174785964-contract

Effective Management of Contract Variations using a Knowledge Based Decision Support System

CEBE Working Paper No. 10 1

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.



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 11 A. Owner Related Changes 12 B. Consultant Related Variations 14 C. Contractor Related Variations 17 D. Other Variations 19

4.2 Potential Effects of Variations 20

4.3 Controls for Variation Orders 24 A. Design Stage Controls for Variation Orders 25 B. Construction Stage Controls for Variation Orders 26 C. Design-Construction Interface Stage Controls for Variation Orders 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 43

9.2 Most Frequent Effects of Variation Orders 45

9.3 Most Effective Controls for Variations 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 (O’Brien, 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 (O’Brien, 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 engineer’s 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; O’Brien, 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 Kong’s 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; O’Brien,

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; O’Brien, 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.



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 (O’Brien, 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; O’Brien, 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.



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; O’Brien, 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.

Owner’s 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. Owner’s financial

problems affect project progress and quality (Clough and Sears, 1994; O’Brien, 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.



Figure 1 Causes of variation orders grouped under four categories

Change in design by consultants

Errors and omissions in design

Conflicts between contract documents

Inadequate scope of work for contractor

Lack of contractor’s involvement in design

Unavailability of equipment

Unavailability of skills

Contractor’s financial difficulties

Weather condition

Safety considerations

Change in govt. regulations

Change in economic conditions

Unforeseen problems

Socio-cultural factors Technology change

Value engineering

Lack of coordination

Design complexity

Inadequate working drawing details

Contractor’s desired profitability

Differing site conditions

Defective workmanship

Unfamiliarity with local conditions

Lack of specialized construction manager

Causes of Variation Orders

A. Owner related variations

B. Consultant related variations

C. Contractor related variations

Change of plans or scope by owner

Change of schedule by owner

Owner’s financial problems

Inadequate project objectives

D. Other variations

Change in specifications by owner

Change in specifications by consultant

Contractor’s obstinate nature

Replacement of materials/procedures

Impediment in prompt decision making process

Obstinate nature of owner

Inadequate shop drawing details

Consultant’s lack of judgment and experience

Lack of consultant’s knowledge of available materials and equipment

Honest wrong belief of consultant

Consultant’s lack of required data

Obstinate nature of consultant

Ambiguous design details

Design discrepancies (inadequate design)

Non-compliance design with govt. regulations

Non-compliance design with owner’s requirement

Fast track construction

Poor procurement process

Lack of communication

Contractor’s lack of judgment & experience

Long lead procurement

Honest wrong belief of contractor

Complex design and technology

Lack of strategic planning

Contractor’s lack of required data



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 (O’Brien, 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.



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

(Dell’Isola, 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.

Consultant’s lack of judgment and experience: Professional experience and judgment is

an important factor for successful completion of a building project (Clough and Sears, 1994;

O’Brien, 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 consultant’s 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 consultant’s lack of knowledge of available materials and equipment can cause

numerous major variations during various project phases.



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.

Consultant’s 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

(O’Brien, 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 owner’s requirements: A comprehensive design is one

that accommodates the owner’s requirements (Cox and Hamilton, 1995). A noncompliance

design with the owner’s requirements is considered an inadequate design (Fisk, 1997).

Eventually, this may cause variations for accommodating the owner’s 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 (O’Brien, 1998). As mentioned



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 contractor’s 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 contractor’s 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 (O’Brien, 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.

Contractor’s 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). Contractor’s financial difficulties may cause major variations during a

project, affecting its quality and progress.

Contractor’s desired profitability: Contractor’s 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 (O’Brien, 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.



Defective workmanship: Defective workmanship may lead to demolition and rework in

construction projects (Fisk, 1997; O’Brien, 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).

Contractor’s lack of judgment and experience: The consultant’s lack of professional

experience increases the risk of errors during construction (O’Brien, 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



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 (O’Brien,

1998).

Contractor’s lack of required data: A lack of required data may affect the contractor’s

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).

Contractor’s 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; O’Brien, 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.



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 (O’Brien, 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; O’Brien, 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



the free floats in the construction schedules. Hence, the variations affect the progress but

without any delay in the project completion.

