Microsoft Word - AJCEB Vol.2 No.1 compiled sized.doc S. THOMAS NG, DUC THANH LUU AND SWEE ENG CHEN 70 THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 DECISION CRITERIA AND THEIR SUBJECTIVITY IN CONSTRUCTION PROCUREMENT SELECTION S.Thomas Ng, University of Hong Kong Duc Thanh Luu and Swee Eng Chen, University of Newcastle, Australia INTRODUCTION The escalating requirements of clients on project time, cost, quality and risk have given rise to the development and use of alternative construction procurement sys- tems (Fellows, 1993). However, since each procurement system has its distinctive characteristics, advantages and constraints, there is hardly any single best system that could suit all kinds of clients and projects (Nahapiet and Nahapiet, 1985). The selec- tion of procurement system therefore be- comes a very important task for clients, as employing an inappropriate procurement system may lead to project failure (Chua et al., 1999). The consequence may be time and cost overruns and/or general dissatis- faction (Bennett and Grice, 1990; Sharif and Morledge, 1994). Despite its significance, many clients have been selecting procurement systems in a cursory manner, and some clients even use a specific procurement system by default without making a deliberate choice (Mas- terman, 1992). A recent UK study (Hibberd and Djebarni, 1996) showed that 89% of re- spondents were dissatisfied with the pro- curement system they had previously employed. Inexperienced clients often have to rely on expert advice when selecting a procurement approach and this could result in inappropriate decisions with unforesee- able consequences (NEDO, 1985). Experi- enced clients may also suffer if they simply based their selection upon biased past ex- perience and the conservative decisions of their in-house experts (Masterman, 1992). The need for selecting and using an appro- priate procurement system for a particular construction project, together with the pro- liferation of differing procurement systems, calls for more systematic methods of selec- tion (Skitmore and Marsden, 1988). To do this, decision criteria pertinent to the selec- tion of procurement approaches and their properties (i.e. subjectivity) must be carefully identified and evaluated. This paper reports the findings of an Australian study focusing on procurement selection criteria. The sub- jectivity of the identified criteria is consid- ered and their effects on procurement selection are examined. CRITERIA FOR PROCUREMENT SELECTION According to Masterman and Gameson (1994), the selection of an appropriate pro- curement system depends largely on the accurate identification of client require- ments. Many researchers have attempted to arrive at a list of client requirements that might affect the selection of a procurement system, and the outcomes of these studies are summarised in Table 1. Speed This refers to the need to complete a project more quickly than other projects of similar nature, complexity and size. Shorter con- struction duration can be achieved by accel- erating or fast-tracking some key phases in the construction project, and this would fa- vour the use of design and build or man- agement contracting (Rowlinson and McDermott, 1999). Since the requirement for a speedier completion could often result in a premium both in the price and quality of construction, a strong justification for speed would be desirable. Certainty of completion time This relates to the degree of certainty that a project will be completed on the exact date and time specified in the contract. Time cer- tainty is a crucial need of clients, particu- larly for those involved in large or prestigious projects scheduled for a particu- lar function or event. There is a strong con- nection between the certainty of time and speed: the greater speed a procurement system can offer, the higher the degree of certainty that the project can be completed on time. DECISION CRITERIA AND THEIR SUBJECTIVITY IN CONSTRUCTION PROCUREMENT SELECTION THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 71 Table 1: Summary of client’s needs for a construction project Client’s needs Description Authors Speed Speedy procurement process, e.g. a desire to have the project completed as soon as possible. Bennett and Flanagan (1983) NEDO (1985) Skitmore and Marsden (1988) Singh (1990) Cost certainty Price and the stipulated time and knowledge of how much the client has to pay at each period during the construction phase. A reduction in unanticipated extra cost over-run Hewitt (1985) NEDO (1985) Skitmore and Marsden (1988) Singh (1990) Masterman and Duff (1994) Time certainty Degree of certainty that the project will be completed on the date, which is agreed by client and contractor when signing the contract. A reduction in unanticipated extra time over-run. Hewitt (1985) NEDO (1985) Skitmore and Marsden (1988) Singh (1990) Masterman and Duff (1994) Flexibility Ability to accommodate design changes during both design and construction periods Bennett and Flanagan (1983) Hewitt (1985) NEDO (1985) Skitmore and Marsden (1988) Singh (1990) Responsibility An involvement in, and a need to be kept informed about, the project throughout its life Bennett and Flanagan (1983) Hewitt (1985) NEDO (1985) Skitmore and Marsden (1988) Singh (1990) Masterman and Duff (1994) Complexity Client may specify innovative design/ high technology building and require particular subcontractor, or constructability analysis Bennett and Flanagan (1983) NEDO (1985) Skitmore and Marsden (1988) Singh (1990) Quality