SUPPORT SYSTEM (KBDSS) 52

10.1 Knowledge-Base 54 Macro layer ...........................................................................................54 Micro layer ............................................................................................55 Effects and controls layer ........................................................................56 10.2 Decision Support Shell 57 Main panel ............................................................................................57 Building the hierarchy between criterions and controls ................................58
Rating the controls .................................................................................58 Selecting the best controls ......................................................................59
11.0 CONCLUSIONS ..................................................................... 60 12.0 RECOMMENDATIONS............................................................ 62 13.0 PRACTICAL APPLICATION OF RESEARCH ............................. 64 14.0 DISTINCT FEATURES OF THE SYSTEM .................................. 67 16.0 FUTURE WORK ..................................................................... 69 ACKNOWLEDGEMENTS ................................................................. 70 REFERENCES ................................................................................ 71 APPENDIX 1: RELATIVE IMPORTANT INDEX OF CAUSES AND CONTROLS ................................................................................... 75 APPENDIX 2: KNOWLEDGE BASED DECISION SUPPORT SYSTEM (KBDSS)....................................................................................... 77
1.0 Introduction
Designing and constructing a building project is a collaborative effort among professionals from independent disciplines, such as architecture, structure and electrical and mechanical etc. As variations are common in all types of construction projects (OBrien, 1998; Ibbs et al., 2001) the subject of variations in building projects is very significant for students in architecture, building/construction management and quantity surveying. The knowledge based decision support system (KBDSS) and the study would assist students in learning about variance performance in the particular case studies reported i.e., educational building projects in Singapore. Specifically, the KBDSS and the study would be useful for 4th year Architecture, Building and Quantity Surveying students. For Architecture students, the relevant modules would be Professional Practice or Architectural Practice; for Building and Quantity Surveying students, the relevant modules would be Contract Administration or Professional Practice. The system would assist them in learning about the issues of designs, contracts, management and project variance through the wealth of information based on past educational projects provided in the KBDSS. The KBDSS would be useful to the students as a more general research tool as students could populate it with their own data and use with the educational projects reported in this paper for comparison. Construction projects are complex because they involve many human and non-human factors and variables. They usually have a long duration, various uncertainties and complex relationships among the participants. The need to make changes in a construction project is a matter of practical reality. Even the most thoughtfully planned project may necessitate changes due to various factors (OBrien, 1998). The high living standards in Singapore have generated many manufacturing and building employment opportunities. The growth of towns has accelerated as a result of high population growth. Large and complex projects have been built, attracting contractors from all over the world. Most of these contractors appear to lack a sufficient understanding of the social, cultural and physical environment of Singapore (Dulaimi and Hwa, 2001). This situation, coupled with inexperienced owners, has led to inadequate designs resulting in many changes to plans, specifications, and contract terms. The education sector development and the new modes of teaching and learning foster the need for renovation or extension of existing academic institutions. The change of space in academic institutions is required to cater for the new technology used. The construction of an educational building also poses risks as in the construction of any other huge projects. Variations during the design and construction processes are to be expected. Variations are inevitable in any construction project (Mokhtar et al., 2000). Needs of the owner may change
in the course of design or construction, market conditions may impose changes to the parameters of the project, and technological developments may alter the design and the choice of the engineer. The engineers review of the design may bring about changes to improve or optimize the design and hence the operations of the project. Furthermore, errors and omissions in engineering or construction may force a change. All these factors and many others necessitate changes that are costly and generally un-welcomed by all parties. Consideration must be given from the initial stages (inception) of the project until commissioning. Contractual provision is required to define the conduct of owner, consultant and contractor to participate in and manage variations. Systematic and proper procedures must be set in place to process a change from conceptual development until it materializes in the field. The reality is that an adverse environment exists among parties in the construction industry. Variations could be perceived as positive or negative to the preconceived goals of the professionals involved in a project. Therefore, a major variation must be managed and handled professionally in order to minimize its cost, schedule and consequential impacts that may divert the project away from its targeted goals. To identify and analyze potential variations in a project as early as possible can enhance the management of project. Learning from these variations is imperative because the professionals can improve and apply their experience in the future.
2.0 Significance and Rationale of Study

Variation orders are an unwanted, but inevitable reality of every construction project (Clough and Sears, 1994; OBrien, 1998; Mokhtar et al., 2000). Construction projects are bound to encounter variation orders; the goal of the owner, design or construction manager is to limit the number of such changes (CII, 1994b; Ibbs, 1997a). Proper management of variation orders is very significant for all types of construction projects. Variations in drawings and contract documents usually lead to a change in contract price or contract schedule. Variation also increases the possibility of contractual disputes. Conventionally, variations present problems to all the parties involved in the construction process. Mendelsohn (1997) observed that probably 75% of the problems encountered on site were generated at the design phase. This is not to say that contractors do not create a slew of problems of their own, but that these problems were often compounded by inherent design flaws. If one were to seriously consider ways to reduce problems on site, an obvious place to begin is to focus on what the project team can do to eliminate these problems at the design phase. There are many reasons for issuing construction variation orders in the construction process. It can be a result of the non-availability or slow delivery of required materials, or the correction of contract document errors and omissions. Identifying the causes of variation orders is very important in order to avoid potential changes in future projects, or to minimize their effects. The construction process is influenced by highly changing variables and unpredictable factors that could result from different sources. These sources include the performance of construction parties, resources availability, environmental conditions, involvement of other parties and contractual relations. As a consequence of these sources, the construction of projects may face problems which could cause delay in the project completion time (Clough and Sears, 1994). Kumaraswamy et al. (1998) studied claims for extension of time due to excusable delays in Hong Kongs civil engineering projects. Their findings suggested that 15-20% time over run was mainly caused by inclement weather, 50% of the projects surveyed were delayed because of variations. Kaming et al. (1997) studied the factors influencing construction time and cost over runs for high rise projects in Indonesia where 31 project managers working in high rise buildings were surveyed. Kaming et al. (1997) pointed out that the major factors influencing cost over run were material cost increase due to inflation, inaccurate material estimating and the degree of
project complexity. In the case of time over run, the most important factors that caused delays were design changes, poor labour productivity, inadequate planning, and resource shortage. The magnitude of schedule average slippage due to variations was reported as 18% (CII, 1990b; Burati et al., 1992; Zeitoun and Oberlender, 1993, Kumaraswamy et al., 1998). The deviation (variation) cost amounted to an average of 14% of the total cost of the project (CII, 1990b; Burati et al., 1992; Zeitoun and Oberlender, 1993). Although there have been cases where variation cost accounted for as high as 100% of the budgeted funds, the industry norm of this percentage was about 10%. The impact of variations varies from one project to another. However, it is generally accepted that the variations affect the construction projects with unpalatable consequences in time and cost (Hester et al., 1991; Barrie and Paulson, 1992; CII, 1994b; Ibbs et al., 2001; Arain et al., 2004). Variations are common in all types of construction projects (CII, 1994b; Fisk, 1997; OBrien, 1998; Ibbs et al., 2001). The nature and frequency of variations occurrence vary from one project to another depending on various factors (CII, 1986b; Kaming et al., 1997). Variations in construction projects can cause substantial adjustment to the contract duration, total direct and indirect cost, or both (Tiong, 1990; Odell, 1995; Ibbs, 1997b; Ibbs et al., 1998). Therefore, project management teams must have the ability to respond to variations effectively in order to minimize their adverse impact to the project. Great concern has been expressed in recent years regarding the impact of variations to the construction projects. As mentioned earlier, variations are frequent in construction projects and can cause considerable adjustment to the project time, cost and quality. The causes of variation orders are diverse, thus making the task of variation management difficult for most clients. However, the undesirable situation can be minimized as long as a mechanism for handling variation orders and making more informed decisions based on the past projects can be understood and built into project management. The litmus test for successful management should not be whether the project was free of variation orders, but rather, if variation orders were resolved in a timely manner to the benefit of all the parties and the project (Ibbs et al., 2001). A clearer view of the causes and their impacts and controls will enable the project team to take advantage of beneficial variations when the opportunity arises, without an inordinate fear of negative impacts. A clearer and comprehensive view of causes, their effects and potential controls, based on past projects, will assist the project team to learn from past experiences and to make more informed decisions for effective management of variation orders. No such studies have been undertaken on the management and control of variation orders on a large scale using a
knowledge-based decision support system (KBDSS) platform. The KBDSS will assist professionals in analyzing variations, and selecting the appropriate controls for minimizing their adverse impacts by providing timely information. Furthermore, by having a systematic way to manage variations, the efficiency of project work and the likelihood of project success should increase. Furthermore, the KBDSS will provide an excellent opportunity for the project managers to learn from past experiences. The KBDSS will help to enhance productivity and cost savings by providing timely information for decision makers/project managers to make more informed decisions. The undesirable effects (i.e., delays and disputes) of variations may be avoided as the decision makers/project managers would be prompted to guard against these effects through the KBDSS. The knowledge base and pertinent information displayed by the KBDSS will provide useful lessons for decision makers/project managers to exercise more informed judgments in deciding where cost savings may be achieved in future educational building projects. Furthermore, the KBDSS will provide a useful tool for training new staff whose work scope includes educational building projects. This is a timely study as a programme of rebuilding and improving existing educational buildings is currently underway in Singapore; it provides the best opportunity to address the contemporary issues relevant to the management of variation orders.
3.0 Objective of the Study

Successful management of variation orders and claims begins even before the start of construction (Ibbs et al., 2001). The project owner must accept that no construction method is guaranteed free of changes and claims. Accordingly, the project owner must look to a construction method most advantageous to its own goals and limitations rather than theoretical goals or limitations. Through timely notification and documentation of variation orders, participants will have kept their rights and thereby their option to pursue a subsequent claim or to defend against a claim (Ibbs et al., 1986; Cox, 1997; OBrien, 1998). The variations and variation orders should always be documented for future reference. A documented source of knowledge about previous variation orders would be helpful in making decisions concerning the appropriate handling of variation orders. Decision making is a significant characteristic that occurs in each phase of a project (Arain, 2005). Often, these decisions will, or can affect the other tasks that will take place. To achieve an effective decision making process, project managers and the other personnel of one project need to have a general understanding of other related or similar past projects (CII, 1994b). This underscores the importance of having a good communication and documentation system for better and prompt decision making during various project phases. If professionals have a knowledge-base established on past similar projects, it would assist the professional team to plan effectively before starting a project, during the design phase as well as during the construction phase to minimize and control variations and their effects (Miresco and Pomerol, 1995). From the outset, project strategies and philosophies should take advantage of lessons learned from past similar projects (Ibbs et al., 2001). It signifies the importance of an organized knowledge-base of similar past projects. The importance of a knowledge-base for better project control was recommended by many researchers (Miresco and Pomerol, 1995; Mokhtar et al., 2000; Gray and Hughes, 2001; Ibbs et al., 2001). As discussed above, it is therefore important to determine the potential causes, their relevant effects and possible controls for variations orders, and then to develop a knowledge-based system for management of variations in educational projects. The main objectives of this study are therefore to: a. Identify and examine the potential causes, their effects and controls for variations in educational building projects in Singapore.
b. Develop a knowledge-based decision support system for management of variations in educational building projects in Singapore.
10
4.0 Holistic View of Variation Orders

In a perfect world changes will be confined to the planning stages. However, late changes often occur during construction, and frequently cause serious disruption to the project (Cameron et al., 2004). In these circumstances, decisions are being made under pressure and cost and time invariably dominate the decision making process (OBrien, 1998; Arain, 2005). Most forms of contract for construction projects allow a process for variations. Even though there may be a process in place to deal with these late changes, cost and time invariably dominate the decision making process. If the change affects the design, it will impact on the construction process and, quite possibly, operation and maintenance as well (Cameron et al., 2004). To overcome the problems associated with changes to a project, the project team must be able to effectively analyze the variation and its immediate and downstream effects (CII, 1994b). An effective analysis of variations and variation orders requires a comprehensive understanding of the root causes of variations and their potential downstream effects. To manage a variation means being able to anticipate its effects and to control, or at least monitor the associated cost and schedule impact (Hester et al., 1991). Hence the structure of management of variation orders presented in this paper includes three main sections i.e., causes of variation orders, effects of variation orders and controls for variation orders.
4.1 Potential Causes of Variations

An effective analysis of variations and variation orders requires a comprehensive understanding of the root causes of variations (Hester et al., 1991) and 53 causes of variation orders were identified. As shown in Figure 1, the causes of variations were grouped under four categories: owner related variations, consultant related variations, contractor related variations and other variations. These groups assisted in developing a comprehensive enumeration of the potential causes of variations. Causes of variation orders have been identified by many researchers (CII, 1990a; Thomas and Napolitan, 1994; Clough and Sears, 1994; Fisk, 1997; Ibbs et al., 1998; OBrien, 1998; Mokhtar et al., 2000; Gray and Hughes, 2001; Arain et al., 2004). The causes of variations can be categorized according to the originators (CII, 1990a; Thomas and Napolitan, 1994). The 53 causes identified from the literature review are discussed below. These will also form the basis for the survey of the professionals described later.
11
A. Owner Related Changes

This section discusses the causes of variations that were initiated by the owner. In some cases, the owner directly initiates variations or the variations are required because the owner fails to fulfill certain requirements for carrying out the project. Change of plans or scope by owner: Change of plan or scope of project is one of the most significant causes of variation in construction projects (CII, 1990b) and is usually the result of insufficient planning at the project definition stage, or because of lack of involvement of the owner in the design phase (Arain et al., 2004). This cause of variations affects the project severely during the later phases. Change of schedule by owner: A change of schedule during the project construction phase may result in major resource reallocation (Fisk, 1997; OBrien, 1998). Time has an equivalent money value. A change in schedule means that the contractor will either provide additional resources, or keep some resources idle. In both cases additional cost is incurred. Owners financial problems: The owner of the facility may run into difficult financial situations that force him to make changes in an attempt to reduce cost. Owners financial problems affect project progress and quality (Clough and Sears, 1994; OBrien, 1998). Proper planning and review of project cash flow would be effective in eliminating this problem. Inadequate project objectives: Inadequate project objectives are important causes of variation in construction projects (Ibbs and Allen, 1995). Due to inadequate project objectives, the designer would not be able to develop a comprehensive design which leads to numerous variations during the project construction phase. Replacement of materials or procedures: Replacement of materials or procedures may cause major variations during the construction phase. The substitution of procedures includes variations in application methods (Chappell and Willis, 1996). Therefore, an adjustment to the original contract value is required if there is a change in procedures. Impediment in prompt decision making process: Prompt decision making is an important factor for project success (Sanvido et al., 1992; Gray and Hughes, 2001). A delay in decision making may hinder subsequent construction activities that may eventually delay the project progress.
12
Causes of Variation Orders
A. Owner related variations
B. Consultant related variations
C. Contractor related variations
D. Other variations
Change of plans or scope by owner Change of schedule by owner Owners financial problems Inadequate project objectives Replacement of materials/procedures
Change in design by consultants Errors and omissions in design Conflicts between contract documents Inadequate scope of work for contractor
Lack of contractors involvement in design Unavailability of equipment Unavailability of skills
Weather condition Safety considerations Change in govt. regulations Change in economic conditions Socio-cultural factors
Contractors financial difficulties Contractors desired profitability Differing site conditions Defective workmanship Unfamiliarity with local conditions Lack of specialized construction manager Fast track construction Poor procurement process Lack of communication Contractors lack of judgment & experience Long lead procurement Honest wrong belief of contractor
Technology change
Impediment in prompt decision making process Obstinate nature of owner Change in specifications by owner
Value engineering
Unforeseen problems
Lack of coordination
Design complexity Inadequate working drawing details Inadequate shop drawing details Consultants lack of judgment and experience Lack of consultants knowledge of available materials and equipment Honest wrong belief of consultant Consultants lack of required data
Obstinate nature of consultant Complex design and technology Ambiguous design details Lack of strategic planning Contractors lack of required data Contractors obstinate nature
Design discrepancies (inadequate design) Non-compliance design with govt. regulations Non-compliance design with owners requirement Change in specifications by consultant
Figure 1 Causes of variation orders grouped under four categories
13
Obstinate nature of owner: A building project is the result of the combined efforts of the professionals. They have to work at the various interfaces of a project (Wang, 2000; Arain et al., 2004). If the owner is obstinate, he may not accommodate other creative and beneficial ideas. Eventually, this may cause major variations in the later stages and affect the project adversely. Change in specifications by owner: Changes in specifications are frequent in construction projects with inadequate project objectives (OBrien, 1998). In a multi-player environment like any construction project, change in specifications by the owner during the construction phase may require major variations and adjustments in project planning and procurement activities.
B. Consultant Related Variations