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

4.2 Effects of variation orders

Progress is affected but without anydelay

Increase in project cost

Hiring new professionals

Increase in overhead expenses

Delay in payment

Quality degradation

Productivity degradation

Poor safety conditions

Completion schedule delay

Procurement delay

Rework and demolition

Logistic delay

Tarnish firm’s reputation

Poor professional relations

Additional payment for contractor

Dispute among professionals



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 (O’Brien, 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 (O’Brien, 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



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 firm’s 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 firm’s 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

(O’Brien, 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 (O’Brien, 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.



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 Kong’s 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 Design-

Construction 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 (Dell’Isola, 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 (O’Brien, 1998). Practical ideas that are not

accommodated during the design phase may affect the project adversely. Variation during



the construction phase is a costly activity as it may initiate numerous changes to construction

activities.

Figure 3 Controls for variation orders

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

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 Owner’s involvement during construction phase Avoid the use of open tendering Use of project scheduling techniques Comprehensive documentation of VO

Con

trols

for v

aria

tion

orde

rs

A. D

esign

stag

e B.

Con

struc

tion s

tage

C. D

esign

-Con

struc

tion i

nterfa

ce st

age

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



Owner’s 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

(O’Brien, 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 (O’Brien, 1998) as it helps in clarifying the project

objectives to all the participants. Eventually, this may reduce the design errors and

noncompliance with the owner’s 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



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.

Owner’s 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



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 contractor’s 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; O’Brien, 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.



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



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.



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.



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



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 developer’s side,

directors, principal architects, senior architects and project architects from the consultant’s

side, and directors, senior project managers, project managers and construction managers

from the contractor’s 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 developer’s side, directors, principal architects, senior

architects and project architects from the consultant’s side, and directors, senior project

managers, project managers and construction managers from the contractor’s side.



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 Questionnaires sent

Responses received Percentage Valid

responses Percentage

Developers 35 29 82.86% 28 80.00%

Consultants 82 36 43.90% 33 40.24%

Contractors 61 33 54.10% 31 50.82%

Total 178 98 55.06% 92 51.69%

Table 2 shows the detailed breakdown of the respondents from the developers’ side.

Table 2: Developers’ response to the survey

Respondents Appointments Responses received Percentage

Directors 2 7.14%

Senior Project Managers 4 14.28%

Senior Development Officers 2 7.14%

Project Managers 8 28.57%

Developers

Project Officers 12 42.85%

Total 28



Table 3 shows the detailed breakdown of respondents from the consultants.

Table 3: Consultants’ response to the survey


Directors 5 15.15%

Principal Architects 8 24.24%

Senior Architects 11 33.33% Consultants

Project Architects 9 27.27%

Total 33

Table 4 shows a detailed breakdown of respondents, from the contractors’ side.

Table 4: Contractors’ response to the survey


Directors 2 6.45%

Senior Project Managers 14 45.16%

Project Managers 9 29.03% Contractors

Construction Managers 6 19.35%

Total 31

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.



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. Causes Mean Std. Dev.

1 Change of plans or scope by owner 3.40 1.12

2 Change of schedule by owner 2.61 1.26

3 Owner’s financial problems 1.88 1.01

4 Inadequate project objectives 2.43 1.51

5 Replacement of materials or procedures 2.68 0.97

6 Impediment in prompt decision making process 2.46 0.89

7 Obstinate nature of owner 1.91 0.93

8 Change in specifications by owner 3.49 1.19

9 Change in design by consultant 3.14 1.12

10 Errors and omissions in design 3.53 1.14

11 Conflicts between contract documents 3.22 1.15

12 Inadequate scope of work for contractor 2.97 1.35

13 Technology change 2.26 0.94

14 Value engineering 2.50 1.11

15 Lack of coordination 3.15 1.19

16 Design complexity 2.65 1.04

17 Inadequate working drawing details 3.13 1.17

18 Inadequate shop drawing details 2.87 1.11

19 Consultant’s lack of judgment and experience 2.73 1.05

20 Lack of consultant’s knowledge of available materials and equipment 2.54 1.24

21 Honest wrong belief of consultant 2.30 0.92

22 Consultant’s lack of required data 2.61 1.25

23 Obstinate nature of consultant 2.07 0.89

24 Ambiguous design details 3.02 1.12

25 Design discrepancies (Inadequate Design) 3.36 1.21



Table 5: Mean and standard deviation of causes of variations

S No. Causes Mean Std. Dev.