level Contractor’s reputation, aesthetics and confidence in design. A building which reflects the clients activities and image Bennett and Flanagan (1983) NEDO (1985) Skitmore and Marsden (1988) Singh (1990) Risk allocation / avoidance A wish to identify risks and uncertainties during the procuring process Bennett and Flanagan (1983) NEDO (1985) Skitmore and Marsden (1988) Singh (1990) Price competition Covering such issues as value for money, maintenance, costs and competitive tendering. Bennett and Flanagan (1983) NEDO (1985) Skitmore and Marsden (1988) Singh (1990) Masterman and Duff (1994) Disputes and arbitration NEDO (1985) Skitmore and Marsden (1988) Singh (1990) S. THOMAS NG, DUC THANH LUU AND SWEE ENG CHEN 72 THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 Certainty of price Some clients may need to have a firm price for their project before committing to it. Price may include design fees, construction costs, financing costs and management fees. According to Turner (1990), “certainty” should not be conceived as an absolute as- suredness, but instead a relative or sliding scale, i.e. “how certain” the price that a pro- curement approach could offer. Procure- ment approaches offering the highest price certainty include design and build or the traditional lump sum method. Quality level This requirement has three components: quality of materials, workmanship and the design concept. When high levels of quality of materials and workmanship are required, a more stringent supervisory and checking process must be adopted, and one would expect that the speed and price should be more flexible to cater for the required qual- ity standard. Design quality is determined by the experience of the designer, and the cost and time available. The contractor’s con- struction experience may contribute to the quality of design solutions if management contracting is employed. Flexibility Flexibility is about the ability to accommo- date variations, such as design changes (Bennett and Flanagan, 1983), during the construction phase. Flexibility is particularly needed for large and complex projects or when the exact requirements cannot be carefully established before tendering. Management contracting allows more varia- tions to be introduced without provoking significant contractual claims. Responsibility Responsibility is directly related to the de- gree of client involvement and control over the procurement process. Some clients may prefer to have a single point of responsibil- ity, and hence reduce their exposure to risk. If the clients have in-house expertise to manage the diversified responsibilities cre- ated in a project, traditional and manage- ment systems will be more suitable. Complexity This reflects the client’s desire for the final building product to be highly specialised, technologically advanced or highly serviced (NEDO, 1985). Projects with greater complexity may call for the use of traditional methods, as design can be fully developed before ten- dering proceeds. Management-type pro- curement approaches may also suit complex projects as a management con- tractor can participate in the early design stage and provide advice on buildability. Price competition Price competition covers such issues as value for money, maintenance, costs and competitive tendering (NEDO, 1985). Many public clients, to satisfy public accountability requirements, must seek competitive ten- ders. Private clients also favour competitive tendering for commercial reasons. Turner (1990), however, asserts that speed, time certainty, quality level and the complexity of the building may restrict the level of price competition. Risk allocation/avoidance This requirement reflects the degree to which the client wishes to transfer the risks of cost and time slippage to the contractor. In choosing a certain procurement system, it is important for the client to know how and to what extent the risk has consciously been passed to another organisation, how it has been shared, how the risk may not have been passed on at all, or indeed how the risk to his organisation may have been increased by the employment of another organisation. RESEARCH METHOD Semi-structured interviews were conducted to establish the decision criteria currently used for selecting procurement systems and to determine the subjectivity of those criteria. To ensure that in-depth knowledge of procurement selection was obtained, people in the sample had to: have good theoretical knowledge and practical experience in different building procurement methods have been actively involved in the process of selecting building procurement systems understand the methods of procurement selection. Since it would be difficult to identify suitable samples that meet all above considerations, purposive sampling and snowball sampling were used (Burgess, 1989). Purposive sam- pling requires researchers to identify experts who have the potential to provide DECISION CRITERIA AND THEIR SUBJECTIVITY IN CONSTRUCTION PROCUREMENT SELECTION THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 73 the necessary information. Snowball sam- pling, however, requires individuals engag- ing in the initial interviews to identify and recommend other experts suitable for the study. A protocol was developed to drive the inter- views. The protocol consisted of three parts. Part I — respondent’s profile seeks to es- tablish the knowledge and experience of interviewees on construction procurement Part II — procurement selection criteria reveals the criteria influencing the client’s choices