This section discusses the causes of variations that were initiated by the consultant. In some cases, the consultant directly initiates variations or the variations are required because the consultant fails to fulfill certain requirements for carrying out the project. Change in design by consultant: Change in design for improvement by the consultant is a norm in contemporary professional practice (Arain et al., 2004). The changes in design are frequent in projects where construction starts before the design is finalized (Fisk, 1997). Design changes can affect a project adversely depending on the timing of the occurrence of the changes. Errors and omissions in design: Errors and omissions in design are an important cause of project delays (Arain et al., 2004). Design errors and omissions may lead to loss of productivity and delay in project schedule (Assaf et al., 1995). Hence, errors and omissions in design can affect a project adversely depending on the timing of the occurrence of the errors. Conflicts between contract documents: Conflict between contract documents can result in misinterpretation of the actual requirement of a project (CII, 1986a). To convey complete project scope for participants, the contract documents must be clear and concise. Insufficient details in contract documents may adversely affect the project, leading to delay in project completion. Inadequate scope of work for contractor: In a multi-player environment like construction, the scope of work for all the players must be clear and unambiguous for successful project completion (Fisk, 1997; Arain et al., 2004). Inadequate scope of work for the contractor can cause major variations that may adversely affect the project, leading to changes in construction planning.
14
Technology change: Technology change is a potential cause of variations in a project. Project planning should be flexible for accommodating new beneficial variations (CII, 1994b). This is because the new technology can be beneficial in the project life cycle, for instance, reducing maintenance cost of the project. Value engineering: Value engineering should ideally be carried out during the design phase (DellIsola, 1982). During the construction phase, value engineering can be a costly exercise, as variation in any design element would initiate down stream variations to other relevant design components (Mokhtar et al., 2000). Lack of coordination: A lack of coordination between parties may cause major variations that could eventually impact the project adversely (Arain et al., 2004). Detrimental variations, which affect the projects adversely, can usually be managed at an early stage with due diligence in coordination. Design complexity: Complex designs require unique skills and construction methods (Arain et al., 2004). Complexity affects the flow of construction activities, whereas simple and linear construction works are relatively easy to handle (Fisk, 1997). Hence, complexity may cause major variations in construction projects. Inadequate working drawing details: To convey a complete concept of the project design, the working drawings must be clear and concise (Geok, 2002). Insufficient working drawing details can result in misinterpretation of the actual requirement of a project (Arain et al., 2004). Thorough reviewing of design details would assist in minimizing variations. Inadequate shop drawing details: Shop drawings are usually developed for construction work details for site professionals (Cox and Hamilton, 1995). As mentioned earlier with regard to working drawing details, likewise, inadequacy of shop drawing details can be a potential cause of variations in the construction projects. Consultants lack of judgment and experience: Professional experience and judgment is an important factor for successful completion of a building project (Clough and Sears, 1994; OBrien, 1998). The lack of professional experience increases the risk of errors in design as well as during construction. Eventually, this may affect the project quality and delay the project completion. Lack of consultants knowledge of available materials and equipment: Knowledge of available materials and equipment is an important factor for developing a comprehensive design (Geok, 2002). In the construction industry where material standardization is not common, the consultants lack of knowledge of available materials and equipment can cause numerous major variations during various project phases.
15
Honest wrong beliefs of consultant: Honest wrong beliefs may cause construction professionals to contribute poor value add in projects (Arain, 2002; Arain et al., 2004). Consultants, without having first hand knowledge, may make decisions based on their wrong beliefs which would adversely affect the pace of the project. Consultants lack of required data: A lack of data can result in misinterpretation of the actual requirements of a project (Assaf et al., 1995; Arain, 2002). Where there is insufficient data, consultants are compelled to develop designs based on their own perceptions, which may not be what the client wants. Eventually, this may cause major variations and affect the project adversely. Obstinate nature of consultant: In a multi-player environment like construction, the professionals have to work as team at the various interfaces of a project (Wang, 2000; Arain et al., 2004). If the consultant is obstinate, he may not accommodate other creative and beneficial ideas. Eventually, this may cause major variations in the later stages and affect the project adversely. Ambiguous design details: A clearer design tends to be comprehended more readily (OBrien, 1998). Ambiguity in design is a potential cause of variations in a project. This is because ambiguity in design can be misinterpreted by project participants, leading to rework and delay in the project completion. Eventually, this may affect the project adversely. Design discrepancies (inadequate design): Inadequate design can be a frequent cause of variations in construction projects (CII, 1990a; Fisk, 1997). Design discrepancies affect the project functionality and quality. Eventually, this can affect a project adversely depending on the timing of the occurrence of the variations. Noncompliance of design with government regulations: Noncompliance of design with government regulations would render the project difficult to execute (Clough and Sears, 1994). Noncompliance with government regulations may affect the project safety and progress adversely, leading to serious accidents and delays in the project completion. Noncompliance of design with owners requirements: A comprehensive design is one that accommodates the owners requirements (Cox and Hamilton, 1995). A noncompliance design with the owners requirements is considered an inadequate design (Fisk, 1997). Eventually, this may cause variations for accommodating the owners requirements. This may affect the project adversely during the construction phase. Change in specifications by consultant: Changes in specifications are frequent in construction projects with inadequate project objectives (OBrien, 1998). As mentioned
16
earlier with respect to changes in specifications by the owner, this is also a potential cause of variations in a project, leading to reworks and delays in the project completion.
C. Contractor Related Variations

This section discusses the causes of variations that were related to the contractor. In some cases, the contractor may suggest variations to the project, or the variations may be required because the contractor fails to fulfill certain requirements for carrying out the project. Lack of contractors involvement in design: Involvement of the contractor in the design may assist in developing better designs by accommodating his creative and practical ideas (Arain et al., 2004). Lack of contractors involvement in design may eventually cause variations. Practical ideas which are not accommodated during the design phase will eventually affect the project adversely. Unavailability of equipment: Unavailability of equipment is a procurement problem that can affect the project completion (OBrien, 1998). Occasionally, the lack of equipment may cause major design variations or adjustments to project scheduling to accommodate the replacement. Unavailability of skills (shortage of skilled manpower): Skilled manpower is one of the major resources required for complex technological projects (Arain et al., 2004). Shortage of skilled manpower is more likely to occur in complex technological projects. This lack can be a cause for variations that may delay the project completion. Contractors financial difficulties: Construction is a labour intensive industry. Whether the contractor has been paid or not, the wages of the worker must still be paid (Thomas and Napolitan, 1994). Contractors financial difficulties may cause major variations during a project, affecting its quality and progress. Contractors desired profitability: Contractors desired profitability can be a potential cause of variations in construction projects. This is because variations are considered a common source of additional works for the contractor (OBrien, 1998). The contractor may eventually strive to convince the project owner to allow certain variations, leading to additional financial benefits for him. Differing site conditions: Differing site condition can be an important cause of delays in large building projects (Assaf et al., 1995). The contractor may face different soil conditions than those indicated in the tender documents. Eventually this may affect his cost estimates and schedule adversely.
17
Defective workmanship: Defective workmanship may lead to demolition and rework in construction projects (Fisk, 1997; OBrien, 1998). Defective workmanship results in low quality in construction projects (Arain et al., 2004). Eventually, this cause may affect the project adversely, leading to rework and delay in the project completion. Unfamiliarity with local conditions: Familiarity with local conditions is an important factor for the successful completion of a construction project (Clough and Sears, 1994). If the contractor is not aware of local conditions, it would be extremely difficult for him to carry out the project. Eventually, project delays may occur that end up with vital variations in the entire design entity. Lack of a specialized construction manager: The construction manager carries out the construction phase in an organized way to eliminate the risks of delays and other problems. Lack of a specialized construction manager may lead to defective workmanship and delay in the construction project. Fast track construction: Fast track construction requires an organized system to concurrently carry out interdependent project activities (Fisk, 1997). When the public and private sectors have large funds and want to complete projects in a very short time, complete plans and specifications may not be available when the contractor starts work (Arain et al., 2004). Eventually, this procurement mode may cause major variations. Poor procurement process: Procurement delays have various adverse effects on other processes in the construction cycle (Fisk, 1997). Occasionally, the procurement delay may cause an entire change or replacement for originally specified materials or equipment for the project (Arain et al., 2004). This may therefore cause a need for project activities to be reworked. Lack of communication: Detrimental variations, which affect the projects adversely, can usually be managed at an early stage with strong and incessant communication. A lack of coordination and communication between parties may cause major variations that could eventually impact the project adversely (Arain et al., 2004). Contractors lack of judgment and experience: The consultants lack of professional experience increases the risk of errors during construction (OBrien, 1998). This lack may cause major construction variations in a project. Eventually, this may affect the project quality and delay the project completion. Long lead procurement: Procurement delays have various adverse affects on other processes in the construction cycle (Fisk, 1997). Occasionally, the procurement delay may cause an entire change or replacement for originally specified materials or equipment for the
18
project. Delay in long lead procurement is a common cause of delays in building projects (Assaf et al., 1995). Honest wrong beliefs of contractor: As mentioned earlier with respect to honest wrong beliefs of the consultant, honest wrong beliefs of the contractor can also be a potential cause of variations in construction projects. Contractors, without having first hand knowledge, may make decisions based on their wrong beliefs which would adversely affect the quality and pace of the project. Complex design and technology: Complex design and technology require detailed interpretations by the designer to make it comprehensible for the contractor (Arain, 2002). A complex design may be experienced for the first time by the contractor. Eventually, the complexity may affect the flow of construction activities, leading to delays in the project completion. Lack of strategic planning: Proper strategic planning is an important factor for successful completion of a building project (Clough and Sears, 1994; CII, 1994a). The lack of strategic planning is a common cause of variations in projects where construction starts before the design is finalized, for instance, in concurrent design and construction contracts (OBrien, 1998). Contractors lack of required data: A lack of required data may affect the contractors strategic planning for successful project completion, leading to frequent disruptions during the construction process. This is because a lack of data can result in misinterpretation of the actual requirements of a project (Assaf et al., 1995; Arain et al., 2004). Contractors obstinate nature: As mentioned earlier with regard to the obstinate nature of consultant, likewise, this can be a potential cause of variations in construction projects. If the contractor is obstinate, he may not accommodate creative and beneficial ideas suggested by others. Eventually, this may cause major variations in the later stages and affect the project adversely.
D. Other Variations
This section discusses the causes of variations that were not directly related to the participants. Weather conditions: Adverse weather conditions can affect outside activities in construction projects (Fisk, 1997; OBrien, 1998). When weather conditions vary, the contractor needs to adjust the construction schedule accordingly. Occasionally, this may affect the project progress adversely, leading to delays in construction.
19
Safety considerations: Safety is an important factor for the successful completion of a building project (Clough and Sears, 1994). Noncompliance with safety requirements may cause major variations in design. Lack of safety considerations may affect the project progress adversely, leading to serious accidents and delays in the project completion. Change in government regulations: Local authorities may have specific codes and regulations that need to be accommodated in the design (Arain et al., 2004). Change in government regulations during the project construction phase may cause major variations in design and construction. This can affect a project adversely depending on the timing of the occurrence of the changes. Change in economic conditions: Economic conditions is one of the influential factors that may affect a construction project (Fisk, 1997). The economic situation of a country can affect the whole construction industry and its participants. Eventually, this may affect the project adversely, depending on the timing of the occurrence of the variations. Socio-cultural factors: Professionals with different socio-cultural backgrounds may encounter problems due to different perceptions, and this may affect the working environment of the construction project (Arain et al., 2004). Lack of coordination is common between professionals with different socio-cultural backgrounds (OBrien, 1998). Eventually, project delays may occur that end up with vital changes in the entire project team. Unforeseen problems: Unforeseen conditions are usually faced by professionals in the construction industry (Clough and Sears, 1994; OBrien, 1998). If these conditions are not solved spontaneously, they may cause major variations in the construction projects. Eventually, this may affect the project adversely, leading to reworks and delays in the project completion.
4.2 Potential Effects of Variations

Effects of variations were observed by many researchers (CII, 1986a; CII, 1990a; Clough and Sears, 1994; CII, 1994a; Thomas and Napolitan, 1995; Fisk, 1997; Ibbs et al., 1998). The 16 effects identified from the literature review, as shown in Figure 2, are discussed below. These will also form the basis for the survey of the professionals described later. Progress is affected but without any delay: Variations during the project may affect the project progress and quality (CII, 1994a; Assaf et al., 1995). Time has an equivalent monetary value even if the professional team tries its best to keep the project completion schedule intact. However, only major variations during the project may affect the project completion time. The contractor would usually try to accommodate the variations by utilizing
20
the free floats in the construction schedules. Hence, the variations affect the progress but without any delay in the project completion.
Progress is affected but without any delay Increase in project cost Hiring new professionals Increase in overhead expenses Delay in payment Quality degradation Productivity degradation Procurement delay Rework and demolition Logistic delay Tarnish firms reputation Poor safety conditions Poor professional relations Additional payment for contractor Dispute among professionals Completion schedule delay
4.2 Effects of variation orders
Figure 2 Potential effects of variation orders Increase in project cost: The most common effect of variations during the construction phase, is the increase in project cost (CII, 1990a). Any major additions or alterations in the design may eventually increase the project cost (Clough and Sears, 1994; Assaf et al., 1995). In every construction project, a contingency sum is usually allocated to cater for possible variations in the project, while keeping the overall project cost intact. Hiring new professionals: Variations in complex technological projects may affect the project adversely (CII, 1995). Specialized manpower is one of the integral resources required for complex technological projects (Fisk, 1997). Depending on the nature, the
21
variations may occasionally require hiring new professionals or change in the entire project team. Increase in overhead expenses: Variations require processing procedures, paper work and reviews before they can even be implemented (OBrien, 1998). The process and implementation of variations in construction projects would increase the overhead expenses for all the participants concerned. Normally these overhead charges are provided for from the contingency fund allocated for the construction project. Delay in payment: Delay in payment occurred frequently due to variations in construction projects (CII, 1990a). Variations may hinder the project progress, leading to delays in achieving the targeted milestones during construction (CII, 1995). Eventually, this may affect payment to the contractors. Occasionally this delay may cause severe problems that end up in delays in payment to the subcontractors; this is because main contractors may not be able to pay the subcontractors unless they get paid by the owner first. Quality degradation: Variations, if frequent, may affect the quality of work adversely (Fisk, 1997). According to CII (1995), the quality of work was usually poor because of frequent variations because contractors tended to compensate for the losses by cutting corners. Productivity degradation: Interruption, delays and redirection of work that are associated with variation orders have a negative impact on labour productivity. These in turn can be translated into labour cost or monetary value (Ibbs, 1997b). Hester et al. (1991) argued that the productivity of workers was expected to be greatly affected in cases where they were required to work overtime for prolonged periods to compensate for schedule delays. Thomas and Napolitan (1995) concluded that variations normally led to disruptions and these disruptions were responsible for labour productivity degradation. The most significant types of disruptions were due to the lack of materials and information as well as the work out of sequence. Lack of material was reported as the most serious disruption. Hence, to manage variation, one needed to manage these disruptions. However, the disruptive effects could not be avoided in many instances. Procurement delay: Variations which are imposed when construction is underway may require revised procurement requests (OBrien, 1998). Procurement delays can be frequent due to variations that require new materials and specialized equipment. Hester et al. (1991) observed that procurement delays were common effects of variations related to new resources for construction projects. Rework and demolition: Rework and demolition are frequent occurrences due to variations in construction projects (Clough and Sears, 1994). Variations which are imposed when
22
construction is underway or even completed, usually lead to reworks and delays in project completion (CII, 1990a). Rework and demolition are potential effects of variations in construction, depending on the timing of the occurrence of the variations. These effects are to be expected due to variations during the construction phase. This is because the variations during the design phase do not require any rework or demolition on construction sites. Logistics delays: Logistics delays may occur due to variations requiring new materials and equipment (Fisk, 1997). Hester et al. (1991) observed that logistics delays were significant effects of variations in construction projects. Logistics delays were experienced in construction projects where variations in the construction phase required new materials, tools and equipments. Tarnish firms reputation: Variations are referred to as a major source of construction claims and disputes (Fisk, 1997; Kumaraswamy et al., 1998). The claims and disputes may affect the firms reputation adversely, leading to insolvency in severe cases. Variations also increase the possibility of professional disputes. Conventionally, variations present problems to all the parties involved in the construction process. Poor safety conditions: Variations may affect the safety conditions in construction projects (OBrien, 1998) as variations in construction methods, materials and equipment may require additional safety measures during the construction phase. Poor professional relations: Construction changes are a major source of construction dispute (Fisk, 1997). Eventually, variations may affect professional relations, leading to disputes. Clear procedures that are presented in the contract and fair allocation of risks can help in resolving disputes through negotiation rather than litigation (CII, 1986a). Additional payments for contractor: Additional payments for the contractor can be a potential effect of variations in construction projects. Variations are considered to be a common source of additional works for the contractor (OBrien, 1998). Due to additional payments, the contractor looks forward to variations in the construction project. Disputes among professionals: Like poor professional relations, disputes among professionals are also potential effects of frequent variations in construction projects. The disputes over variation orders and claims are inevitable and the variation clauses are often the source of project disputes (CII, 1986a). Clear procedures presented in the contract and fair allocation of risks can help in resolving disputes through negotiation rather than litigation (CII, 1986a). Frequent communication and strong coordination can assist in eliminating the disputes between professionals.
23
Completion schedule delay: Completion schedule delay is a frequent result of variations in construction projects (Ibbs, 1997a). The magnitude of the schedule being delayed due to variations was reported by Zeitoun and Oberlender (1993) to be 9% of the original schedule for 71 fixed price projects studied. Kumaraswamy et al. (1998) studied claims for extension of time due to excusable delays in Hong Kongs civil engineering projects. Their findings suggested that 50% of the projects surveyed were delayed because of variations.
4.3 Controls for Variation Orders

Controls for variations and variation orders have been suggested by many researchers (Mokhtar et al., 2000; Ibbs et al., 2001). Discussed below are 30 controls identified from a literature review. These will also form the basis for the survey of the professionals later. The controls were grouped under three categories: Design stage, Construction stage and DesignConstruction interface stage as shown in Figure 3. These groups assisted in developing a comprehensive enumeration of potential controls for variation orders.
A. Design Stage Controls for Variation Orders