26 Noncompliance design with govt. regulations 3.01 1.24

27 Noncompliance design with owner’s requirement 2.84 1.08

28 Change in specifications by consultant 3.03 1.09

29 Lack of contractor’s involvement in design 2.88 1.34

30 Unavailability of equipment 2.23 1.00

31 Unavailability of skills 2.24 1.00

32 Contractor’s financial difficulties 2.59 1.03

33 Contractor’s desired profitability 2.71 1.08

34 Differing site conditions 3.27 1.15

35 Defective workmanship 2.83 1.02

36 Unfamiliarity with local conditions 2.13 1.02

37 Lack of specialized construction manager 2.25 1.13

38 Fast track construction 2.64 1.13

39 Poor procurement process 2.42 1.01

40 Lack of communication 2.91 1.08

41 Contractor’s lack of judgment & experience 2.71 1.03

42 Long lead procurement 2.54 1.03

43 Honest wrong belief of contractor 2.32 0.99

44 Complex design and technology 2.27 0.95

45 Lack of strategic planning 2.71 1.01

46 Contractor’s lack of required data 2.53 1.02

47 Contractor’s obstinate nature 2.05 0.99

48 Weather conditions 3.03 1.17

49 Safety considerations 3.15 1.00

50 Change in government regulations 3.04 1.06

51 Change in economic conditions 2.60 0.84

52 Socio-cultural factors 2.21 0.79

53 Unforeseen problems 3.41 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



developers, the consultants and the contractors were correlated, Spearman’s rank

correlation was used. The Spearman’s 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: Spearman’s rank correlation for causes of variations

Correlations Spearman's rho Developers Consultants Contractors

Developers Correlation Coefficient 1 0.275* 0.743**

Sig. (2-tailed) . 0.046 0

Consultants Correlation Coefficient 0.275* 1 0.156

Sig. (2-tailed) 0.046 . 0.264

Contractors Correlation Coefficient 0.743** 0.156 1

Sig. (2-tailed) 0 0.264 .

N 53

** Correlation is significant at the .01 level (2-tailed). * Correlation is significant at the .05 level (2-tailed).

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. Causes Mean Std. Dev. Rank

10 Errors and omissions in design 3.53 1.14 1

8 Change in specifications by owner 3.49 1.19 2

53 Unforeseen problems 3.41 1.07 3

1 Change of plans or scope by owner 3.40 1.12 4

25 Design discrepancies (Inadequate Design) 3.36 1.21 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



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. Effects Mean Std. Dev.

1 Progress is affected but without any delay 3.39 1.03

2 Increase in project cost 3.89 1.00

3 Hiring new professionals 2.02 0.99

4 Increase in overhead expenses 3.29 1.36

5 Delay in payment 3.09 1.51

6 Quality degradation 2.49 1.06

7 Productivity degradation 2.80 1.09

8 Procurement delay 2.92 1.00

9 Rework and demolition 3.26 1.18

10 Logistic delay 2.91 0.91

11 Tarnish firm’s reputation 2.23 0.95

12 Poor safety conditions 2.28 1.10

13 Poor professional relations 2.18 0.89

14 Additional payment for contractor 3.35 1.10

15 Dispute among professionals 2.52 0.92

16 Completion schedule delay 3.25 1.09

The Spearman’s 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.



Table 9: Spearman’s rank correlation for effects of variations


Developers Correlation Coefficient 1 0.868** 0.654** Sig. (2-tailed) . 0 0.006 Consultants Correlation Coefficient 0.868** 1 0.7** Sig. (2-tailed) 0 . 0.003 Contractors Correlation Coefficient 0.654** 0.7** 1 Sig. (2-tailed) 0.006 0.003 . N 16 ** Correlation is significant at the .01 level (2-tailed). * Correlation is significant at the .05 level (2-tailed).

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. Effects Mean Std. Dev. Rank

2 Increase in project cost 3.89 1.00 1

1 Progress is affected but without any delay 3.39 1.03 2

14 Additional payment for contractor 3.35 1.10 3

4 Increase in overhead expenses 3.29 1.36 4

9 Rework and demolition 3.26 1.18 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 Spearman’s rank correlation results indicated that the ranking by the consultants

and the contractors were strongly correlated.