in selecting procurement systems Part III — characteristics of procurement selection criteria uncovers the interviewees’ perceptions of the subjectivity of procure- ment selection criteria. The protocol was piloted by two experts not participating in the final interviews. The pilot studies revealed that some questions were ambiguous, while others might lead to bi- ased responses. For instance, some terms in the protocol were not clearly understood by the experts during the pilot studies. The protocol was therefore edited to address the above issues, and definitions of some key terms were incorporated. RESPONDENTS’ PROFILES Five Australian client organisations were identified that agreed to participate in this study. These included one private and four public client organisations. All interviewees were responsible for managing the pro- curement of construction works in their or- ganisation (Table 2). The results of the interviews confirmed that all respondents always performed procure- ment selection for their construction pro- jects. They all recognised the need to select an appropriate procurement system for each construction project, satisfying the time, cost, quality and risk requirements. This indicated that the respondents pos- sessed a good understanding of procure- ment selection. The above finding was confirmed by the revelation of the respondents’ practical ex- perience in procurement selection. As shown in Table 2, except for respondent B, who had been involved in procurement se- lection for almost 10 years, all other re- spondents had over 15 years experience in procurement selection. To further establish the suitability of the in- terviewees, indirect assessments of respon- dents’ knowledge on various procurement systems were conducted. Issues like the different types of procurement systems available in Australia, their advantages and disadvantages, and their application in dif- ferent circumstances were raised with each respondent. It was found that all respon- dents had comprehensive knowledge and experience of all those issues, and that they were suitable for this study. PROCUREMENT SELECTION PRACTICE All organisations surveyed had their own procedures for procurement selection. Re- spondent A indicated that many consultants were invited to advise on project character- istics, such as design, complexity, budget costing, and special management require- ments. Their advice serves as the basic in- formation for procurement selection. In- house experts then examine the organisa- tion’s requirements and, based upon their experience, determine the most appropriate procurement system for the project. Table 2: Details of interviewees Ref. Type of organisation Position of interviewees Experience A Private client Manager, Dept. of Planning and Physical Estate > 15 years B City council Manager, Dept. of Project Management 10 years C City council Manager, Procurement Dept. > 15 years D City council Manager, Procurement Strategist Dept. > 20 years E Road authority Manager, Dept. of Project Management > 15 years S. THOMAS NG, DUC THANH LUU AND SWEE ENG CHEN 74 THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 Figure 1: Considerations in procurement selection Procure ment selectio n Orga nisational procedure s In-house expertise Consultant’s advice Previous expertise According to organisations A, B, C and E, the traditional procurement approach was still the more favourable choice for small projects. For organisations B and C, small projects were those valued at less than $300,000, while small projects were defined by organisation E as those valued at less than $1,000,000. According to the respon- dents, small projects were normally rather straightforward, and did not require a great deal of design and management skills from the contractors. For projects above the stated value limits, a formal process of pro- curement selection was required. A variety of issues such as organisational policies, financial regulations, advice from in-house experts, advice from external consultants, previous experience, risks and quality as- surance would be considered. Only organisation D had occasionally at- tempted to adopt the concepts of theoretical procurement selection as proposed by re- searchers (e.g. Alhazmi and McCaffer, 2000; Franks, 1990; Griffith and Headley, 1997; Skitmore and Marsden, 1988). However, such application is limited and has never been taken seriously. PROCUREMENT SELECTION CRITERIA A list of procurement selection criteria as suggested by the respondents is presented in Table 3. These criteria include speed, time certainty, price certainty, complexity, flexibility, responsibility, risk allocation, quality level, price competition, public ac- countability, client requirement and political issues. Compared with the list of procurement se- lection criteria found in relevant literature (e.g. Bennett and Flanagan, 1983; Hewitt, 1985; Masterman and Duff, 1994; NEDO, 1985; Skitmore and Marsden, 1988; Singh, 1990), the current findings are very similar to those identified in previous studies. The only differences were public accountability, political issues and client requirements, which were not emphasised in previous studies, whereas disputes and arbitration were not suggested in this study. Respon- dent B explained that public accountability and political issues were considered during procurement selection as the local govern- ment was required to demonstrate account- ability to the community. As a result, the selected procurement system should be in favour of public accountability