Review of contract documents: Contract documents are the main source of information for any project. Comprehensive and balanced variation clauses would be helpful in improving coordination and communication quality (CII, 1994a). Conflicts between contract documents can result in misinterpretation of the actual requirement of a project. Freezing design: Variations in design can affect a project adversely depending on the timing of the occurrence of the changes. Therefore, freezing the design is a strong control method. Many owners freeze the design and close the door for variations after the completion of the drawings (CII, 1990a). However, this control requires that the design of the construction project should be comprehensive; otherwise, it may affect the project objectives adversely. Value engineering at conceptual phase: During the design phase, value engineering can be a cost saving exercise, as at this stage, variation in any design element would not require rework or demolition at the construction site. Value engineering at the conceptual stage can assist in clarifying project objectives and reducing design discrepancies (DellIsola, 1982). Involvement of professionals at initial stages of project: Involvement of professionals in design may assist in developing better designs by accommodating their creative and practical ideas (Arain et al., 2004). This practice would assist in developing a comprehensive design with minimum discrepancies (OBrien, 1998). Practical ideas that are not accommodated during the design phase may affect the project adversely. Variation during
24
the construction phase is a costly activity as it may initiate numerous changes to construction activities.
Review of contract documents Freezing design Value engineering at conceptual phase Involvement of professionals at initial stages of project Owner involvement at planning and design phases Involvement of contractor at planning and scheduling process Thorough detailing of design Clear and thorough project brief Reducing contingency sum
A. Design stage
Controls for variation orders
Clarity of variation order procedures Written approvals Variation order scope Variation logic and justification Project manager from an independent firm to manage the project Restricted pre-qualification system for awarding projects Owners involvement during construction phase Avoid the use of open tendering Use of project scheduling techniques Comprehensive documentation of VO
B. Construction stage
Prompt approval procedures Ability to negotiate variation Valuation of indirect effects Team effort by owner, consultant and contractor to control variation orders Utilize work breakdown structure Continuous coordination and direct communication Control the potential for variation orders to arise through contractual clauses Comprehensive site investigation Use of collected and organized project data compiled by owner, consultant and contractor (share database) Knowledge-base of previous similar projects Comprehensive analysis and prompt decision making through computerized knowledge-based decision support system
Figure 3 Controls for variation orders
C. Design-Construction interface stage
25
Owners involvement at planning and design phase: Involvement of the owner at the design phase would assist in clarifying the project objectives and identifying noncompliance with their requirements at the early stage (Fisk, 1997). Hence, this may help in eliminating variations during the construction stage where the impact of the variations can be severe. Involvement of contractor at planning and scheduling process: Involvement of the contractor at planning and scheduling may assist in developing better plans and schedules by accommodating practical ideas suggested by the contractor (Arain et al., 2004). Eventually, this may eliminate the major variations in the later stages of the construction project where the impact of the variations can be severe. Thorough detailing of design: A clearer design tends to be comprehended more readily (OBrien, 1998). This would also assist in identifying the errors and omissions in design at an early stage. Eventually, thorough detailing of design can eliminate variations arising from ambiguities and errors in design. Clear and thorough project brief: A clear and thorough project brief is an important control for variations in construction projects (OBrien, 1998) as it helps in clarifying the project objectives to all the participants. Eventually, this may reduce the design errors and noncompliance with the owners requirements. Reducing contingency sum: The provision of a large contingency sum may affect the participants working approaches. This is because the designer may not develop a comprehensive design and would consequently carry out the rectifications in design as variation orders during the later stages of the construction project. Therefore, reducing the contingency sum would be helpful in ensuring that the professionals carry out their jobs with diligence.
B. Construction Stage Controls for Variation Orders

Clarity of variation order procedures: Clarity of variation order procedures is an integral part of effective management of variation orders (Mokhtar et al., 2000). Early in the project life, the procedures should be identified and made clear to all parties. Clarity of variation order procedures would help in reducing the processing time and other mishandling issues (Ibbs et al., 2001). Written approvals: Any variation in the work that involves a change in the original price must be approved in writing by the owner before a variation order can be executed (CII, 1990a; Hester et al., 1991; Cox, 1997). Any party signing of behalf of the owner must have written authorization from the owner. It is difficult to prove the right for compensation if there is no such authorization from the owner. In the hectic environment of construction, many
26
verbal agreements can be forgotten, leaving the contractor without any legal proof to get compensation for the variations. Variation order scope: A well defined scope can assist the professional team in recognizing and planning appropriately to minimize the negative impact of the variation (Ibbs et al. 2001). The original scope should be clear and well defined to distinguish between a variation of scope and a variation due to design development. CII (1994b) pointed out that a common disagreement between parties in a project was about defining the variation scope. Thus, the effective definition of the scope of work is of paramount importance to identify and manage variations. Variation logic and justification: Variation logic and justification for implementation was one of the principles of effective change management proposed by Ibbs et al. (2001). This principle required a change to be classified as required or elective. Required changes were required to meet original objectives of the project while elective changes were additional features that enhanced the project. Knowing the logic and justification behind the proposed variations assists the professionals in promoting beneficial variations and eliminating detrimental variations. Project manager from an independent firm to manage the project: Involvement of a project manager from an independent firm would assist in eliminating variations that arise due to the lack of coordination among professionals (Arain et al., 2004). This practice may assist in reducing design discrepancies through early reviews of the contract documents and drawings. Restricted pre-qualification system for awarding projects: A restricted pre-qualification system for awarding projects would act as a filter to select only the capable parties for project bids (Chan and Yeong, 1995; Fisk, 1997). However, the lack of a restricted pre-qualification system may allow incapable parties to bid. This may eventually lead to numerous problems in the later stages of a construction project. Owners involvement during construction phase: Involvement of the owner during the construction phase would assist in identifying noncompliance with the requirements and in approving the variations promptly (Ibbs et al., 2001). Eventually, the involvement of the owner during the construction phase may keep him aware of ongoing activities and assist in prompt decision making. Avoid use of open tendering: Competitive open tendering usually encourages the main contractor to price very low to win the contract, especially in bad times when they are in need of jobs. This practice would give rise to the contractor trying to claim more to compensate for
27
the low price (Chan and Yeong, 1995). Avoiding the use of open tender would assist in eliminating the risks of unfair bids. This may eventually help in eliminating variations that may arise due to the contractors bidding strategy. Use of project scheduling/management techniques: To manage a variation means being able to anticipate its effects and to control, or at least monitor, the associated cost and schedule impact (Hester et al., 1991). The most known scheduling techniques in the construction industry are CPM, PERT and Gantt chart (Clough and Sears, 1994). These techniques are helpful in identifying the downstream effects of any variations on subsequent construction activities (Mokhtar et al., 2000). Eventually, these may assist in eliminating detrimental variations. Comprehensive documentation of variation order: Through timely notification and documentation of variation orders, participants will have kept their rights and thereby their option to pursue a subsequent claim or to defend against a claim (Cox, 1997; OBrien, 1998). One of the most aggravating conditions is the length of time that elapses between the time when a proposed contract modification is first announced and when the matter is finally rejected or approved as a variation order (Fisk, 1997). Cox (1997) suggested that the documentation of variation orders and claims had assisted in tracking the effects of the variation and claim events on time and cost. A documented source of knowledge about previous variation orders would be helpful in making decisions concerning the appropriate handling of variation orders.
C. Design-Construction Interface Stage Controls for Variation Orders

Prompt approval procedures: One of the most aggravating conditions is the length of time that elapses between the time when a proposed contract modification is first announced and when the matter is finally rejected or approved as a variation order (Fisk, 1997). However, the longer the period between recognition and implementation, the more costly the change will be. Hence, prompt approval procedures would assist in reducing the adverse effects of variations in the construction project. Ability to negotiate variation: Ability to negotiate variation is an important factor for the effective control of variation orders (Clough and Sears, 1994). Effective negotiation can assist the professional team in minimizing the negative impacts of the variation (Cushman and Butler, 1994). There are certain skills required for effective negotiation of variation orders, i.e., the knowledge of contract terms, project details, technology, labour rates, equipment, methods and communication skills.
28
Valuation of indirect effects: Consequential effects can occur later in the downstream phases of a project. Therefore, it is essential to acknowledge this possibility and establish the mechanism to evaluate its consequences (Ibbs et al., 2001). Indirect effects of variations can be substantial in the downstream phases of a complex project (Fisk, 1997). Professionals should thus evaluate the total overall effects a change may have on the downstream phases of a project, to manage the variation order effectively. Team effort by owner, consultant and contractor to control variation orders: Coordination is important in a multi-participant environment as in most construction projects (CII, 1994a; Assaf et al., 1995). Detrimental variations, which affect the projects adversely, can usually be managed at an early stage with due diligence in coordination. Utilize work breakdown structure: A work breakdown structure (WBS) is a management tool for identifying and defining work (Hester et al., 1991; Mokhtar et al., 2000). A contractor should consider using the WBS as an evaluation tool, especially on large projects. If a variation involves work not previously included in the WBS, it can be logically added to the WBS and its relationship with the other WBS element can be easily checked. Ripple effects can also be traced by the use of WBS (Hester et al., 1991). Continuous coordination and direct communication: Coordination and communication are important in a multi-participant environment as in most construction projects (Assaf et al., 1995). Detrimental variations, which affect the projects adversely, can usually be managed at the early stage with due diligence in coordination, and frequent communication. Control the potential for variation orders to arise through contractual clauses: Selection of the appropriate contract form with the necessary and unambiguous variation clauses would be helpful in the management of variation orders (Cox, 1997). Shifting risks and improved communication channels could result from properly prepared variation clauses (CII, 1990a). Clear procedures presented in the contract and fair allocation of risks can help in resolving disputes through negotiation rather than litigation. Comprehensive site investigation: Comprehensive site investigations assist in proper planning for construction activities (Fisk, 1997). As mentioned earlier, differing site conditions are an important cause of delays in large building projects (Assaf et al., 1995). Therefore, a comprehensive site investigation would help in reducing potential variations in a project. Use of collected and organized project data compiled by owner, consultant and contractor: The variation orders should always be documented for future references (Fisk, 1997). In a research study by CII (1994b) on the control of project changes, the research
29
team concluded that better controls for variation orders were achievable by sharing a database compiled by all the participants. Knowledge-base of previous similar projects: A knowledge-base would facilitate an effective management process (CII, 1994b; Miresco and Pomerol, 1995; Ibbs et al., 2001). From the outset, project strategies and philosophies should take advantage of lessons learned from past similar projects (CII, 1994b). If professionals have a knowledge-base established on past similar projects, it would assist the professional team to plan effectively before starting a project, both during the design phase as well as during the construction phase, minimize and control variations and their effects. Comprehensive analysis and prompt decision making through computerized knowledge-based decision support system: A Decision Support System (DSS) approach for management decisions seems to be the most natural idea to follow (Miresco and Pomerol, 1995). The knowledge-based system would be helpful in presenting a comprehensive scenario of the causes of variations, their relevant effects and potential controls that would assist in decision making at the early stage of the variations occurring.
30
5.0 Scope of Research

The government of Singapore initiated a major program of rebuilding and improving existing educational buildings to ensure that the new generation of Singaporeans would get the best opportunities to equip them with the information technology (IT) available. The new and upgraded facilities in these educational buildings will include computer laboratories, media resource libraries, IT learning resource rooms, pastoral care rooms and health and fitness rooms. The occupants can also look forward to bigger classrooms and staff-rooms, as well as more interaction areas. A total of about 290 educational buildings will be upgraded or rebuilt by a government agency over a period of seven years, at an estimated cost of $4.46 billion from 1999 to 2005 (Note: at the time of writing , US$ 1 is about S$ 1.80). The projects are of three types, namely, upgrade, rebuild, and grass root (new) buildings. This is a timely study as the major programme of rebuilding and improving is currently under way. It is important to assess the causes, their relevant effects and the potential controls for variation orders for educational building projects. Developing a knowledge-based decision support system for management of variation orders for educational building projects will contribute towards the better control of variation orders through prompt and more informed decisions. Therefore this research concentrated on the educational building projects under this major rebuilding and improvement programme in Singapore. Furthermore, the survey was restricted to the developers (governmental agency), the consultants and contractors who have carried out these educational projects.
31
6.0 Research Methodology

To develop the knowledge-base decision support system for management of variation orders for educational building projects in Singapore, a case study approach and questionnaire survey were carried out. Information for the study was obtained from source documents of the educational projects completed, and through personal interviews and in-depth discussions with the professionals with a government agency responsible for the rebuilding and improvement programme, the consultants and the contractors who were involved in the educational projects. A case study approach encompassing 79 educational building projects was carried out in Singapore to collect the information required for in-depth study and analysis. The projects were documented and analyzed between February to September 2004. The purpose of the case study approach was to obtain data from the source documents of the completed projects. The source documents included the contract documents, variation orders documents, contract drawings and as-built drawings. Through the above literature review, 53 causes of variation orders were identified, together with 16 potential effects and 30 controls for variation orders. These provided the basis for the formulation of a questionnaire which was restricted to the professionals who were involved in the educational building projects under the rebuilding and improving programme in Singapore. With these parameters in mind, the target population of 35 developers, 82 consultants, and 61 contractors in Singapore were identified. The sample size of the required each population was determined statistically (Kish, 1995). n0 = (p*q)/ V2 . (1)
n = n0 / [1+ (n0 / N)] .. (2) Where: n0: First estimate of sample size p: The proportion of the characteristic being measured in the target population q: Complement of p or 1-p V: The maximum standard error allowed N: The population size n: The sample size To maximize n, p was set at 0.5. The target populations, N were 35, 82, and 61 for the developers, consultants and contractors respectively. To account for possible error in the qualitative answers from the questionnaire, the maximum standard error V was set at 10% or 0.1. Substituting in Equations 1 and 2 above, the minimum required samples were
32
calculated to be 14.58, 19.16, and 17.70 for the developers, consultants and contractors respectively. This means that the minimum sample size of 15, 19 and 18 for the developers, consultants and contractors respectively, is statically acceptable for analysis of the responses. A survey of 178 professionals, who have carried out the educational projects under the rebuilding and improvement programme in Singapore, was carried out. They included directors, senior managers, project managers and project officers from the developers side, directors, principal architects, senior architects and project architects from the consultants side, and directors, senior project managers, project managers and construction managers from the contractors side. A 5-point likert scale was used in the questionnaire to gauge the most important causes, their effects and controls for variation orders for the educational building projects in Singapore. In addition to collecting information from the source documents and sending out the questionnaires, 62 face-to-face interviews using the questionnaire and the collected data were also conducted to ensure that all questions were answered, that the information was accurate and the respondents have a chance to clarify any doubts with the research team. Interviews of 28 professionals with the government agency responsible for the rebuilding and improvement programme, 16 consultants and 18 contractors, who were involved in these educational projects, were carried out. They included directors, senior managers, project managers and project officers from the developers side, directors, principal architects, senior architects and project architects from the consultants side, and directors, senior project managers, project managers and construction managers from the contractors side.
33
7.0 Background of Respondents

The survey packages i.e., the final questionnaire along with a covering letter stating the main objectives of the questionnaire, and a self addressed stamped envelope, were sent out in May 2004. Responses were received between May and July 2004. The survey packages were sent to the 178 professionals. They included 31 developers, 82 consultants and 61 contractors who carried out the educational projects under the rebuilding and improvement programme. Of the 178 professionals, 98 professionals responded to the survey. 29 (82.86%), 36 (43.90%), and 33 (54.10%) responses were received from developers, consultants and contractors respectively. After checking though the completed questionnaires, 92 questionnaires were found to be suitable for data analysis. This yielded a response rate of about 51.69%. Table 1 shows the details of the responses. Table 1: Survey response rates
Respondents Developers Consultants Contractors Total Questionnaires sent 35 82 61 178 Responses received 29 36 33 98 Percentage 82.86% 43.90% 54.10% 55.06% Valid responses 28 33 31 92 Percentage 80.00% 40.24% 50.82% 51.69%
Table 2 shows the detailed breakdown of the respondents from the developers side. Table 2: Developers response to the survey
Respondents Appointments Directors Senior Project Managers Developers Senior Development Officers Project Managers Project Officers Total Responses received 2 4 2 8 12 28 Percentage 7.14% 14.28% 7.14% 28.57% 42.85%
34
Table 3 shows the detailed breakdown of respondents from the consultants. Table 3: Consultants response to the survey
Respondents Appointments Directors Principal Architects Consultants Senior Architects Project Architects Total 11 9 33 33.33% 27.27% Responses received 5 8 Percentage 15.15% 24.24%
Table 4 shows a detailed breakdown of respondents, from the contractors side. Table 4: Contractors response to the survey
Respondents Appointments Directors Senior Project Managers Contractors Project Managers Construction Managers Total 9 6 31 29.03% 19.35% Responses received 2 14 Percentage 6.45% 45.16%
As all the respondents were involved with the educational projects under the rebuilding and improvement programme and professionally positioned at management level or higher, a certain level of accuracy in the data collected was also assured.
35
8.0 Analysis of Results

The questionnaire listed 53 causes, 16 effects and 30 controls for variations orders for educational buildings in Singapore. Each respondent was asked to rate each issue based on his/her professional judgment. The causes of variation orders were analyzed and ranked according to their responses. Table 5: Mean and standard deviation of causes of variations
S No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Causes Change of plans or scope by owner Change of schedule by owner Owners financial problems Inadequate project objectives Replacement of materials or procedures Impediment in prompt decision making process Obstinate nature of owner Change in specifications by owner Change in design by consultant Errors and omissions in design Conflicts between contract documents Inadequate scope of work for contractor Technology change Value engineering Lack of coordination Design complexity Inadequate working drawing details Inadequate shop drawing details Consultants lack of judgment and experience Lack of consultants knowledge of available materials and equipment Honest wrong belief of consultant Consultants lack of required data Obstinate nature of consultant Ambiguous design details Mean 3.40 2.61 1.88 2.43 2.68 2.46 1.91 3.49 3.14 3.53 3.22 2.97 2.26 2.50 3.15 2.65 3.13 2.87 2.73 2.54 2.30 2.61 2.07 3.02 Std. Dev. 1.12 1.26 1.01 1.51 0.97 0.89 0.93 1.19 1.12 1.14 1.15 1.35 0.94 1.11 1.19 1.04 1.17 1.11 1.05 1.24 0.92 1.25 0.89 1.12
25
Design discrepancies (Inadequate Design)
3.36
1.21
36
Table 5: Mean and standard deviation of causes of variations