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 Review of contract documents 3.63 0.82

2 Freezing design 3.30 1.41

3 Value engineering at conceptual phase 3.54 1.04

4 Involvement of professionals at initial stages of project 3.86 1.10

5 Owner involvement at planning and design phases 4.29 0.86

6 Involvement of contractor at planning and scheduling process 3.53 1.18

7 Thorough detailing of design 4.17 0.75

8 Clear and thorough project brief 4.20 0.76

9 Reducing contingency sum 2.73 1.47

Construction stage

10 Clarity of variation order procedures 3.85 0.77

11 Written approvals 3.79 1.13

12 Variation order scope 3.60 1.03

13 Variation logic and justification 3.76 0.92

14 Project manager from an independent firm to manage the project 2.83 1.21

15 Restricted pre-qualification system for awarding projects 3.36 0.98

16 Owner’s involvement during construction phase 3.55 1.05

17 Avoid the use of open tendering 2.96 1.06

18 Use of project scheduling techniques 3.11 0.85

19 Comprehensive documentation of VO 3.83 0.90

Design-Construction interface stage

20 Prompt approval procedures 3.90 1.03

21 Ability to negotiate variation 3.32 0.95

22 Valuation of indirect effects 3.43 0.96

23 Team effort by owner, consultant and contractor to control variation orders

4.09 0.82

24 Utilize work breakdown structure 3.36 0.90

25 Continuous coordination and direct communication 4.13 0.70

26 Control the potential for variation orders to arise through contractual clauses

3.50 0.99



Table 11: Mean and standard deviation of controls for variation orders

S/No. Controls Mean Std. Dev.

27 Comprehensive site investigation 3.79 0.82

28 Use of collected and organized project data compiled by owner, consultant and contractor (share database)

3.64 0.81

29 Knowledge-base of previous similar projects 3.95 0.88

30 Comprehensive analysis and prompt decision making through computerized knowledge-based decision support system

3.61 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: Spearman’s rank correlation for controls for variations


Developers Correlation Coefficient 1 0.631** 0.559**

Sig. (2-tailed) . 0 0.011

Consultants Correlation Coefficient 0.631** 1 0.714**

Sig. (2-tailed) 0 . 0

Contractors Correlation Coefficient 0.559** 0.714** 1

Sig. (2-tailed) 0.011 0 .

N 30

** Correlation is significant at the .01 level (2-tailed). * Correlation is significant at the .05 level (2-tailed).

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.



Table13: Most effective controls for variation orders

S No. Controls Mean Std. Dev. Rank

5 Owner involvement at planning and design phases 4.29 0.86 1

8 Clear and thorough project brief 4.20 0.76 2

7 Thorough detailing of design 4.17 0.75 3

25 Continuous coordination and direct communication 4.13 0.70 4

23 Team effort by owner, consultant and contractor to control variation orders 4.09 0.82 5

The top five most effective controls were owner’s 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.



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



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.



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 project’s 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.



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

Owner’s 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.



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 owner’s 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



2n1 + 1n2 2N

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 cross-

tabulation 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 =

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.


50


Figure 4 Most important causes, their frequent effects and effective controls for variation orders

2) Change in specifications by owner

MOST

EFF

ECTI

VE C

ONTR

OLS

Thorough detailing of design

Clear and thorough project brief

Owner involvement at planning and design process

Knowledge-base of previous similar projects

Team effort by owner, consultant and contractor to control variation orders

MOST

FRE

QUEN

T EF

FECT

S

Progress is affected but without any delay

Increase in project costs


Delay in payment



3) Unforeseen problems


Avoid the use of open tendering

Comprehensive analysis and prompt decision making through computerized knowledge-based decision support system

Restricted pre-qualification system for awarding projects

Owner’s involvement during construction phase

MO

ST E

FFEC

TIV

E C

ON

TRO

LS




Delay in payment


MO

ST F

REQ

UEN

T EF

FEC

TS

1) Errors and omissions in design






MOST

EFF

ECTI

VE C

ONTR

OLS




Delay in payment


MOST

FRE

QUEN

T EF

FECT

S

Freezing design (i.e., no changes after final design)




Involvement of contractor at planning and scheduling process



MOST

EFF

ECTI

VE C

ONTR

OLS

5) Design discrepancies (inadequate design)

Completion schedule delay



Delay in payment


MOST

FRE

QUEN

T EF

FECT

S

Prompt approval procedures




Involvement of professionals at initial stages of project


Knowledge-base of previous similar projects MO

ST E

FFEC

TIVE

CON

TROL

S

4) Change of plans or scope by owner




Delay in payment


MOST

FRE

QUEN

T EF

FECT

S



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.



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.