criteria such as cost reduction and environmental friendliness. DECISION CRITERIA AND THEIR SUBJECTIVITY IN CONSTRUCTION PROCUREMENT SELECTION THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 75 Table 3: Criteria used for procurement selection Respondents Criteria A B C D E Speed a a a a a Price certainty a a a a a Time certainty a a a a a Complexity a a a a a Flexibility a a a a a Responsibility a a a a a Quality level a a a a a Risk allocation a a a a a Price competition a a a a a Others Public accountability Client's requirements, political issues Client's requirements, political issues Note: a represents common selection criterion from the respondent’s perspective Table 4: Subjectivity of procurement selection criteria Respondents Criteria A B C D E Speed a a a a a Price certainty a Time certainty a Complexity a a a a a Flexibility a a a a a Responsibility a a a a a Quality level a a a a a Risk allocation a a a a a Price competition a a a a a Note: arepresents subjective criterion from the respondent’s perspective Respondents C and D also considered politi- cal issues (along with client requirements) as important criteria for procurement selec- tion. Being the public agents representing other governmental departments in con- struction works, their decisions or their cli- ents’ decisions in procurement selection were largely affected by governments’ po- litical policies. Although disputes and arbitration were not directly mentioned by the respondents in the interviews, as key procurement selection criteria, these factors had been implied by the respondents as part of risk allocation. For instance, respondent E stated that there was a need to allocate risks or manage safety issues upfront so that when problems occurred they could be resolved easily. Re- spondents C and D conceived disputes and arbitration as components of risk allocation as they pointed out the disadvantages of traditional lump sum procurement when disputes occur. They also advocated that design and build approaches favoured cli- ents during the disputes and arbitration processes as all the risks would have been transferred to the contractor. SUBJECTIVITY OF PROCUREMENT SELECTION CRITERIA Respondents were asked to express their perceptions on the subjectivity of the deci- sion criteria identified, and the results are summarised in Table 4. Four out of five respondents believed that time certainty and price certainty could be measured objectively. Time certainty was unequivocal as the completion date could be reasonably predicted by measuring the job requirements. Likewise, price certainty was considered as an objective criterion as price can also be reasonably predicted before- hand. The certainty of price can then be S. THOMAS NG, DUC THANH LUU AND SWEE ENG CHEN 76 THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 measured against the contractor’s require- ments. All respondents believed that other selec- tion criteria including speed, complexity, flexibility, responsibility, quality level, risk allocation and price competition were vague and subjective. These findings have sup- ported some researcher’s assertions (e.g. Rowlinson and McDermott, 1999; Cheung et al., 2001) that some procurement selection criteria are intangible in nature. Speed Respondent D claimed that speed was an arbitrary factor, which depended very much on the client’s situations. The speed to be specified depends on the level of extra pre- mium a client is prepared to pay for speed- ing up the design and construction processes. The definition of high speed may therefore differ from one client to another. Respondent B suggested that speed might be determined by contractor’s experience. A contractor with ample experience in the prospective project type and construction method may complete the work in a much quicker time. As a result, the level of speed as specified by the client may not totally cor- respond to what the contractor can offer. Complexity All respondents believed that complexity was very difficult to define, and the definition usually varied from person to person. Re- spondent C suggested that the opposite of complexity was routine, repetitiveness or standardisation. As a result, a project could be very simple for someone who has done a similar job before, but extremely complex for someone with no prior experience of the project type. According to respondent A, complexity was vague as, apart from the complexity of physical design and work method statements, complexity could be caused by the public and/or people involved, and these are rather difficult to predict. The views of respondent A concur with those of Turner (1990) who claimed that complexity was a non-quantifiable criterion since it could not be clearly depicted in the specifi- cation. Flexibility All respondents conceived flexibility to be subjective. A good explanation was provided by respondent B who claimed that, in a construction project, flexibility depends on human factors such as stakeholders’ situa- tion, experience and competence, and non- human factors such as project type, project situations, external factors (weather, strikes, political impact). He added that as both human and non-human factors are unpredictable and very difficult to manage, flexibility could become relatively ambigu- ous to the decision-makers. Quality All respondents shared the same view that quality could be difficult to measure objec- tively, as it was partially determined by vague standards such as form, commodity, delight, and comfort. Turner (1990) claimed that the