S No. 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 Causes Noncompliance design with govt. regulations Noncompliance design with owners requirement Change in specifications by consultant Lack of contractors involvement in design Unavailability of equipment Unavailability of skills Contractors financial difficulties Contractors desired profitability Differing site conditions Defective workmanship Unfamiliarity with local conditions Lack of specialized construction manager Fast track construction Poor procurement process Lack of communication Contractors lack of judgment & experience Long lead procurement Honest wrong belief of contractor Complex design and technology Lack of strategic planning Contractors lack of required data Contractors obstinate nature Weather conditions Safety considerations Change in government regulations Change in economic conditions Socio-cultural factors Unforeseen problems Mean 3.01 2.84 3.03 2.88 2.23 2.24 2.59 2.71 3.27 2.83 2.13 2.25 2.64 2.42 2.91 2.71 2.54 2.32 2.27 2.71 2.53 2.05 3.03 3.15 3.04 2.60 2.21 3.41 Std. Dev. 1.24 1.08 1.09 1.34 1.00 1.00 1.03 1.08 1.15 1.02 1.02 1.13 1.13 1.01 1.08 1.03 1.03 0.99 0.95 1.01 1.02 0.99 1.17 1.00 1.06 0.84 0.79 1.07
As shown in Table 5, 53 causes of variation orders were tabulated according to their means and standard deviations. To ascertain whether the rankings of the 53 causes by the
37
developers, the consultants and the contractors were correlated, Spearmans rank correlation was used. The Spearmans rank correlation results indicated that the ranking by the developers and the contractors were strongly correlated, nevertheless, the ranking by the developers and the consultants were not correlated. Furthermore, as shown in Table 6, the ranking by the consultants and the contractors were also not correlated. This was not unexpected because both the developers and the contractors pointed towards the consultants for initiating most of the causes of variations. Table 6: Spearmans rank correlation for causes of variations
Correlations Spearman's rho Developers Correlation Coefficient Sig. (2-tailed) Consultants Correlation Coefficient Sig. (2-tailed) Contractors Correlation Coefficient Sig. (2-tailed) N
** Correlation is significant at the .01 level (2-tailed). * Correlation is significant at the .05 level (2-tailed).
Developers 1 . 0.275* 0.046 0.743** 0 53
Consultants 0.275* 0.046 1 . 0.156 0.264
Contractors 0.743** 0 0.156 0.264 1 .
However, the consultants ranking of the causes indicated mostly contractor and developer related variations. Furthermore, the causes of variations in educational projects were categorized into the most important ones as shown in Table 7. Table 7: Most important causes of variations in educational building projects
S No. 10 8 53 1 25 Causes Errors and omissions in design Change in specifications by owner Unforeseen problems Change of plans or scope by owner Design discrepancies (Inadequate Design) Mean 3.53 3.49 3.41 3.40 3.36 Std. Dev. 1.14 1.19 1.07 1.12 1.21 Rank 1 2 3 4 5
The results suggest that the errors and omissions in design, change in specifications by owner, unforeseen problems, change in plans or scope by owner, and design discrepancies were considered to be the most important causes of variation orders for educational building projects in Singapore. It was revealed that of the top five most important causes of
38
variations, four causes were from both owner related variations and the consultant related variations groups. A section of the questionnaire listed 16 effects of variations orders for educational buildings in Singapore. Each respondent was asked to rate each issue based on his/her professional judgment. The list of effects of variation orders were analyzed and ranked according to their responses. The 16 effects of variation orders were tabulated according to their means and standard deviations as shown in Table 8. Table 8: Mean and standard deviation of effects of variation orders
S No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Effects Progress is affected but without any delay Increase in project cost Hiring new professionals Increase in overhead expenses Delay in payment Quality degradation Productivity degradation Procurement delay Rework and demolition Logistic delay Tarnish firms reputation Poor safety conditions Poor professional relations Additional payment for contractor Dispute among professionals Completion schedule delay Mean 3.39 3.89 2.02 3.29 3.09 2.49 2.80 2.92 3.26 2.91 2.23 2.28 2.18 3.35 2.52 3.25 Std. Dev. 1.03 1.00 0.99 1.36 1.51 1.06 1.09 1.00 1.18 0.91 0.95 1.10 0.89 1.10 0.92 1.09
The Spearmans rank correlation results of the 16 effects indicated that the ranking by all respondents were correlated as shown in Table 9. It shows that the professionals agreed on the potential effects of variations in educational building projects.
39
Table 9: Spearmans rank correlation for effects of variations

Correlations Spearman's rho Developers Consultants Contractors Correlation Coefficient Sig. (2-tailed) Correlation Coefficient Sig. (2-tailed) Correlation Coefficient Sig. (2-tailed) N Developers 1 . 0.868** 0 0.654** 0.006 16 Consultants 0.868** 0 1 . 0.7** 0.003 Contractors 0.654** 0.006 0.7** 0.003 1 .
The 16 potential effects of variations in educational building projects were categorized into the most frequent ones as shown in Table 10. Table 10: Most frequent effects of variation orders for institutional buildings
S No. 2 1 14 4 9 Effects Increase in project cost Progress is affected but without any delay Additional payment for contractor Increase in overhead expenses Rework and demolition Mean 3.89 3.39 3.35 3.29 3.26 Std. Dev. 1.00 1.03 1.10 1.36 1.18 Rank 1 2 3 4 5
The results present that project cost increase, progress is affected but without any delay, additional payment for contractor, overhead expenses increase and rework and demolition were considered to be the most frequent effects of variation orders for educational buildings in Singapore. The respondents rated the 30 controls for variation orders based on his/her professional judgment and these are tabulated according to their means and standard deviations in Table 11. The Spearmans rank correlation results indicated that the ranking by the consultants and the contractors were strongly correlated.
40
Leveraging on Information Technology for Effective Management of Variations in Educational Building Projects: A KBDSS approach
Table 11: Mean and standard deviation of controls for variation orders
S/No. Controls Mean Std. Dev.
Design stage 1 2 3 4 5 6 7 8 9 Review of contract documents Freezing design Value engineering at conceptual phase Involvement of professionals at initial stages of project Owner involvement at planning and design phases Involvement of contractor at planning and scheduling process Thorough detailing of design Clear and thorough project brief Reducing contingency sum 3.63 3.30 3.54 3.86 4.29 3.53 4.17 4.20 2.73 0.82 1.41 1.04 1.10 0.86 1.18 0.75 0.76 1.47
Construction stage 10 11 12 13 14 15 16 17 18 19 Clarity of variation order procedures Written approvals Variation order scope Variation logic and justification Project manager from an independent firm to manage the project Restricted pre-qualification system for awarding projects Owners involvement during construction phase Avoid the use of open tendering Use of project scheduling techniques Comprehensive documentation of VO 3.85 3.79 3.60 3.76 2.83 3.36 3.55 2.96 3.11 3.83 0.77 1.13 1.03 0.92 1.21 0.98 1.05 1.06 0.85 0.90
Design-Construction interface stage 20 21 22 23 24 25 26 Prompt approval procedures Ability to negotiate variation Valuation of indirect effects Team effort by owner, consultant and contractor to control variation orders Utilize work breakdown structure Continuous coordination and direct communication Control the potential for variation orders to arise through contractual clauses 3.90 3.32 3.43 4.09 3.36 4.13 3.50 1.03 0.95 0.96 0.82 0.90 0.70 0.99
41
Table 11: Mean and standard deviation of controls for variation orders
S/No. Controls Mean Std. Dev.
27 28 29 30
Comprehensive site investigation Use of collected and organized project data compiled by owner, consultant and contractor (share database) Knowledge-base of previous similar projects Comprehensive analysis and prompt decision making through computerized knowledge-based decision support system
3.79 3.64 3.95 3.61
0.82 0.81 0.88 0.98
The results as shown in Table 12 suggested that the ranking by the developers and the consultants were moderately correlated. Furthermore, the ranking by the developers and the contractors were also moderately correlated. Overall, the results of the raking correlation exercise suggested that the professionals however agreed on the effectiveness of the suggested controls for variations in educational building projects. Table 12: Spearmans rank correlation for controls for variations
Correlations Spearman's rho Developers Correlation Coefficient Sig. (2-tailed) Consultants Correlation Coefficient Sig. (2-tailed) Contractors Correlation Coefficient Sig. (2-tailed) N Developers 1 . 0.631** 0 0.559** 0.011 30 Consultants 0.631** 0 1 . 0.714** 0 Contractors 0.559** 0.011 0.714** 0 1 .
As shown in Table 11, the results indicated that the design stage was considered as the most important time-line for implementing the most effective controls for variations. A majority of controls that were ranked as very effective were from the design stage and design and construction interface stage categories. Furthermore, the controls for variation orders were also categorized according to their effectiveness as shown in Table 13.
42
Table13: Most effective controls for variation orders

S No. 5 8 7 25 23 Controls Owner involvement at planning and design phases Clear and thorough project brief Thorough detailing of design Continuous coordination and direct communication Team effort by owner, consultant and contractor to control variation orders Mean 4.29 4.20 4.17 4.13 4.09 Std. Dev. 0.86 0.76 0.75 0.70 0.82 Rank 1 2 3 4 5
The top five most effective controls were owners involvement at the planning and design phase, clear and thorough project brief, thorough detailings of design, continuous coordination and direct communication, and team effort by owner, consultant and contractor to control variation orders. The results indicated that the design stage and design and construction interface stages were considered as the most effective phases for implementing controls for minimizing the adverse impact of variations in educational building projects.
43
9.0 Discussion
Through the questionnaire survey and interviews with the professionals, the most important causes, their frequent effects and effective controls for variation orders for educational building projects were revealed and tabulated. The five most significant causes of variation orders for educational buildings are discussed below.
9.1 Most Significant Causes of Variation Orders

Errors and omissions in design The errors and omissions in design were ranked by the professionals as the most important cause of variation orders for educational buildings, as design errors and omissions may lead to loss of productivity and delay in the project schedule. These errors, if not rectified during the design phase, would eventually appear in the construction phase where the impact could be more severe than in the design phase. It was revealed through in-depth interviews with the professionals that during the early phases of the programme, large numbers of projects were awarded to consultants who did not have prior experience of educational buildings, and the time given for design development was not sufficient, thus leading to numerous errors and omissions in design. Hence, in order to reduce design errors and omissions, it is imperative that the professionals concentrate more on allocating sufficient time for design development and improving design detailings that would assist in reducing the design variations. Change in specifications by owner Change in specifications by owners was ranked as the second most important cause of variation orders. In a multi-player environment like construction, change in specifications by the owner during the construction phase may require major variations and adjustments in project planning and procurement activities. Such changes were frequent in educational projects with inadequate project objectives. Many problems were frequently faced during the initial phase of the rebuilding and improvement programme because of the changes in specifications, leading to frequent revisions of specifications during the construction phase. Unforeseen problems The third most important cause of variation orders was unforeseen problems. Unforeseen conditions are usually faced by professionals in the construction industry (Arain et al., 2004). Unforeseen conditions would render the project difficult to execute. If these conditions are not solved promptly, they may cause major variations in the construction projects. Eventually, this may affect the project adversely, leading to rework and delays in the project
44
programme. It was also revealed from the in-depth interview sessions that numerous variations were initiated because of unforeseen problems in the upgrading projects. This was not unexpected because there were numerous constraints in upgrading projects, due to existing structure and discrepancies between design and construction that were caused by nonconformity of as-built drawings and information. Change in plan or scope by owner Changes in plan or scope by the owner was perceived as the fourth most important cause of variation orders for educational buildings. Many problems were frequently faced during the initial phase of the rebuilding and improvement programme, as plans were not finalized by the owner, leading to frequent revisions of plans during the construction phase and significant rework. It was also revealed from the in-depth interview sessions with the professionals that a majority of the educational projects were completed during the initial phases of the programme of rebuilding and improvement, hence large numbers of design changes were expected, as during the initial phases of the programme, the project objectives were not very clear. Eventually, this may affect the project adversely, leading to rework and delays in the project programme. Design discrepancies The fifth most important cause of variation orders was design discrepancies which may affect the project functionality and quality. Eventually, this can affect a project adversely, depending on the timing of the occurrence of the variations. Inadequate design was a frequent cause of variations as the designs were not comprehensive and eventually the discrepancies were rectified through variation orders. Furthermore, it was also revealed through in-depth interviews with the professionals that the time allocated for the design process during the early phases of the programme was insufficient because a large number of projects were targeted during these phases. Hence, the design discrepancies were frequent, which was not unexpected. As mentioned earlier with regard to errors and omissions in design, it is likewise recommended that the professionals should concentrate more on improving design detailings and compliance with government regulations. This would assist in reducing variations due to design discrepancies. As shown in Table 10, the 16 potential effects of variations in educational building projects were categorized into the most frequent ones. These most frequent effects of variations are discussed below.
45
9.2 Most Frequent Effects of Variation Orders

Increase in project cost According to the survey findings, the most frequent effect of variation orders was the increase in project cost. It was not unexpected for the project cost to increase due to frequent variations in the project, as variation orders may affect the projects total direct and indirect costs. Therefore, any major addition or alteration in the design may eventually increase the project cost. In every construction project, a contingency sum is usually allocated to cater for possible variations in the project, while keeping the overall project cost intact. However, frequent major variations may lead to cost overrun in the contingency sum. Progress is affected but without any delay The second most frequent effect of variation orders was where progress is affected but without any delay. This was because the professional team usually strives to keep the project completion schedule intact because time has an equivalent monetary value. The contractors are usually compelled to accommodate the implementation time for variations by utilizing the free floats in the construction schedules. Hence, the variations affect the progress but without any delay in the overall project completion. It was revealed through the in-depth interviews with the professionals that in most of the cases the contractors agreed to carry out the variations without claiming for extension of time for the overall project schedule. Nevertheless, in some cases the progress was affected. Additional payment for contractor Additional payment for contractor was perceived as the third most frequent effect of variation orders. This was because variations are considered as a common source of additional works for the contractor. The contractors would consider variations in the project as additional opportunities to achieve their desired profit margins. This situation was frequently faced by the owner in projects where the terms for valuing the variations were not considered at the inception of the project. Increase in overhead expenses The fourth most frequent effect of variation orders for educational building projects was the increase in overhead expenses. This was because the process and implementation of variations in construction projects increased the overhead expenses for all the concerned participants. Normally these overhead charges are provided for from the contingency fund allocated for the construction project.
46
Rework and demolition Rework and demolition were considered as the fifth most frequent effect of variation orders. Any additions or alterations in design during the construction phase may result in reworks and demolitions on site. Furthermore, the reworks and demolitions may affect the subsequent construction activities, eventually leading to delays in the project completion. Therefore, the impact of a variation in design during the construction phase is more severe than in the design phase. It was revealed through in-depth interview sessions that during the initial phases of the rebuilding and improvement programme the reworks and demolitions were very frequent because the drawings were finalized and the specifications were frequently changed by the owner. The five most effective controls for variations in educational building projects are discussed below.
9.3 Most Effective Controls for Variations