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

Data base Data obtained from various sources

Questionnaire survey

Interview with experts

Knowledge from literature

Data from source documents of past similar projects

Data from ongoing 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

User interface

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

Software interface Import export knowledge between KB and DSS

User



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



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



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 project’s 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



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



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.



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 sub-

criteria, 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.



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



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.



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.



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.



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 in-

depth 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



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



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.



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.



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 time-

lags 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.



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.



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. R296-

000-078-112 is gratefully acknowledged. The authors would particularly like to thank

Ms. Diane Bowden for her help and support throughout.



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. 593-

599.

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.



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) Architect’s 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.

Dell’Isola, 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.



Ibbs, C.W. (1997a) Change’s 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 tracking’s 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.283-

294.

Mendelsohn, R. (1997) The constructability review process: a constructor’s 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.



O’Brien, 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. 290-

296.

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.



APPENDIX 1 Table 14: Relative Importance Index (RII) of causes and effects

<< Back

Ca 1 Ca 2 Ca 3 Ca 4 Ca 5 Ca 6 Ca 7 Ca 8 Ca 9 Ca 10 Ca 11 Ca 12 Ca 13 Ca 14 Ca 15 Ca 16 Ca 17 Ca 18 Ca 19E 1 0.190 0.136 0.022 0.141 0.076 0.033 0.022 0.212 0.125 0.196 0.158 0.125 0.065 0.071 0.163 0.109 0.147 0.109 0.076E 2 0.201 0.136 0.027 0.152 0.092 0.043 0.027 0.277 0.207 0.310 0.255 0.190 0.065 0.087 0.261 0.136 0.234 0.141 0.158E 3 0.054 0.043 0.011 0.054 0.043 0.011 0.011 0.054 0.043 0.043 0.043 0.043 0.022 0.027 0.049 0.027 0.043 0.033 0.022E 4 0.245 0.158 0.027 0.196 0.103 0.016 0.022 0.239 0.207 0.250 0.217 0.201 0.103 0.136 0.212 0.158 0.212 0.136 0.092E 5 0.250 0.174 0.038 0.179 0.103 0.043 0.021 0.228 0.205 0.250 0.223 0.201 0.092 0.130 0.234 0.168 0.228 0.141 0.120E 6 0.065 0.060 0.005 0.060 0.049 0.011 0.005 0.092 0.049 0.054 0.049 0.038 0.016 0.027 0.049 0.027 0.038 0.038 0.027E 7 0.130 0.120 0.011 0.120 0.049 0.005 0.005 0.141 0.120 0.120 0.109 0.109 0.082 0.092 0.098 0.087 0.103 0.098 0.049E 8 0.098 0.087 0.011 0.087 0.043 0.005 0.005 0.125 0.109 0.103 0.087 0.092 0.049 0.060 0.087 0.065 0.092 0.076 0.054E 9 0.212 0.163 0.011 0.168 0.071 0.022 0.016 0.212 0.185 0.217 0.212 0.179 0.098 0.109 0.212 0.141 0.217 0.147 0.103E 10 0.092 0.054 0.011 0.054 0.043 0.016 0.011 0.082 0.060 0.071 0.082 0.060 0.011 0.016 0.076 0.054 0.087 0.049 0.043E 11 0.033 0.022 0.016 0.027 0.027 0.016 0.010 0.033 0.027 0.033 0.033 0.033 0.011 0.016 0.038 0.027 0.038 0.022 0.016E 12 0.065 0.071 0.005 0.065 0.033 0.011 0.000 0.071 0.043 0.049 0.049 0.054 0.033 0.038 0.054 0.038 0.054 0.054 0.027E 13 0.022 0.082 0.022 0.087 0.016 0.033 0.005 0.027 0.082 0.114 0.022 0.022 0.005 0.005 0.022 0.071 0.027 0.022 0.005E 14 0.141 0.065 0.016 0.071 0.065 0.027 0.016 0.168 0.120 0.195 0.168 0.087 0.016 0.027 0.147 0.060 0.158 0.082 0.103E 15 0.033 0.022 0.000 0.011 0.005 0.016 0.010 0.022 0.022 0.038 0.038 0.016 0.000 0.000 0.033 0.016 0.038 0.022 0.022E 16 0.174 0.136 0.005 0.141 0.060 0.027 0.021 0.152 0.125 0.190 0.179 0.130 0.076 0.087 0.168 0.092 0.174 0.120 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 38E 1 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 0.082E 2 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 0.125E 3 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 0.022E 4 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 0.087E 5 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 0.103E 6 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 0.043E 7 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 0.043E 8 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 0.033E 9 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 0.087E 10 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 0.043E 11 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 0.016E 12 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 0.027E 13 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 0.016E 14 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 0.098E 15 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 0.027E 16 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