Owners involvement at planning and design phases The involvement of the owner at the planning and design phases was perceived as the most effective control of variations. This was because the involvement of the owner in the design phase would assist in clarifying the project objectives and in identifying noncompliance with their requirements at an early stage. Eventually, this may help in eliminating the occurrence of variations during the construction stage where the impact of the variations can be more severe. The survey results indicated that the involvement of owner not only in the design phase but also in the construction phase is highly appreciated by the professionals. This was not unexpected because it provides a better opportunity for all parties to understand the actual requirements and design brief and to make prompt decisions during the project. It was revealed through in-depth interviews with the professionals that the owner was not involved at planning and design phases during the initial phases of the rebuilding and improvement programme. This initiated numerous variations during the construction phase where the impact of the variations was more severe than in the design phase. In most of the cases, the owner was able to indicate his intentions during the construction phase where he was able to view the actual output of the design. Furthermore, this initiated numerous reworks and demolitions in the projects. Hence, the professionals strongly recommended the involvement of owner at planning and design phases that would eventually assist in reducing potential variations in the construction projects.
47
Clear and thorough project brief The second most effective control of variations was a clear and thorough project brief which helps in clarifying the project objectives to all the participants. As mentioned earlier, the project brief was not clear and thorough during the initial phases of the rebuilding and improvement programme. Hence, many variations were encountered by the professionals because of an unclear and inadequate project brief. A clear and thorough project brief can eliminate variations that may arise due to unclear project objectives. Eventually, this may reduce the design errors and noncompliance with owners requirement. Thorough detailing of design Thorough detailing of design was ranked as the third most effective control of variations in educational building projects. This has always been an important control mechanism because it assists in reducing ambiguities and discrepancies in design. A clearer design tends to be comprehended more readily. Furthermore, thorough detailing of design was considered as an effective control of variation orders because it can assist in identifying the errors and omissions in design at an early stage. Eventually, it can eliminate the variations arising from ambiguities and errors in design. As mentioned in the previous section, the time allocated for design exercise during the initial phases of the programme was not sufficient, which eventually resulted in inadequate detailings of design. Hence, the professionals faced numerous variations due to the inadequate detailings of designs. Continuous coordination and direct communications Coordination is important in the multi-participant environment found in most construction projects. Detrimental variations, which affect the projects adversely, can usually be managed at an early stage with due diligence in coordination. Continuous coordination and direct communication was perceived as an effective control for variation orders in educational building projects. It was considered as the fourth most effective control of variation orders. This was because coordination and communication are integral for the successful completion of construction projects. These also assist in managing variations, which can affect the projects adversely, at an early stage where the impact of the variations would be less severe than during the construction phase. Team effort by owner, consultant and contractor to control variation orders According to the survey findings, team effort by owner, consultant and contractor was perceived as the fifth most effective control of variations. A delay in decision making may hinder subsequent construction activities that would eventually delay the project progress. Hence, team effort by all participants would assist in reducing the adverse effects of
48
variations in the construction project. As mentioned earlier, coordination is important in a multi-participant environment as in most construction projects. Variations, which affect the projects adversely, can usually be managed at an early stage with due diligence in coordination. Construction projects are bound to encounter variation orders; the goal of any owner, designer, or construction manager is to control the number of variations. Therefore, team effort by the participants would be helpful in reducing the adverse effects of variations when the project progresses. The survey findings are discussed above. The questionnaire responses were also used for revealing the most frequent effects and effective controls for each of the 53 causes of variations in educational building projects. Chan and Kumaraswamy (1997) used the relative importance index method. This method was also adopted to analyze the data colleted from the questionnaire survey. The analysis was carried out for all three groups of respondents. Firstly, the questionnaire responses were used for carrying out crosstabulation analyses between causes and effects, and between causes and controls. The cross-tabulation analyses assisted in identifying the important cores i.e., the causes and effects, and causes and controls that were considered important by the respondents. The number of responses that rated the causes and effects as important were extracted from the cross-tabulation analysis and used for developing the Relative Importance Index (RII). The RII method has been adopted by many researchers (Kometa et al., 1994; Aibinu and Jagboro, 2002) in earlier studies. The RII was calculated for each cause of variations as follows: RII =
2n1 + 1n2 2N
Where: n1 = number of respondents for very important n2 = number of respondents for important N = total number of respondents As shown in Table 14, (Appendix 1) the causes and their effects were tabulated according to their RII values. Likewise, the causes and their potential controls were also tabulated according to their RII values as shown in Table 15 (Appendix 1). These analyses assisted in identifying the most frequent effects and most effective controls for each cause of variation order. Furthermore, Figure 4 presents the most frequent effects and effective controls for the most important causes of variations that were identified in Table 7.
49
1) Errors and omissions in design
2) Change in specifications by owner
3) Unforeseen problems
4) Change of plans or scope by owner
5) Design discrepancies (inadequate design)
Increase in project costs MOST FREQUENT EFFECTS
Increase in project costs MOST FREQUENT EFFECTS
Delay in payment MOST FREQUENT EFFECTS
Progress is affected but without any delay MOST FREQUENT EFFECTS Increase in project costs Increase in overhead expenses Delay in payment Rework and demolition
Increase in project costs MOST FREQUENT EFFECTS MOST EFFECTIVE CONTROLS
Increase in overhead expenses Delay in payment
Increase in overhead expenses Delay in payment
Increase in overhead expenses Rework and demolition
Delay in payment
Increase in overhead expenses Rework and demolition
Rework and demolition
Progress is affected but without any delay Rework and demolition
Increase in project costs
Progress is affected but without any delay Additional payment for contractor
Progress is affected but without any delay
Completion schedule delay
Additional payment for contractor Owner involvement at planning and design process Owner involvement at planning and design process
Clear and thorough project brief MOST EFFECTIVE CONTROLS MOST EFFECTIVE CONTROLS Owner involvement at planning and design process MOST EFFECTIVE CONTROLS Owner involvement at planning and design process MOST EFFECTIVE CONTROLS Restricted pre-qualification system for awarding projects Comprehensive analysis and prompt decision making through computerized knowledge-based decision support system Avoid the use of open tendering
Knowledge-base of previous similar projects Clear and thorough project brief
Thorough detailing of design
Prompt approval procedures
Team effort by owner, consultant and contractor to control variation orders Clear and thorough project brief
Clear and thorough project brief
Knowledge-base of previous similar projects Team effort by owner, consultant and contractor to control variation orders Involvement of professionals at initial stages of project
Team effort by owner, consultant and contractor to control variation orders Knowledge-base of previous similar projects
Team effort by owner, consultant and contractor to control variation orders Freezing design (i.e., no changes after final design)
Clear and thorough project brief Owners involvement during construction phase
Knowledge-base of previous similar projects
Involvement of contractor at planning and scheduling process
Figure 4 Most important causes, their frequent effects and effective controls for variation orders
50
This will benefit the professionals involved with educational projects. The professionals would learn about the root causes of variation orders and their downstream effects that would assist in the proactive evaluation of variation orders. The comprehensive tabulation of the 53 causes and their frequent effects as shown in Table 14, and effective controls as shown in Table 15, assisted in developing the knowledge-based decision support system (KBDSS) which is presented in the following section.
51
10.0 Knowledge-Based Decision Support System (KBDSS)

A knowledge-based decision support system is a system that can undertake intelligent tasks in a specific domain that is normally performed by highly skilled people (Miresco and Pomerol, 1995). Typically, the success of such a system relies on the ability to represent the knowledge for a particular subject. Computerized decision support systems can be used by project participants to help make more informed decisions regarding the management of variation orders in projects by providing access to useful, organized and timely information. It is important to understand that the KBDSS for the management of variation orders is not designed to make decisions for users, but rather it provides pertinent information in an efficient and easy-to-access format that allows users to make more informed decisions. The architecture of the main components of the KBDSS is shown in Figure 5. The model contains two main components, i.e., a knowledge-base and a decision support shell, for selecting appropriate potential controls for variation orders for educational buildings. The database is developed through collecting data from source documents of the 79 educational projects, questionnaire survey, literature review and interview sessions with the professionals. The knowledge-base was developed through initial sieving and organization of data from the database. Furthermore, the knowledge-base was divided into three main segments namely, macro layer, micro layer, and effects and controls layer. The segment that contained information pertinent to possible effects and controls of the causes of variation orders for educational buildings was integrated with a decision support shell.
52
Questionnaire survey
Data from ongoing projects
Knowledge from literature
Interview with experts
Data base Data obtained from various sources Data from source documents of past similar projects
Knowledge acquisition Knowledge obtained through initial sieving and organization.
Knowledge-base (KB) Three main layers: Macro layer (level 1) Knowledge-base consists of major information about all the past projects. Micro layer (level 2) Detailed information of variation orders in a particular project. Effects/controls layer (level 3) Effects of a particular cause of variation and suggested solutions Suggested potential controls of causes Inference engine Categorizing by types Sieving information by rules Maintaining compatibility between model interfaces Calculating cost implications Calculating time implications Calculating frequency of variations and variation orders Calculating percentages
Software interface User interface Import export knowledge between KB and DSS
Decision support shell (DSS) Decision support through building the hierarchy among the main criteria and the suggested controls, rating the controls, and analyzing the controls for selection through multiple analytical techniques
User
Figure 5 Main components of the Knowledge-based decision support system The decision support shell provided decision support through a structured process consisting of building the hierarchy among the main criterions and the suggested controls, rating the
53
controls, and techniques. The KBDSS is developed in the MS Excel environment using numerous macros for developing the user-interface that carry out stipulated functions. These are incorporated within a decision support shell.
10.1 Knowledge-Base
The knowledge-base contains the sieved and organized information about the variations and variation orders for educational building projects in Singapore. The knowledge-base was divided into three main segments, namely, macro layer, micro layer and effects/controls layer. These three main segments of the knowledge-base are discussed below. Macro layer The first segment of the knowledge-base is the macro layer that consists of the information gathered from source documents of 79 educational projects and through interview sessions with the professionals. The macro layer contains such information as project name, program phase, work scope, institutional level, date of commencement, project duration, date of completion, actual completion, schedule completion status, schedule difference, contract final sum, contingency sum percent, contingency sum, contingency sum used, total number of variation orders, total cost of variation orders, total time implication, total number of variations, frequency of variation orders, frequency of variations, main contractors and consultants (see Figure 6, Appendix 2). The user interface allows the user to access, edit, modify, add and view the information displayed on the macro layer. To add new project information, the user needs to input the project name, program phase, work scope, institutional level, date of commencement, date of completion, actual completion, contract final sum, contingency sum percent, main contractor and consultant. The inference engine computes the project duration, schedule completion status, schedule difference, and contingency sum from the information given in the macro layer. The graphical user interface (GUI) assists users in interacting with the system on every level of the KBDSS. In addition, the GUI and inference engine will maintain the compatibility between layers and the decision shell. The GUI and inference engine create interface between the macro layer and the micro layer to retrieve the information about the total number of variation orders, total cost of variation orders, total time implication, total number of variations, frequency of variation orders and frequency of variations in each individual project. Furthermore, a variety of filters are provided on the macro layer that assists in sieving information by certain rules. The user would be able to apply multiple filters for analyzing the information by certain rules, for instance, the user would be able to view the
54
information about the educational projects that were completed behind schedule and among these, the projects with the highest frequency of variation orders, highest contingency sum used, highest number of variations, etc. This analysis assists the user in identifying the nature and frequency of variation orders in certain type of educational projects. The inference engine provides a comprehensive summary of the information available on the macro layer. As shown in Figure 7, (Appendix 2) the inference engine computes the total number of projects, subtotal (that assists in identifying the projects when multiple filters are applied), total number of projects based on program phases (P1, P2, P3), subtotal of projects based on program phases, total number of projects categorized according to work scope, subtotal of projects categorized according to work scope, total number of projects categorized based on institutional levels, subtotal of projects categorized based on institutional levels, total number of projects based on schedule completion status (ahead of schedule, on schedule, behind schedule), subtotal of projects based on schedule completion status, total number of projects based on three levels of contingency sum usage, subtotal of projects based on three levels of contingency sum usage, total number projects categorized based on time implications, and subtotal of projects based on time implications. Furthermore, the inference engine also computes the percentages for each category mentioned above and shown in Figure 7. This assists the user in analyzing and identifying the nature and frequency of variation orders in certain type of educational projects. Micro layer Information about the 79 educational projects were computed and documented on the macro layer as shown in Figure 6a, where the macro layer is integrated with the micro layer through the GUI. As shown in Figure 6b, the project names enumerated on the macro layer are included in the KBDSS query form that assists in accessing the micro layer. The micro layer is the second segment of the knowledge-base that contains 79 sub-layers based on the 79 educational projects respectively. The micro layer (Figure 8, Appendix 2) contains detailed information regarding variations and variation orders for the educational project, including the variation order code that assists in sieving information, detailed description of particular variation collected from source documents, reason for carrying out the particular variation provided by the consultant, root cause of variation, type of variation, cost implication, time implication, approving authority, and endorsing authority. Here, the information regarding the description of the particular variation, reason, type of variation, cost implication, time implication, approving authority, and endorsing authority were obtained from the source documents of the 79 educational projects. The root causes were determined based on the description of variations, reasons given by the consultants, and the project source
55
documents. The root causes were verified through the in-depth interview sessions with the developers, consultants and contractors later. The user interface provided at the micro layer allows the user to access, edit, modify, add and view the information. In this layer, the user may add new projects detailed information by inputting the description of a particular variation, reason, causes, type of variation, cost implication, time implication, approving authority, and endorsing authority. The inference engine provides drop-down lists for inputting information regarding the cause of variations, variation type, time implication, approving authority, preparation and endorsement. The inference engine provides a comprehensive summary of the information available on the micro layer. As shown in Figure 9, Appendix 2, the inference engine computes the total number of variation orders, subtotal (that assists in identifying the information when multiple filters are applied), total number of variations, subtotal of variations, total cost of variation orders, subtotal cost, and total time implication for the particular project. In addition to computing the abovementioned information, the inference engine also computes and enumerates the number of variations according to various types of variations. The inference engine also assists in computing the actual contingency sum by deducting the cost of variations requested and funded by the institution or other sources. A variety of filters are provided on the micro layer that assists in sieving information by certain rules. The user would be able to apply multiple filters for finding out the most frequent causes of variations, most frequent types of variations, and variations with most significant cost implication and time implication. The multiple summaries that can be generated by applying filters and using the KBDSS query form are presented in Figure 9. The summary section of the micro layer can saved for future reference. This feature of the KBDSS assists in carrying out comparative analyses of the information provided in all the layers of the KBDSS. The inference engine integrates the summary section with the filter applications that assist in indicating the multiple filters application results in the summary section. The results in the summary section assist the user in determining the most important causes of variations in each project. However, the micro layer also provides detailed information (as mentioned above) about all the 79 educational projects for a comprehensive analysis. The effect and control tab creates an interface between the micro layers and the effect and control layers of the KBDSS (see Figure 10, Appendix 2). Effects and controls layer The third segment of the knowledge-base is the effects and controls layer that suggests most important effects and most effective controls for each cause of variations. This layer
56
contains 53 sub-layers based on the potential causes of variations and 10 sub-layers of most important causes combined (Note: the 53 causes were identified from the literature review, analysis of information given in the source documents and in-depth interviews with the developers, consultants and contractors). The graphical presentation of the 5 most important effects and 5 most effective controls for the cause of variations is presented (Figure 11, Appendix 2). The upper part of the graphical presentation displays the potential effects of the cause of variations and the lower part presents the most effective controls for the cause of variations. Here the effects and controls for variation orders are tabulated according to the survey results discussed in the previous sections of this report. The CDP form is provided in the effects and controls layer, which enables the user to switch among the effects and controls layer, micro layer and the macro layer that contains major information about all the 79 projects. The names of the projects can be selected in the CDP form that links with the corresponding micro layers. The user interface in the effects and controls layer allows the user to access, edit, modify, add and view the graphical presentation of the cause of variations and its potential effects and effective controls. The controls selection tab is provided in the CDP form. This feature assists in linking the knowledge-base with the decision support shell.
10.2 Decision Support Shell

As mentioned in the previous section, the 5 most effective controls for the cause of variations were presented on the effects and controls layer, and the layer was linked with the decision support shell, as shown in Figure 11. The decision support shell is integrated with the knowledge-base to assist the user in selecting the appropriate controls of variations and variation orders. The decision support shell provides decision support through a structured process consisting of building the hierarchy among the main criterions and the suggested controls, rating the controls, and analyzing the controls for selection through multiple analytical techniques, for instance, the analytical hierarchy process, multi-attribute rating technique, and direct trade-offs. The decision support shell contains four layers that are based on the structured process of decision making, namely, main panel, building the hierarchy between criterions and controls, rating the controls, selecting the best controls. Main panel The main panel contains the goal, main criteria and the most effective controls for variations as shown in Figure 12. As discussed above, the CDP form links with the corresponding main panel that contains the main criteria, and the suggested controls. Hence, the decision support shell contains 53 layers based on the each cause of variations and their most
57
effective controls. These layers are developed considering the three main criterions, i.e., time, cost and quality, for evaluating the suggested controls. The suggested controls would be evaluated based on the given criteria. In this layer the user may add any suggested controls that are considered to be important. Hence, this stage is named as the brainstorm session. The provision of the facility for adding more controls based on the brainstorm session, or deducting any suggested controls from the panel of suggested controls is available till the next step that generates the hierarchy among the criteria and the controls for variations. Building the hierarchy between criterions and controls This layer displays the root cause of variations and the most effective controls for the cause derived from the 53 sub-layers. The main objective of this layer is to generate the hierarchy between the main criteria and the suggested controls for variations. For building the hierarchy, the user may use the function key given in the main menu. The shell generates hierarchy among the goal, the criteria and the suggested controls. The shell graphically presented the hierarchy among the goal, the criteria and the suggested controls for the cause of variations (Figure 13, Appendix 2). The hierarchy assists in rating all the suggested controls. Rating the controls The rating process includes four main activities: choosing a rating method, selecting rating scale views, assigning rating scales and entering weights or scores. This layer provides analytical hierarchy process (AHP) as a rating technique. This is because the decision will be based on purely qualitative assessments of the suggested controls. There are three rating methods available, i.e., direct comparison, full pair-wise comparison, and abbreviated pair-wise comparison. Direct comparison is used to enter quantitative data about each criterion. These values come from a previous analysis or from experience and detailed understanding of the issue. Full pair-wise comparison means comparing in pairs and is useful if the quantitative data is not available for each criterion, or most of the criteria are similar in nature. Each criterion in a rating set is compared against every other criterion in the same set as shown in Figure 14, Appendix 2. Abbreviated pair-wise comparison is similar to full pair-wise comparison except that it contains smaller sets. It omits comparisons that are obvious, for instance, if time is more important than cost, and cost is more important than quality, then time is also more important than quality. The latter comparison is omitted.
58
Three types of scale views are provided for entering weights, i.e., numerical, verbal and graphical. These three types are provided for a user-friendly interface, any of the scale views provided can be used to input the weights for the criteria. The direct method is the default rating method and is used for entering weights for this decision process. The first step for rating the controls was to assign weight to the criteria, i.e., time, cost and quality. The main criterion for assigning weight to the sub criteria, i.e., time, cost and quality, was selecting the controls. This assisted in indicting the importance of each criterion in selecting the controls for the cause of variations. The second step was to rate the suggested controls with respect to quality, as quality was rated critical (Figure 16). The rating priority is based on hierarchy of the main criteria rated earlier in the first step. The user rated all the suggested controls and assigned weights to each alternative (control) (Figure 17). The third step was to rate the suggested controls with respect to time, as time was rated very important. The fourth step was to rate the suggested controls with respect to cost, as this was also rated as important. The user rated all the suggested controls and assigned weights to each alternative (control) (Figure 19). The abovementioned steps are dependent on the number of criterions, for instance, the user may add sub-criteria to the given three main criteria. Depending on the number of subcriteria, the steps of assigning weights will be increased accordingly. The shell does not let the user miss a rating. Once the rating is completed, then the user may go to the next step i.e., selecting the best controls. Selecting the best controls Once rating is completed, the shell calculates the decision scores and displays a graphical presentation of the results as shown in Figure 21. The decision score can be sorted according to ascending or descending orders, which assist in viewing the comprehensive scenario. The suggested controls are displayed with their corresponding decision score and its graphical presentation. The user can easily select the best controls based on the decision scores. Furthermore, the results can be analyzed according to various contributions by criteria. The graphical presentation (stacked horizontal bar) of the results is shown in Figure 22 according to the contributions by criteria. The user may analyze the suggested controls by selecting any one of the criteria. For further analysis, various analysis modes are also provided, i.e., sensitivity by weights, data scatter plots, and trade-offs of lowest criteria. All these modes assist in analyzing and presenting the decision. Furthermore, the shell also presents various other options for displaying the results, i.e., decision score sheet, pie charts, stacked bars, stacked horizontal bars, and trend.
59
11.0 Conclusions
This paper initially presented the professionals views of the causes, effects and controls for variation orders for educational building projects in Singapore. Through the questionnaire survey and in-depth interviews with the professionals who were involved with the educational building projects, the potential causes, frequent effects and effective controls for variation orders for these projects were identified. Arising there from, a comprehensive tabulation of the 53 causes of variation orders and their frequent effects and effective controls was also developed, that assisted in developing a knowledge-based decision support system (KBDSS). The study will benefit the professionals involved with educational building projects. As discussed in the previous sections, the most important causes were from owner related variations and consultant related variations groups, and the suggested controls for variation orders were mostly for the design stage. Hence, it is imperative that the professionals should concentrate more on defining the scope of projects, allocating sufficient time for design development and improving design detailings and compliance with government regulations that would assist in reducing variations related to these groups. Furthermore, the study also suggested that variations can be reduced with due diligence during the design stages. The management of variation orders is considered successful if the variation orders are resolved in a timely manner to the benefit of all the parties and the project (Cox, 1997). The study identified the most likely areas on which to focus to reduce the variations in future educational projects. Hence, the suggested controls would assist professionals in taking proactive measures for reducing variation orders. Furthermore, the study suggests that the successful management of variation orders must begin before the start of construction and continue through to the close-out of the last contract. Successful management of variations demands awareness, preparation and input from the project owner as well as the project contractors. Eventually, the study presents research into the development of a KBDSS for the management of variation orders for educational buildings in Singapore. The KBDSS consists of two main components, i.e., a knowledge-base and a decision support shell for selecting appropriate controls. The database is developed through data collected from source documents of 79 educational projects, a questionnaire survey, literature review and interview sessions with the professionals who were involved in the educational projects. The knowledge-base was developed through initial sieving and organization of data from the database. Furthermore, the knowledge-base was divided into three main segments namely, macro layer, micro layer and effects/controls layer. These three segments assisted in
60
analyzing and presenting accurate and timely information about variations and variation orders for educational building projects in Singapore. The decision support shell was integrated with the knowledge-base to assist the user in selecting the appropriate controls of variations and variation orders. The decision support shell provided decision support through a structured process consisting of building the hierarchy between the main criteria and the suggested controls, rating the controls, and analyzing the controls for selection through multiple analytical techniques. The benefits of the KBDSS include the ability to assist the professional team (decision makers) to select the appropriate controlling methods to minimize variations and their effects. The KBDSS is capable of displaying variations and their relevant details, a variety of filtered knowledge, and various analyses of available knowledge of the completed educational projects. This would eventually lead the decision makers to the various suggested controls for the variations and assist in selecting the most appropriate controls. The decision makers can interact with the system so that the decision makers can constantly refine and add data to keep the system up-to-date. Various filters are provided in the KBDSS that assist in viewing the exact information through multiple filters that are applicable simultaneously. The KBDSS provides an extremely fast response to the queries and also provides user-friendly interfaces that assist the decision maker to add, edit or modify the information given in all layers of the KBDSS. The user can add potential controls of variations and rate these controls with multiple techniques provided in the KBDSS for analyzing and selecting the best controls for variation orders. The development of the KBDSS was based on the information gathered from the source documents of completed educational projects and in-depth interviews with the professionals and would help decision makers in taking proactive measures for reducing potential variations. In short, the KBDSS is able to assist project managers by providing accurate and timely information for decision making, and a user-friendly tool for analyzing and selecting the suggested controls for variation orders for educational buildings in Singapore.
61
12.0 Recommendations
The study identifies the most likely areas on which to focus to reduce the variations in future educational projects. Recommendations are now suggested based on the findings of the research and literature review. A clear and thorough project brief would assist in eliminating variations that may arise due to unclear project objectives. Eventually, this may assist in developing a clear scope of work for the professionals. The involvement of the owner in the design phase would assist in clarifying the project objectives and in identifying the noncompliance with their requirements at an early stage. Eventually, this may help in eliminating the occurrence of variations arising from errors and design discrepancies during the construction stage where the impact of the variations can be severe. The controls for the errors and omissions in design, design discrepancies and frequent change in design, would be through thorough detailings of design. Thorough detailings of design was perceived as one of the most effective controls for variation orders for educational building projects. This will provide an opportunity for the consultant to review and finalize the design during the design phase. This would assist in reducing the variation occurrences during the construction phase where the impact of variations can be severe. As discussed in the previous sections, the most important causes of variation orders were mostly owner related variations and consultant related variations. Hence, the study suggested that variations can be reduced with due diligence during the design stages. Furthermore, the suggested controls also emphasized the involvement of all the parties for a collaborative effort in reducing variations. If professionals have a knowledge-base established based on past similar projects, it would assist the professional team to plan effectively before starting a project, during the design phase as well as during the construction phase to minimize and control variations and their effects. The knowledge-base would assist project managers by providing accurate and timely information for making more informed decisions for effective management of variation orders. Therefore, a comprehensive knowledge-based system established based on past similar projects is highly recommended.
62
The identification and analysis of potential variations possible in a project as early as possible can enhance the management of variations in the projects. Learning from these variations is imperative because the professionals can improve and apply their experience in the future. The KBDSS placed emphasis on sharing the lessons learned from existing projects with project teams of the future projects.
The KBDSS provides an excellent opportunity to project managers to learn from past experiences. The KBDSS should be applied in the early stages of the construction projects. In providing a systematic way to manage variations through the KBDSS, the efficiency of the building project and the likelihood of project success can be enhanced.
This paper presented the in-depth analyses of the causes, their frequent effects and effective controls for variations in educational building projects in Singapore. This may assist professionals in analyzing variations and taking proactive measures for reducing variation orders. The KBDSS is able to assist project managers by providing accurate and timely information for decision making, and a user-friendly tool for analyzing and selecting the suggested controls for variation orders for educational buildings in Singapore. The study will not only benefit the professionals involved with educational building projects but also be useful for students in understanding the issues. The building professionals and students would be able to learn about the root causes of variation orders and their downstream effects that may assist them in their evaluation of variation orders. Furthermore, with appropriate modifications, the KBDSS will also be useful for the management of variations in other types of building projects.
63
13.0 Practical Application of Research

This is a timely study as the programme of rebuilding and improving existing educational buildings is currently underway in Singapore; it provides the best opportunity to address the contemporary issues relevant to the management of variation orders. The study presents indepth analyses of the causes, their effects and controls for variations in educational building projects. This would assist professionals in analyzing variations and selecting the most appropriate controls for minimizing variation orders. The study is valuable for all the professionals involved with developing educational projects. The litmus test for successful management should not be whether the project was free of variation orders, but rather, if variation orders were resolved in a timely manner to the benefit of all the parties and the project. A clearer view of the causes and their impacts on the projects will enable the project team to take advantage of beneficial variations when the opportunity arises, without an inordinate fear of the negative impacts. Eventually, a clearer and comprehensive view of the causes, their effects and potential controls will result in more informed decisions for effective management of variation orders. Furthermore, considering the fact that the variations are common in all types of construction projects, this study also contributes to effective management of variation orders as the in-depth analyses of the causes, their frequent effects and effective controls, can be used by professionals to take proactive measures for reducing and controlling variation orders in various other types of residential and commercial projects, etc. Although variations are frequently unavoidable in the construction industry, negative variations are undesirable for building projects as these would have an adverse impact on time, cost and quality. In the worst case scenario, negative variations would cause a building project to overrun its budget as well as time schedule, leading to a delay in handing a completed educational building project to the users (i.e. the principal, teachers and students). The KBDSS is a unique system developed specially for the effective management of variation orders for educational building projects under the rebuilding and improvement programme for the first time. Primarily, the KBDSS is developed based on six fundamental principles of effective variation management. The system provides an excellent opportunity to project managers to learn from past experiences. It is important to note that this system for the management of variations is not designed to make decisions for users, but rather it provides pertinent information in an efficient and easy-to-access format that allows users to make more informed decisions and judgments. Although this system does not try to take over the role of the human experts or force them to accept the output of the system, it provides more
64
relevant evidence and facts to facilitate the human experts in making well-informed final decisions. In providing a systematic way to manage variations through the KBDSS, the efficiency of the building project and the likelihood of project success can be enhanced. The KBDSS will be used by the governmental agency responsible for developing the educational projects in Singapore. The study initially presented in-depth analyses of the causes, effects and controls for variation orders for educational building projects in Singapore. This may assist professionals in analyzing variations and selecting the appropriate controls for minimizing variation orders. Hence, the study is valuable for all the professionals involved with developing the educational projects. As mentioned earlier, a clearer view of the causes and their impacts on the projects will enable the project team to take advantage of beneficial variations when the opportunity arises without an inordinate fear of negative impacts. Eventually, a clearer and comprehensive view of causes, their effects and potential controls will result in informed decisions for effective management of variation orders. Furthermore, this study also contributed to knowledge as the in-depth analyses of the causes, effects and controls for variation orders for educational building projects, can be used by future researchers to carry out studies on the management and controls of variation orders in various other types of projects. Eventually the in-depth analyses of the causes, effects and controls for variations were used as basis for developing the KBDSS for management of variation orders for educational projects in Singapore. Although there is a body of knowledge relating to the management of variation orders, the relationships between the causes, effects and controls of variation orders remain unclear. The extensive surveys, interviews and literature review undertaken in this present study established these relationships for the first time on a holistic basis. The study went beyond the establishment of these relationships to utilize, again for the first time, information technology to build a KBDSS to aid in decision making. This may assist professionals in analyzing variations, and selecting the appropriate controls for minimizing their adverse impacts. Furthermore, by having a systematic way to manage variations, the efficiency of project work and the likelihood of project success should increase. The system emphasized on sharing the lessons learned from existing projects with project teams of future projects. The KBDSS provides an excellent opportunity to project managers to learn from past experiences. Furthermore, the KBDSS will help to enhance productivity and cost savings in that: (1) timely information is available for decision makers/project managers to make more informed decisions; (2) the undesirable effects (such as delays and disputes) of variations
65
may be avoided as the decision makers/project managers would be prompted to guard against these effects; (3) the knowledge base and pertinent information displayed by the KBDSS will provide useful lessons for decision makers/project managers to exercise more informed judgments in deciding where cost savings may be achieved in future educational building projects; and (4) the KBDSS provides a useful tool for training new staff members (new professionals) whose work scope include educational building projects. The study would assist building professionals in establishing an effective management system. Furthermore, the survey results reported and the KBDSS can be efficiently used to increase students understanding of these issues. It may assist students in learning about variance performance in the particular case studies reported (educational buildings in Singapore). This would be an interesting online resource for students in 4th year Architecture, Building and Quantity Surveying. For Architecture students, the relevant modules would be Professional Practice or Architectural Practice, for Building and Quantity Surveying students, the relevant modules would be Contract Administration or Professional Practice. The study presented an extensive list of potential causes, effects and controls for variations, which can be used as a basis for understanding the issues. The students will be able to analyze the causes, their effects and controls for variations based on the accurate and real knowledge provided in the KBDSS. The system would assist them in learning about the issues of designs, contracts, management and project variance through the wealth of information based on past educational projects provided in the KBDSS. Furthermore, the students would be able to apply numerous filters to the consolidated knowledge to analyze the various situations related to different projects. Likewise, the KBDSS can be used as a more general research tool because the students may populate it with their own data and compare with the educational projects reported in this paper. The students/researchers can also use the KBDSS platform to carry out studies on the management and controls of variation orders in other types of building projects i.e., commercial, residential and industrial projects etc. With further generic enhancement and modification, the KBDSS will also be useful for the management of variation orders in other types of building projects, thus helping to raise the overall level of productivity in the construction industry. The system developed and the findings from this study would also be valuable for all building professionals in general. Furthermore, this study also contributed to knowledge as the research into development of the system can be used by future researchers to carry out studies on the development of similar management systems for other types of building projects.
66
14.0 Distinct Features of the System

As mentioned earlier, the KBDSS is a unique system developed specially for the effective management of variation orders for educational building projects under the rebuilding and improvement programme which is presently underway in Singapore. Furthermore, the KBDSS is unique as this is first time whereby the records for building projects are consolidated and systematically analyzed. The distinct features of the system are described briefly below. The system is based on accurate and real data trawled from source documents of 79 past educational projects completed under the rebuilding and improvement programme in Singapore. The information is verified by the developer, consultants and contractor through in-depth interviews based on the data collected. The system displays actual variations and their relevant in-depth details, a variety of filtered knowledge, and various analyses of the available knowledge. The system suggests, based on detailed feedback from the building professionals, the top five most frequent effects and most effective controls for each cause of variations. The system is dynamic and designed to accommodate information pertinent to variations in ongoing projects that provides a platform for the organization to continuously learn and develop based on current building projects. It has an extremely user-friendly interface. The knowledge consolidation process of the past experience will allow such knowledge to reside within an organization rather than residing within individual staff that may leave over time. Furthermore, as the KBDSS systematically consolidates all the decisions that have been made for numerous projects over time so that individuals, especially new staff would be able to learn from the collective experience and knowledge of everyone. Hence, the KBDSS has a great potential for training new staff members. The new staff will be able to explore the details of all previous actions and decisions taken by other staff involved with the educational projects. This would assist them in learning from past decisions and making more informed decisions for effective management of variations.
67
Accurate analysis generated by KBDSS based on past information will help the professionals to crystallize the past learning experience so that better decisions would be made for future educational building projects/programmes.
It combines the self-learning capacities of a group with inputs and processing from experts (professionals and authoritative publications). This should lead to a more comprehensive learning experience for all and bring learning to a higher level.
It facilitates knowledge/learning harvesting of multiple and recurring projects that occurs over a period of time through the standardization of records and derivations of classification. With the records serving as an objective basis, staff would be able to recall information and participate in learning in a more unbiased manner even though each may be managing different projects and there are timelags between project implementation and discussion.
It retains the learning points in a knowledge base as described in earlier sections. This facilitates multiple reuse of knowledge in a team environment. The knowledge base acts as an authoritative reference for decision making as the learning points have been improved through processing by experts. Also, by constantly adding new learning points to the knowledge base as more projects are analysed, the knowledge base is updated.
The KBDSS emphasizes the importance of a learning from past experience culture, promotes the use of a structured learning methodology and seeks to transfer individual knowledge to the institutional knowledge of the organization.
The KBDSS can be extended to university staff teaching contract administration and project management. It would assist them in teaching the students about the issues of contracts and project variance through the wealth of information based on past educational projects provided in the KBDSS.
The KBDSS and the study would assist students in learning about variance performance in the particular case studies reported i.e., educational building projects in Singapore. As the system is dynamic and designed to accommodate information pertinent to variations in projects, the students may use it as a more general research tool for example, the students may fill it with their own data and compare with the educational projects reported in this paper.
68
16.0 Future Work