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 53E 1 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 0.152E 2 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 0.207E 3 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 0.054E 4 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 0.228E 5 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 0.245E 6 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 0.076E 7 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 0.136E 8 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 0.109E 9 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 0.212E 10 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 0.092E 11 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 0.038E 12 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 0.065E 13 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 0.027E 14 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 0.141E 15 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 0.022E 16 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



Table 15: Relative Importance Index (RII) of causes and controls

<< Back

Ca 1 Ca 2 Ca 3 Ca 4 Ca 5 Ca 6 Ca 7 Ca 8 Ca 9 Ca 10 Ca 11 Ca 12 Ca 13 Ca 14 Ca 15 Ca 16 Ca 17 Ca 18 Ca 19C 1 0.168 0.087 0.016 0.092 0.033 0.033 0.087 0.168 0.109 0.207 0.141 0.120 0.049 0.049 0.152 0.092 0.125 0.092 0.092C 2 0.190 0.109 0.016 0.125 0.071 0.033 0.092 0.228 0.168 0.266 0.234 0.163 0.076 0.076 0.201 0.130 0.190 0.141 0.092C 3 0.174 0.103 0.011 0.125 0.071 0.022 0.071 0.196 0.130 0.223 0.163 0.147 0.060 0.065 0.174 0.103 0.147 0.109 0.076C 4 0.239 0.174 0.027 0.168 0.071 0.054 0.147 0.239 0.158 0.234 0.196 0.190 0.103 0.120 0.196 0.125 0.196 0.163 0.098C 5 0.321 0.174 0.049 0.201 0.114 0.054 0.196 0.342 0.223 0.348 0.261 0.223 0.103 0.120 0.283 0.168 0.255 0.168 0.114C 6 0.223 0.141 0.027 0.168 0.071 0.033 0.130 0.207 0.185 0.245 0.245 0.196 0.102 0.098 0.239 0.163 0.234 0.141 0.103C 7 0.266 0.163 0.033 0.168 0.092 0.043 0.141 0.321 0.223 0.326 0.228 0.207 0.082 0.096 0.250 0.163 0.207 0.147 0.130C 8 0.266 0.158 0.049 0.179 0.120 0.065 0.168 0.315 0.217 0.310 0.234 0.217 0.060 0.082 0.245 0.168 0.223 0.147 0.120C 9 0.076 0.038 0.005 0.065 0.022 0.011 0.054 0.109 0.065 0.152 0.049 0.082 0.011 0.016 0.076 0.049 0.038 0.027 0.033C 10 0.163 0.103 0.027 0.103 0.054 0.043 0.098 0.196 0.109 0.196 0.120 0.136 0.043 0.054 0.158 0.092 0.130 0.103 0.087C 11 0.212 0.109 0.027 0.120 0.060 0.033 0.103 0.239 0.141 0.272 0.168 0.147 0.043 0.071 0.196 0.098 0.158 0.109 0.092C 12 0.152 0.076 0.016 0.109 0.049 0.022 0.076 0.190 0.098 0.190 0.092 0.125 0.043 0.060 0.120 0.071 0.098 0.082 0.065C 13 0.207 0.120 0.011 0.185 0.087 0.022 0.071 0.266 0.168 0.266 0.174 0.190 0.071 0.076 0.185 0.131 0.158 0.109 0.082C 14 0.071 0.054 0.022 0.033 0.016 0.022 0.060 0.087 0.054 0.060 0.065 0.065 0.027 0.027 0.060 0.049 0.082 0.065 0.054C 15 0.114 0.071 0.043 0.065 0.049 0.054 0.098 0.098 0.038 0.109 0.098 0.087 0.027 0.033 0.103 0.065 0.103 0.071 0.038C 16 0.196 0.141 0.033 0.147 0.065 0.027 0.087 0.223 0.130 0.201 0.147 0.152 0.087 0.092 0.158 0.114 0.152 0.130 0.076C 17 0.082 0.027 0.011 0.033 0.049 0.016 0.049 0.120 0.043 0.076 0.076 0.043 0.011 0.022 0.082 0.043 0.071 0.027 0.027C 18 0.043 0.022 0.011 0.000 0.016 0.027 0.054 0.060 0.022 0.038 0.027 0.027 0.000 0.011 0.038 0.016 0.027 0.027 0.022C 19 0.196 0.114 0.027 0.136 0.071 0.043 0.109 0.217 0.141 0.245 0.168 0.168 0.054 0.087 0.207 0.109 0.163 0.103 0.082C 20 0.261 0.125 0.038 0.141 0.098 0.060 0.125 0.277 0.147 0.266 0.190 0.163 0.054 0.076 0.217 0.125 0.190 0.114 0.082C 21 0.098 0.011 0.022 0.022 0.049 0.027 0.065 0.130 0.054 0.109 0.065 0.060 0.000 0.011 0.082 0.043 0.060 0.033 0.027C 22 0.163 0.109 0.005 0.125 0.054 0.033 0.065 0.185 0.114 0.163 0.130 0.130 0.054 0.060 0.130 0.098 0.136 0.098 0.065C 23 0.239 0.120 0.038 0.141 0.098 0.049 0.158 0.299 0.174 0.321 0.217 0.185 0.043 0.054 0.239 0.129 0.207 0.120 0.109C 24 0.087 0.043 0.022 0.087 0.049 0.033 0.092 0.125 0.071 0.141 0.082 0.125 0.016 0.016 0.092 0.076 0.087 0.054 0.060C 25 0.212 0.120 0.049 0.163 0.120 0.033 0.125 0.277 0.190 0.283 0.174 0.187 0.060 0.098 0.196 0.129 0.158 0.109 0.092C 26 0.098 0.027 0.022 0.049 0.027 0.016 0.082 0.141 0.103 0.174 0.087 0.103 0.022 0.033 0.109 0.065 0.076 0.049 0.060C 27 0.174 0.114 0.022 0.109 0.054 0.043 0.125 0.201 0.114 0.207 0.158 0.130 0.054 0.071 0.179 0.098 0.147 0.098 0.076C 28 0.174 0.098 0.033 0.114 0.065 0.038 0.141 0.185 0.136 0.190 0.152 0.130 0.049 0.049 0.179 0.114 0.158 0.087 0.087C 29 0.245 0.158 0.043 0.185 0.087 0.043 0.168 0.283 0.201 0.293 0.239 0.217 0.082 0.109 0.261 0.160 0.234 0.136 0.120C 30 0.179 0.130 0.033 0.158 0.071 0.033 0.103 0.212 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 