The study focuses on developing the knowledge-based decision support system for effective management of variation orders that would enable the professionals to be aware of factors which initiate variations, their frequent effects and effective controls. This provides the professionals with requisite knowledge to make more informed decisions and to take proactive measures for reducing potential variations in future projects. Finally, as this study presents in-depth analyses of the causes, effects and controls for variations in educational building projects in Singapore; further works can be extended to other types of construction projects in the future. This study also contributed to knowledge as the research into development of the system can be used by future researchers to carry out studies on the development of similar management systems for other types of building projects.
69
Acknowledgements
The authors wish to acknowledge the government agency, the consultants and the contractors for their kind responses and making available the data needed. The name of the government agency was not revealed in this paper to preserve its anonymity. The financial support provided by the National University of Singapore under research grant no. R296000-078-112 is gratefully acknowledged. The authors would particularly like to thank Ms. Diane Bowden for her help and support throughout.
70
References
Aibinu, A.A. & Jagboro, G.O. (2002) The effects of construction delays on project delivery in Nigerian construction industry. International Journal of Project Management, 20(2), pp. 593599. Arain, F.M. (2002) Design-Construction Interface Dissonances. unpublished MS Thesis, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia. Arain, F.M. (2005) Potential barriers in management of refurbishment projects. Journal of Independent Studies and Research, 3(1), pp. 22-31. Arain, F.M., Assaf, S.A. & Low, S.P. (2004) Causes of discrepancies between design and construction. Architectural Science Review, 47(3), pp.237-249. Assaf, S.A., Al-Khalil, M. & Al-Hazmi, M. (1995) Causes of delays in large building construction projects. Journal of Construction Engineering and Management, ASCE, 11(2), pp. 45-50. Barrie, D.S. & Paulson, B.C. (1992) Professional Construction Management: Including CM, Design-Construct, and General Contracting. McGraw Hill, New York. Burati, J.L., Farrington, J.J. & Ledbetter, W.B. (1992) Causes of quality deviations in design and construction. Journal of Construction Engineering and Management, 118(1), pp. 34-49. Cameron, I., Duff, R. & Hare, B. (2004) Integrated Gateways: Planning out Health and Safety Risk. Research Report 263, Glasgow Caledonian University, UK. Chan, D.W.M. & Kumaraswamy, M.M. (1997) A survey of time-cost relationships in Hong Kong construction projects. Building Technology and Management Journal, 20(2), pp. 54-72. Chan, A. & Yeong, C. (1995) A comparison of strategies for reducing variations. Construction Management and Economics, 13(6), pp. 467-473. Chappell, D. & Willis, A. (1996) The Architect in Practice. (8th edition) Blackwell Science Ltd, USA. CII (1986a) Constructability: A Primer. Construction Industry Institute, University of Texas at Austin, TX. CII (1986b) Impact of Various Construction Contract Types and Clauses on Project Performance. Publication 5-1, Construction Industry Institute, University of Texas at Austin, TX.
71
CII (1990a) The Impact of Changes on Construction Cost and Schedule. Publication 6-10, Construction Industry Institute, University of Texas at Austin, TX. CII (1990b) Scope Definition and Control. Publication 6-2, Construction Industry Institute, University of Texas at Austin, TX. CII (1994a) Pre-project Planning: Beginning a Project the Right Way. Publication 39-1, Construction Industry Institute, University of Texas at Austin, TX. CII (1994b) Project Change Management. Special Publication 43-1, Construction Industry Institute, University of Texas at Austin, TX. CII (1995) Qualitative Effects of Project Changes. Publication 43-2, Construction Industry Institute, University of Texas at Austin, TX. Clough, R.H. & Sears, G.A. (1994) Construction Contracting. (6th edition) John Wiley & Sons Inc., New York. Cox, R.K. (1997) Managing change orders and claims. Journal of Management in Engineering, ASCE, 13(1), pp.24-30. Cox, S. & Hamilton, A. (1995) Architects Job Book. (6th Edition) RIBA Publications, Royal Institute of British Architects, UK. Cushman, R.F. & Butler, S.D. (1994) Construction Change Order Claims. Wiley Law Publications, John Wiley & Sons Inc., New York. DellIsola, A.J. (1982) Value Engineering in the Construction Industry. Van Nostrand Reinhold, New York. Dulaimi, M.F. & Hwa, T.F. (2001) Developing world class construction companies in Singapore. Construction Management and Economics, 19(3), pp. 591-599. Fisk, E. R. (1997) Construction Project Administration. (5th edition) Prentice Hall, New Jersey. Geok, O.S. (2002) Causes and Improvement for Quality Problems in Design & Build Projects. unpublished B.Sc. Thesis, National University of Singapore. Singapore. Gray, C. & Hughes, W. (2001) Building Design Management. Butterworth-Heinemann, Oxford, UK. Hester, W., Kuprenas, J.A. & Chang T.C. (1991) Construction Changes and Change Orders: Their Magnitude and Impact. CII Source Document 66, University of California- Berkeley.
72
Ibbs, C.W. (1997a) Changes impact on construction productivity. Journal of Construction Engineering and Management, ASCE, 123(1), pp. 89-97. Ibbs, C.W. (1997b) Quantitative impacts of project change: size issues. Journal of Construction Engineering and Management, ASCE, 123(3), pp. 308-311. Ibbs, C.W. & Allen, W.E. (1995) Quantitative Impacts of Project Chang. Construction Management Technical Report no.23, University of California at Berkeley, USA. Ibbs, C.W., Lee, S.A. & Li, M.I. (1998) Fast trackings impact on project change. Project Management Journal, 29(4), pp. 35-41. Ibbs, C. W., Wong, C.K. & Kwak, Y.H. (2001) Project change management system. Journal of Management in Engineering, ASCE, 17(3), pp. 159-165. Ibbs, C.W., Wall, D.E., Hassanein, M.A., Back, W.E., DeLaGarza, R.K., Twardock, J.J. & Schran, S.M. (1986) Determining the Impact of Various Construction Contract Types and Clauses on Project Performance. CII Source Document 10, Construction Industry Institute, University of Texas at Austin. Kaming, P.F., Olomolaiye, P.O., Holt, G.D. & Harris, F.C. (1997) Factors influencing construction time and cost overruns on high rise projects in Indonesia. Construction Management and Economics, 15(1), pp 83-94. Kish, L. (1995) Survey Sampling. (65th edition) John Wiley and Sons Inc., New York. Kometa, S.T.; Olomolaiye, P.O. & Harris, F.C. (1994) Attributes of UK construction clients influencing project consultants performance. Construction Management and Economics, 12(2), pp. 433-443. Kumaraswamy, M. M., Miller, D. R. A. & Yogeswaran, K. (1998) Claims for extensions of time in civil engineering projects. Construction Management and Economics, 16(3), pp.283294. Mendelsohn, R. (1997) The constructability review process: a constructors perspective. Journal of Management in Engineering, ASCE, 13(3), pp.17-19. Miresco, E.T. & Pomerol, J.C. (1995) A knowledge-based decision support system for construction project management. Proceedings of the Sixth International Conference on Computing in Civil and Building Engineering, (2), pp. 1501-1507. Mokhtar, A., Bedard, C. & Fazio, P. (2000) Collaborative planning and scheduling of interrelated design changes. Journal of Architectural Engineering, ASCE, 6(2), pp. 66-75.
73
OBrien, J.J. (1998) Construction Change Orders. McGraw Hill, New York. Odell, J.H. (1995) School Buildings Planning, Design and Construction. The Association of Independent Schools, NSW Ltd., Australia. Sanvido, V., Parfitt, K., Guvensia, M. & Coyle, M. (1992) Critical success factors for construction projects. Journal of Construction Engineering and Management, ASCE, 118(1), pp. 94-111. Thomas, H.R. and Napolitan, C.L. (1994) The Effects of Changes on Labor Productivity: Why and How Much. CII Document 99, The Pennsylvania State University, USA. Thomas, H.R. & Napolitan, C.L. (1995) Quantitative effects of construction changes on labor productivity. Journal of Construction Engineering and Management, ASCE, 121(3), pp. 290296. Tiong, R.S. (1990) Effective controls for large scale construction projects. Project Management Journal, 11(1), pp. 32-42. Wang, Y. (2000) Coordination issues in Chinese large building projects. Journal of Management in Engineering, ASCE, 16(6), pp. 54-61. Zeitoun, A. & Oberlender, G. (1993) Early Warning Signs of Project Changes. CII Source Document 91, Oklahoma State University, USA.
74
APPENDIX 1
Table 14: Relative Importance Index (RII) of causes and effects
E1 E2 E3 E4 E5 E6 E7 E8 E9 E 10 E 11 E 12 E 13 E 14 E 15 E 16 Ca 1 0.190 0.201 0.054 0.245 0.250 0.065 0.130 0.098 0.212 0.092 0.033 0.065 0.022 0.141 0.033 0.174 Ca 2 0.136 0.136 0.043 0.158 0.174 0.060 0.120 0.087 0.163 0.054 0.022 0.071 0.082 0.065 0.022 0.136 Ca 3 0.022 0.027 0.011 0.027 0.038 0.005 0.011 0.011 0.011 0.011 0.016 0.005 0.022 0.016 0.000 0.005 Ca 4 0.141 0.152 0.054 0.196 0.179 0.060 0.120 0.087 0.168 0.054 0.027 0.065 0.087 0.071 0.011 0.141 Ca 5 0.076 0.092 0.043 0.103 0.103 0.049 0.049 0.043 0.071 0.043 0.027 0.033 0.016 0.065 0.005 0.060 Ca 6 0.033 0.043 0.011 0.016 0.043 0.011 0.005 0.005 0.022 0.016 0.016 0.011 0.033 0.027 0.016 0.027 Ca 7 0.022 0.027 0.011 0.022 0.021 0.005 0.005 0.005 0.016 0.011 0.010 0.000 0.005 0.016 0.010 0.021 Ca 8 0.212 0.277 0.054 0.239 0.228 0.092 0.141 0.125 0.212 0.082 0.033 0.071 0.027 0.168 0.022 0.152 Ca 9 Ca 10 Ca 11 Ca 12 Ca 13 Ca 14 Ca 15 Ca 16 Ca 17 Ca 18 Ca 19 0.125 0.196 0.158 0.125 0.065 0.071 0.163 0.109 0.147 0.109 0.076 0.207 0.310 0.255 0.190 0.065 0.087 0.261 0.136 0.234 0.141 0.158 0.043 0.043 0.043 0.043 0.022 0.027 0.049 0.027 0.043 0.033 0.022 0.207 0.250 0.217 0.201 0.103 0.136 0.212 0.158 0.212 0.136 0.092 0.205 0.250 0.223 0.201 0.092 0.130 0.234 0.168 0.228 0.141 0.120 0.049 0.054 0.049 0.038 0.016 0.027 0.049 0.027 0.038 0.038 0.027 0.120 0.120 0.109 0.109 0.082 0.092 0.098 0.087 0.103 0.098 0.049 0.109 0.103 0.087 0.092 0.049 0.060 0.087 0.065 0.092 0.076 0.054 0.185 0.217 0.212 0.179 0.098 0.109 0.212 0.141 0.217 0.147 0.103 0.060 0.071 0.082 0.060 0.011 0.016 0.076 0.054 0.087 0.049 0.043 0.027 0.033 0.033 0.033 0.011 0.016 0.038 0.027 0.038 0.022 0.016 0.043 0.049 0.049 0.054 0.033 0.038 0.054 0.038 0.054 0.054 0.027 0.082 0.114 0.022 0.022 0.005 0.005 0.022 0.071 0.027 0.022 0.005 0.120 0.195 0.168 0.087 0.016 0.027 0.147 0.060 0.158 0.082 0.103 0.022 0.038 0.038 0.016 0.000 0.000 0.033 0.016 0.038 0.022 0.022 0.125 0.190 0.179 0.130 0.076 0.087 0.168 0.092 0.174 0.120 0.082
E1 E2 E3 E4 E5 E6 E7 E8 E9 E 10 E 11 E 12 E 13 E 14 E 15 E 16
Ca 20 Ca 21 Ca 22 Ca 23 Ca 24 Ca 25 Ca 26 Ca 27 Ca 28 Ca 29 Ca 30 Ca 31 Ca 32 Ca 33 Ca 34 Ca 35 Ca 36 Ca 37 Ca 38 0.109 0.038 0.130 0.033 0.130 0.141 0.125 0.103 0.130 0.130 0.049 0.060 0.087 0.087 0.163 0.065 0.060 0.043 0.082 0.174 0.071 0.163 0.060 0.185 0.272 0.196 0.125 0.168 0.207 0.043 0.059 0.071 0.147 0.255 0.120 0.087 0.087 0.125 0.038 0.022 0.043 0.011 0.033 0.033 0.043 0.033 0.033 0.038 0.033 0.033 0.016 0.038 0.054 0.033 0.033 0.022 0.022 0.158 0.071 0.174 0.022 0.182 0.228 0.185 0.120 0.158 0.190 0.054 0.054 0.049 0.087 0.245 0.076 0.071 0.038 0.087 0.174 0.076 0.174 0.038 0.182 0.239 0.179 0.120 0.152 0.190 0.049 0.053 0.038 0.098 0.239 0.076 0.065 0.054 0.103 0.043 0.022 0.033 0.016 0.038 0.038 0.038 0.027 0.033 0.043 0.042 0.043 0.022 0.071 0.071 0.049 0.027 0.033 0.043 0.082 0.016 0.098 0.005 0.098 0.109 0.103 0.092 0.092 0.098 0.038 0.049 0.027 0.054 0.120 0.049 0.016 0.027 0.043 0.065 0.022 0.065 0.016 0.087 0.092 0.076 0.060 0.071 0.082 0.033 0.038 0.033 0.049 0.103 0.054 0.022 0.033 0.033 0.163 0.054 0.163 0.022 0.179 0.212 0.163 0.119 0.147 0.179 0.041 0.043 0.038 0.071 0.228 0.082 0.049 0.038 0.087 0.065 0.043 0.060 0.016 0.060 0.076 0.060 0.027 0.049 0.082 0.022 0.022 0.037 0.043 0.087 0.043 0.033 0.027 0.043 0.027 0.022 0.027 0.011 0.016 0.016 0.038 0.027 0.022 0.038 0.016 0.016 0.016 0.027 0.033 0.022 0.027 0.027 0.016 0.043 0.000 0.049 0.000 0.049 0.049 0.054 0.054 0.049 0.054 0.027 0.033 0.005 0.038 0.060 0.038 0.016 0.027 0.027 0.011 0.000 0.016 0.000 0.016 0.016 0.027 0.027 0.016 0.022 0.011 0.016 0.000 0.022 0.016 0.022 0.065 0.022 0.016 0.109 0.065 0.071 0.030 0.092 0.152 0.114 0.082 0.103 0.120 0.033 0.043 0.065 0.103 0.163 0.092 0.054 0.065 0.098 0.011 0.011 0.022 0.016 0.043 0.038 0.016 0.016 0.022 0.022 0.000 0.005 0.022 0.016 0.038 0.016 0.011 0.011 0.027 0.120 0.027 0.120 0.011 0.129 0.168 0.130 0.109 0.114 0.136 0.060 0.038 0.022 0.076 0.185 0.065 0.033 0.033 0.082
E1 E2 E3 E4 E5 E6 E7 E8 E9 E 10 E 11 E 12 E 13 E 14 E 15 E 16
Ca 39 Ca 40 Ca 41 Ca 42 Ca 43 Ca 44 Ca 45 Ca 46 Ca 47 Ca 48 Ca 49 Ca 50 Ca 51 Ca 52 Ca 53 0.082 0.130 0.071 0.043 0.043 0.049 0.060 0.065 0.033 0.130 0.141 0.136 0.049 0.022 0.152 0.098 0.185 0.136 0.098 0.076 0.049 0.114 0.092 0.049 0.201 0.190 0.174 0.076 0.021 0.207 0.027 0.043 0.033 0.011 0.027 0.033 0.033 0.038 0.016 0.043 0.033 0.049 0.022 0.020 0.054 0.060 0.152 0.103 0.087 0.087 0.048 0.065 0.087 0.038 0.190 0.152 0.174 0.098 0.020 0.228 0.065 0.174 0.109 0.076 0.070 0.054 0.076 0.082 0.032 0.190 0.158 0.168 0.082 0.019 0.245 0.043 0.065 0.038 0.038 0.033 0.033 0.033 0.027 0.011 0.071 0.065 0.043 0.033 0.017 0.076 0.043 0.092 0.033 0.038 0.027 0.038 0.027 0.027 0.022 0.092 0.103 0.103 0.043 0.011 0.136 0.043 0.082 0.038 0.038 0.033 0.033 0.038 0.022 0.016 0.071 0.082 0.076 0.033 0.011 0.109 0.054 0.130 0.082 0.060 0.060 0.043 0.060 0.065 0.027 0.141 0.130 0.158 0.065 0.016 0.212 0.043 0.071 0.060 0.043 0.043 0.022 0.038 0.049 0.027 0.076 0.038 0.049 0.027 0.005 0.092 0.016 0.027 0.027 0.016 0.027 0.016 0.033 0.033 0.016 0.033 0.022 0.043 0.022 0.016 0.038 0.038 0.049 0.022 0.011 0.005 0.022 0.022 0.016 0.000 0.054 0.071 0.054 0.016 0.011 0.065 0.022 0.022 0.016 0.011 0.005 0.005 0.022 0.011 0.000 0.016 0.022 0.027 0.000 0.016 0.027 0.071 0.114 0.092 0.076 0.060 0.038 0.082 0.071 0.036 0.125 0.109 0.109 0.054 0.016 0.141 0.027 0.022 0.005 0.005 0.000 0.005 0.022 0.016 0.011 0.005 0.016 0.011 0.000 0.000 0.022 0.071 0.109 0.065 0.038 0.033 0.027 0.065 0.049 0.022 0.103 0.125 0.125 0.022 0.011 0.049
<< Back
75
Table 15: Relative Importance Index (RII) of causes and controls