38C 1 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 0.082C 2 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 0.065C 3 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 0.071C 4 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 0.136C 5 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 0.152C 6 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 0.092C 7 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 0.147C 8 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 0.130C 9 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 0.022C 10 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 0.076C 11 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 0.092C 12 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 0.043C 13 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 0.060C 14 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 0.043C 15 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 0.092C 16 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 0.054C 17 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 0.060C 18 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 0.054C 19 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 0.109C 20 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 0.125C 21 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 0.033C 22 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 0.065C 23 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 0.130C 24 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 0.038C 25 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 0.098C 26 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 0.043C 27 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 0.114C 28 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 0.092C 29 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 0.125C 30 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 53C 1 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 0.120C 2 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 0.179C 3 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 0.120C 4 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 0.082C 5 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 0.098C 6 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 0.071C 7 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 0.114C 8 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 0.234C 9 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 0.060C 10 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 0.125C 11 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 0.147C 12 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 0.114C 13 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 0.179C 14 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 0.065C 15 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 0.272C 16 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 0.223C 17 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 0.239C 18 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 0.038C 19 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 0.168C 20 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 0.207C 21 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 0.060C 22 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 0.141C 23 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 0.196C 24 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 0.076C 25 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 0.212C 26 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 0.168C 27 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 0.120C 28 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 0.130C 29 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 0.120C 30 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



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 (cont’d)



Figure 6c: Macro layer of the knowledge-base (cont’d)

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 (cont’d)



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



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)



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



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



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



Figure 19: Rating the controls for variations with respect to cost

Figure 20: The KBDSS prompts the user when the rating is completed

KBDSS



Figure 21: The controls for variations sorted according to the decision scores

Figure 22: The suggested controls sorted according to contributions by criteria

Date post:	28-Nov-2015
Category:	Documents
Upload:	lokuliyanan
View:	24 times
Download:	0 times

174785964-contract

Documents