C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Ca 1 0.168 0.190 0.174 0.239 0.321 0.223 0.266 0.266 0.076 0.163 0.212 0.152 0.207 0.071 0.114 0.196 0.082 0.043 0.196 0.261 0.098 0.163 0.239 0.087 0.212 0.098 0.174 0.174 0.245 0.179
Ca 2 0.087 0.109 0.103 0.174 0.174 0.141 0.163 0.158 0.038 0.103 0.109 0.076 0.120 0.054 0.071 0.141 0.027 0.022 0.114 0.125 0.011 0.109 0.120 0.043 0.120 0.027 0.114 0.098 0.158 0.130
Ca 3 0.016 0.016 0.011 0.027 0.049 0.027 0.033 0.049 0.005 0.027 0.027 0.016 0.011 0.022 0.043 0.033 0.011 0.011 0.027 0.038 0.022 0.005 0.038 0.022 0.049 0.022 0.022 0.033 0.043 0.033
Ca 4 0.092 0.125 0.125 0.168 0.201 0.168 0.168 0.179 0.065 0.103 0.120 0.109 0.185 0.033 0.065 0.147 0.033 0.000 0.136 0.141 0.022 0.125 0.141 0.087 0.163 0.049 0.109 0.114 0.185 0.158
Ca 5 0.033 0.071 0.071 0.071 0.114 0.071 0.092 0.120 0.022 0.054 0.060 0.049 0.087 0.016 0.049 0.065 0.049 0.016 0.071 0.098 0.049 0.054 0.098 0.049 0.120 0.027 0.054 0.065 0.087 0.071
Ca 6 0.033 0.033 0.022 0.054 0.054 0.033 0.043 0.065 0.011 0.043 0.033 0.022 0.022 0.022 0.054 0.027 0.016 0.027 0.043 0.060 0.027 0.033 0.049 0.033 0.033 0.016 0.043 0.038 0.043 0.033
Ca 7 0.087 0.092 0.071 0.147 0.196 0.130 0.141 0.168 0.054 0.098 0.103 0.076 0.071 0.060 0.098 0.087 0.049 0.054 0.109 0.125 0.065 0.065 0.158 0.092 0.125 0.082 0.125 0.141 0.168 0.103
Ca 8 0.168 0.228 0.196 0.239 0.342 0.207 0.321 0.315 0.109 0.196 0.239 0.190 0.266 0.087 0.098 0.223 0.120 0.060 0.217 0.277 0.130 0.185 0.299 0.125 0.277 0.141 0.201 0.185 0.283 0.212
Ca 9 Ca 10 Ca 11 Ca 12 Ca 13 Ca 14 Ca 15 Ca 16 Ca 17 Ca 18 Ca 19 0.109 0.207 0.141 0.120 0.049 0.049 0.152 0.092 0.125 0.092 0.092 0.168 0.266 0.234 0.163 0.076 0.076 0.201 0.130 0.190 0.141 0.092 0.130 0.223 0.163 0.147 0.060 0.065 0.174 0.103 0.147 0.109 0.076 0.158 0.234 0.196 0.190 0.103 0.120 0.196 0.125 0.196 0.163 0.098 0.223 0.348 0.261 0.223 0.103 0.120 0.283 0.168 0.255 0.168 0.114 0.185 0.245 0.245 0.196 0.102 0.098 0.239 0.163 0.234 0.141 0.103 0.223 0.326 0.228 0.207 0.082 0.096 0.250 0.163 0.207 0.147 0.130 0.217 0.310 0.234 0.217 0.060 0.082 0.245 0.168 0.223 0.147 0.120 0.065 0.152 0.049 0.082 0.011 0.016 0.076 0.049 0.038 0.027 0.033 0.109 0.196 0.120 0.136 0.043 0.054 0.158 0.092 0.130 0.103 0.087 0.141 0.272 0.168 0.147 0.043 0.071 0.196 0.098 0.158 0.109 0.092 0.098 0.190 0.092 0.125 0.043 0.060 0.120 0.071 0.098 0.082 0.065 0.168 0.266 0.174 0.190 0.071 0.076 0.185 0.131 0.158 0.109 0.082 0.054 0.060 0.065 0.065 0.027 0.027 0.060 0.049 0.082 0.065 0.054 0.038 0.109 0.098 0.087 0.027 0.033 0.103 0.065 0.103 0.071 0.038 0.130 0.201 0.147 0.152 0.087 0.092 0.158 0.114 0.152 0.130 0.076 0.043 0.076 0.076 0.043 0.011 0.022 0.082 0.043 0.071 0.027 0.027 0.022 0.038 0.027 0.027 0.000 0.011 0.038 0.016 0.027 0.027 0.022 0.141 0.245 0.168 0.168 0.054 0.087 0.207 0.109 0.163 0.103 0.082 0.147 0.266 0.190 0.163 0.054 0.076 0.217 0.125 0.190 0.114 0.082 0.054 0.109 0.065 0.060 0.000 0.011 0.082 0.043 0.060 0.033 0.027 0.114 0.163 0.130 0.130 0.054 0.060 0.130 0.098 0.136 0.098 0.065 0.174 0.321 0.217 0.185 0.043 0.054 0.239 0.129 0.207 0.120 0.109 0.071 0.141 0.082 0.125 0.016 0.016 0.092 0.076 0.087 0.054 0.060 0.190 0.283 0.174 0.187 0.060 0.098 0.196 0.129 0.158 0.109 0.092 0.103 0.174 0.087 0.103 0.022 0.033 0.109 0.065 0.076 0.049 0.060 0.114 0.207 0.158 0.130 0.054 0.071 0.179 0.098 0.147 0.098 0.076 0.136 0.190 0.152 0.130 0.049 0.049 0.179 0.114 0.158 0.087 0.087 0.201 0.293 0.239 0.217 0.082 0.109 0.261 0.160 0.234 0.136 0.120 0.152 0.228 0.174 0.168 0.074 0.103 0.179 0.114 0.168 0.109 0.082
Ca 20 Ca 21 Ca 22 Ca 23 Ca 24 Ca 25 Ca 26 Ca 27 Ca 28 Ca 29 Ca 30 Ca 31 Ca 32 Ca 33 Ca 34 Ca 35 Ca 36 Ca 37 Ca 38 0.098 0.022 0.092 0.033 0.120 0.163 0.114 0.098 0.103 0.152 0.027 0.038 0.076 0.076 0.179 0.071 0.043 0.060 0.082 0.152 0.065 0.152 0.022 0.179 0.223 0.196 0.130 0.163 0.201 0.016 0.038 0.065 0.098 0.201 0.076 0.076 0.076 0.065 0.109 0.022 0.098 0.016 0.136 0.168 0.141 0.122 0.120 0.168 0.038 0.049 0.082 0.076 0.201 0.092 0.065 0.065 0.071 0.120 0.027 0.152 0.036 0.168 0.217 0.163 0.136 0.130 0.223 0.060 0.073 0.120 0.120 0.266 0.136 0.065 0.087 0.136 0.168 0.082 0.185 0.035 0.196 0.283 0.223 0.147 0.190 0.277 0.065 0.071 0.125 0.152 0.315 0.141 0.098 0.087 0.152 0.168 0.065 0.185 0.035 0.158 0.223 0.179 0.136 0.163 0.212 0.033 0.038 0.082 0.071 0.217 0.060 0.065 0.043 0.092 0.163 0.054 0.174 0.060 0.196 0.255 0.217 0.168 0.201 0.234 0.049 0.071 0.125 0.136 0.261 0.136 0.082 0.103 0.147 0.174 0.076 0.168 0.049 0.190 0.261 0.212 0.130 0.168 0.261 0.059 0.081 0.136 0.141 0.272 0.147 0.109 0.102 0.130 0.027 0.011 0.043 0.022 0.060 0.071 0.060 0.033 0.060 0.092 0.005 0.011 0.060 0.043 0.071 0.033 0.038 0.049 0.022 0.082 0.027 0.103 0.027 0.120 0.114 0.130 0.092 0.109 0.179 0.033 0.049 0.114 0.109 0.168 0.109 0.049 0.076 0.076 0.098 0.033 0.114 0.027 0.141 0.207 0.141 0.109 0.130 0.196 0.027 0.049 0.114 0.098 0.207 0.103 0.054 0.071 0.092 0.071 0.033 0.076 0.027 0.098 0.130 0.103 0.082 0.098 0.174 0.027 0.033 0.087 0.071 0.147 0.082 0.043 0.054 0.043 0.136 0.049 0.141 0.016 0.158 0.185 0.163 0.114 0.179 0.217 0.038 0.049 0.049 0.076 0.201 0.092 0.071 0.049 0.060 0.043 0.027 0.049 0.038 0.060 0.076 0.060 0.033 0.038 0.103 0.016 0.011 0.082 0.043 0.098 0.060 0.027 0.049 0.043 0.043 0.027 0.060 0.027 0.049 0.098 0.082 0.060 0.033 0.136 0.033 0.043 0.065 0.060 0.125 0.071 0.043 0.054 0.092 0.109 0.038 0.130 0.035 0.130 0.163 0.147 0.114 0.120 0.179 0.049 0.043 0.071 0.071 0.179 0.092 0.049 0.043 0.054 0.038 0.022 0.033 0.005 0.043 0.071 0.043 0.022 0.033 0.065 0.022 0.022 0.038 0.043 0.098 0.038 0.043 0.033 0.060 0.022 0.011 0.016 0.016 0.033 0.049 0.027 0.016 0.005 0.054 0.016 0.016 0.054 0.038 0.049 0.038 0.016 0.033 0.054 0.109 0.033 0.141 0.022 0.147 0.212 0.130 0.087 0.130 0.190 0.038 0.049 0.092 0.092 0.190 0.076 0.060 0.049 0.109 0.114 0.049 0.130 0.016 0.141 0.201 0.158 0.122 0.147 0.207 0.033 0.060 0.098 0.103 0.223 0.114 0.065 0.071 0.125 0.043 0.038 0.033 0.016 0.049 0.071 0.049 0.022 0.038 0.092 0.016 0.016 0.060 0.043 0.092 0.071 0.049 0.043 0.033 0.103 0.038 0.114 0.027 0.125 0.130 0.109 0.076 0.103 0.158 0.038 0.038 0.065 0.065 0.163 0.082 0.049 0.043 0.065 0.136 0.076 0.141 0.038 0.174 0.250 0.196 0.114 0.179 0.207 0.033 0.054 0.113 0.130 0.245 0.125 0.087 0.092 0.130 0.065 0.033 0.065 0.027 0.076 0.098 0.076 0.043 0.043 0.136 0.016 0.027 0.060 0.049 0.103 0.060 0.054 0.054 0.038 0.120 0.060 0.136 0.033 0.168 0.212 0.174 0.109 0.141 0.212 0.062 0.082 0.130 0.136 0.245 0.130 0.109 0.092 0.098 0.060 0.022 0.060 0.033 0.076 0.120 0.087 0.054 0.071 0.125 0.016 0.022 0.092 0.049 0.103 0.065 0.038 0.049 0.043 0.092 0.016 0.120 0.033 0.114 0.174 0.125 0.087 0.103 0.152 0.033 0.043 0.087 0.092 0.185 0.087 0.043 0.049 0.114 0.103 0.043 0.114 0.049 0.114 0.185 0.125 0.076 0.103 0.141 0.033 0.043 0.098 0.087 0.179 0.076 0.054 0.065 0.092 0.168 0.076 0.183 0.047 0.190 0.283 0.223 0.120 0.179 0.239 0.054 0.054 0.098 0.130 0.272 0.103 0.087 0.082 0.125 0.125 0.038 0.141 0.027 0.152 0.212 0.168 0.103 0.141 0.185 0.054 0.054 0.076 0.087 0.207 0.092 0.065 0.060 0.098
Ca 39 Ca 40 Ca 41 Ca 42 Ca 43 Ca 44 Ca 45 Ca 46 Ca 47 Ca 48 Ca 49 Ca 50 Ca 51 Ca 52 Ca 53 0.071 0.114 0.071 0.054 0.043 0.033 0.082 0.065 0.054 0.109 0.136 0.109 0.022 0.000 0.120 0.043 0.152 0.130 0.065 0.071 0.038 0.082 0.103 0.054 0.136 0.152 0.190 0.082 0.011 0.179 0.076 0.114 0.076 0.054 0.054 0.033 0.082 0.071 0.038 0.120 0.136 0.141 0.049 0.005 0.120 0.087 0.158 0.065 0.049 0.027 0.011 0.065 0.038 0.049 0.092 0.092 0.065 0.033 0.000 0.082 0.120 0.223 0.076 0.092 0.043 0.071 0.109 0.076 0.065 0.141 0.201 0.168 0.054 0.033 0.098 0.054 0.158 0.060 0.065 0.038 0.038 0.076 0.065 0.071 0.114 0.098 0.087 0.016 0.010 0.071 0.114 0.228 0.109 0.071 0.076 0.027 0.065 0.065 0.033 0.114 0.125 0.092 0.054 0.005 0.114 0.147 0.234 0.109 0.092 0.049 0.076 0.147 0.098 0.076 0.163 0.185 0.217 0.049 0.022 0.234 0.033 0.065 0.082 0.049 0.038 0.038 0.054 0.054 0.038 0.114 0.114 0.076 0.011 0.000 0.060 0.087 0.152 0.109 0.071 0.049 0.043 0.092 0.071 0.054 0.147 0.163 0.130 0.033 0.005 0.125 0.098 0.152 0.109 0.071 0.049 0.043 0.109 0.071 0.049 0.168 0.180 0.158 0.033 0.005 0.147 0.076 0.120 0.071 0.038 0.054 0.022 0.087 0.049 0.027 0.130 0.180 0.114 0.038 0.005 0.114 0.060 0.163 0.109 0.060 0.071 0.043 0.054 0.065 0.016 0.168 0.179 0.174 0.071 0.005 0.179 0.043 0.076 0.043 0.043 0.027 0.016 0.060 0.033 0.043 0.060 0.071 0.043 0.022 0.005 0.065 0.087 0.082 0.168 0.141 0.109 0.065 0.120 0.109 0.060 0.261 0.185 0.239 0.103 0.033 0.272 0.071 0.125 0.125 0.082 0.092 0.038 0.098 0.082 0.054 0.158 0.125 0.179 0.071 0.010 0.223 0.038 0.049 0.190 0.136 0.114 0.065 0.152 0.158 0.062 0.239 0.152 0.174 0.080 0.010 0.239 0.043 0.043 0.038 0.049 0.027 0.016 0.043 0.033 0.033 0.054 0.043 0.038 0.011 0.000 0.038 0.082 0.163 0.071 0.049 0.043 0.022 0.065 0.071 0.038 0.136 0.130 0.130 0.054 0.005 0.168 0.087 0.163 0.120 0.087 0.065 0.060 0.098 0.082 0.038 0.185 0.190 0.185 0.065 0.011 0.207 0.027 0.076 0.087 0.071 0.054 0.022 0.049 0.038 0.016 0.207 0.082 0.065 0.038 0.005 0.060 0.065 0.114 0.071 0.043 0.049 0.038 0.060 0.054 0.027 0.114 0.125 0.109 0.049 0.005 0.141 0.098 0.196 0.168 0.098 0.092 0.043 0.120 0.114 0.065 0.212 0.188 0.196 0.082 0.000 0.196 0.033 0.092 0.071 0.049 0.043 0.027 0.043 0.049 0.027 0.103 0.087 0.071 0.033 0.005 0.076 0.103 0.196 0.082 0.054 0.043 0.054 0.098 0.071 0.038 0.152 0.179 0.185 0.082 0.027 0.212 0.049 0.103 0.136 0.120 0.092 0.071 0.114 0.098 0.062 0.217 0.147 0.163 0.070 0.027 0.168 0.087 0.136 0.071 0.071 0.033 0.038 0.087 0.054 0.043 0.098 0.098 0.092 0.027 0.010 0.120 0.087 0.147 0.082 0.076 0.049 0.022 0.092 0.076 0.043 0.120 0.109 0.103 0.054 0.010 0.130 0.103 0.234 0.082 0.076 0.038 0.033 0.103 0.043 0.049 0.120 0.098 0.109 0.033 0.005 0.120 0.087 0.141 0.179 0.103 0.120 0.054 0.130 0.152 0.082 0.261 0.174 0.212 0.098 0.016 0.266
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
<< Back
76
Appendix 2: Knowledge Based Decision Support System (KBDSS)
Figure 6a: Macro layer of the knowledge-base that consists of the major information regarding educational building projects
Figure 6b: Macro layer of the knowledge-base (contd)

CEBE Working Paper No. 10 77
Figure 6c: Macro layer of the knowledge-base (contd)
Figure 7: Summary section displaying the results of the filters applied on the macro layer
<< Back
Figure 8a: Micro layer of the knowledge-base that contains the detailed information regarding variation orders for the educational project
Figure 8b: Micro layer of the knowledge-base that contains the detailed information regarding variation orders for the educational project (contd)
Figure 9: Multiple summary sections displaying the results of the filters applied on the micro layer, and the KBDSS query form showing the effects and controls layer tab that connects the micro layer with the effect and controls layer of the knowledge-base
Figure 10: KBDSS query form showing the effects and controls layer tab that connects the micro layer with the effect and controls layer of the knowledge-base
80
Figure 11: Effects and controls layer of the knowledge-base that pinpoints the most important effects and most effective controls for each cause of variations
Figure 12: Main panel of decision support shell that contains the goal, main criteria and the most effective controls for variations (focusing on Time, Cost and Quality)
81
Figure 13: Building the hierarchy among the goal, main criteria and controls for variations
Figure 14: Full pair-wise rating method that assists in rating each criterion against every other criterion in the same set
82
Figure 15: Abbreviated pair-wise rating method that is similar to the full pair-wise comparison except that it contains smaller sets
Figure 16: Rating the main criteria using the direct method, i.e. the default rating method provided in the KBDSS
83
Figure 17: Rating the controls for variations with respect to quality (Note: the rating priority is based on the hierarchy of the main criteria rated earlier)
Figure 18: Rating the controls for variations with respect to time
84
Figure 19: Rating the controls for variations with respect to cost
KBDSS
Figure 20: The KBDSS prompts the user when the rating is completed
85
Figure 21: The controls for variations sorted according to the decision scores
Figure 22: The suggested controls sorted according to contributions by criteria
86

Contract

Uploaded by

Copyright:

Available Formats

You might also like

Contract

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Contract

Uploaded by

Copyright:

Available Formats

10

ISSN 1744-9847 ISBN 1-905148-09-7

Editor: Chris Webster Editorial Assistant: Diane Bowden

Keywords: Building Contracts, Construction Management, Contract Administration,

CEBE Working Paper No. 10

10.0 KNOWLEDGE-BASED DECISION SUPPORT SYSTEM (KBDSS) 52

CEBE Working Paper No. 10

Rating the controls .................................................................................58 Selecting the best controls ......................................................................59

CEBE Working Paper No. 10

CEBE Working Paper No. 10

CEBE Working Paper No. 10

2.0 Significance and Rationale of Study

CEBE Working Paper No. 10

CEBE Working Paper No. 10

CEBE Working Paper No. 10

3.0 Objective of the Study

CEBE Working Paper No. 10

CEBE Working Paper No. 10

4.0 Holistic View of Variation Orders

4.1 Potential Causes of Variations

CEBE Working Paper No. 10

A. Owner Related Changes

CEBE Working Paper No. 10

A. Owner related variations

B. Consultant related variations

C. Contractor related variations

Lack of contractors involvement in design Unavailability of equipment Unavailability of skills

Figure 1 Causes of variation orders grouped under four categories

CEBE Working Paper No. 10

B. Consultant Related Variations

CEBE Working Paper No. 10

CEBE Working Paper No. 10

CEBE Working Paper No. 10

C. Contractor Related Variations

CEBE Working Paper No. 10

CEBE Working Paper No. 10

CEBE Working Paper No. 10

4.2 Potential Effects of Variations

CEBE Working Paper No. 10

4.2 Effects of variation orders

CEBE Working Paper No. 10

CEBE Working Paper No. 10

CEBE Working Paper No. 10

4.3 Controls for Variation Orders

A. Design Stage Controls for Variation Orders

CEBE Working Paper No. 10

Controls for variation orders

Figure 3 Controls for variation orders

CEBE Working Paper No. 10

C. Design-Construction interface stage

B. Construction Stage Controls for Variation Orders

CEBE Working Paper No. 10

CEBE Working Paper No. 10

C. Design-Construction Interface Stage Controls for Variation Orders

CEBE Working Paper No. 10

CEBE Working Paper No. 10

CEBE Working Paper No. 10