quality of the design concept may not be easily determined in the specification, and may sometimes involve expert opinions. Responsibility Responsibility was not considered to be ob- jective since there is no fixed definition as to what is a high, medium or low level of client involvement in a project, instead the level of responsibility varies from project to project. Respondent C claimed that it was a normal practice to get actively involved in a project for as much as 50% of the total project time to gain the best results for the project. On the other hand, respondent A believed that allocating 50% of his total time to a project would be rather high. Risk allocation Four out of five respondents indicated that an interlacing relationship exists between risk allocation and responsibility, as the more responsibility one has been assigned in a construction project, the more risk one would have to assume. Not only did they think that responsibility was not fully struc- tured, they also believed risk allocation could not be measured objectively. Respon- dent E elaborated by saying that when he transfers 50% of the total risk that might occur in a project to the contractor, he re- gards that amount of risk transfer as a high risk allocation and feels safe. On the other hand, a client who does not have any experi- ence or knowledge in construction might think that it is necessary to transfer up to 90% of risk (i.e. a high risk allocation) to the contractor in order to feel comfortable. DECISION CRITERIA AND THEIR SUBJECTIVITY IN CONSTRUCTION PROCUREMENT SELECTION THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 77 Table 5: Procurement selection methods Methods Authors Operational Research Procurement path decision chart NEDO (1985) Procurement rating system Franks (1990) Multi-attribute approach Singh (1990) Bennett and Grice (1990) Ambrose and Tucker (1999) Weighted score model Griffith and Headley (1997) Analytical hierarchy process Alhazmi and McCaffer (2000) Multi-attribute utility approach Cheung et al. (2001) Statistical Discriminant approach Skitmore and Marsden (1988) Computerised Rule-based expert computer system (ELSIE) Brandon et al. (1988) Price competition The respondents believed that price compe- tition was vague, as the definitions of low or high price competition would vary with dif- ferent clients. Respondent A regarded a saving of 10% of the originally estimated project sum due to competitive tendering activities as high price competition, while respondent E indicated that he expected up to 20% saving to qualify as high price com- petition. Respondent B suggested that around 15% could be reasonably seen as high price competition. DISCUSSION Over the last two decades several theoreti- cal selection models have been introduced with the aim of improving the objectiveness of procurement selection. These methods can be classified into three main categories, namely operational research, statistical and computerised models (Table 5). While these models provide the means to improve the decision process, they fail to address the subjective characteristics of certain procurement selection criteria, which are used as primary input in these models. In fact, subjective criteria are usu- ally linguistic in nature, which may contain a certain level of vagueness (fuzziness) in the description of semantic meanings (Zimmermann, 1991). Consequently these criteria may not be adequately handled by traditional probability theory, which assumes a precise definition of the situa- tions to be dealt with (Kolmogoroff, 1956). Since probability theory is adopted in some procurement selection methods, there is a possibility that those methods may not properly capture the vagueness of the se- lection criteria used, and the decisions de- rived by these methods may be prone to error. To illustrate the effects of misinterpreting a subjective criterion in the assessment proc- ess, an example based upon Skitmore and Marsden’s multi-attribute approach is pro- vided here. Assuming “complexity” as the only subjective criterion involved in the as- sessment, as a principle of probability the- ory, the client needs to select a priority rating scale (say from 1 to 20) to represent his/her perception on how complex the pro- ject would be. To reflect his/her perception on a highly complex project, the client may select a priority scale of, for instance, 15, 17, 20 or any other large number within the range. Tables 6 and 7 illustrate the pre- ferred procurement options based on a pri- ority scale for “complexity” of 17 and 20 respectively. With other selection criteria being equal, a different perception on “high complexity” could yield very different re- sults. In this example, Procurement System E was the most preferred option should “high complexity” be interpreted as 17 (Ta- ble 6), while Procurement System B should be chosen if a priority scale of 20 was used (Table 7). As a result, the client may end up with different recommendations if there were different interpretations of the mean- ing of “high complexity”. S. THOMAS NG, DUC THANH LUU AND SWEE ENG CHEN 78 THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 Table 6: Result of procurement selection when priority rating for “high complexity” is 17 Procurement paths A B C D E F G Client's priority criteria Client's priority rating Rationalise priority rating Utility result factor Utility result factor Utility result factor Utility result factor Utility result factor Utility result factor Utility result factor Speed 11 0.14 40 5.4 10 1.4 60 8.1 100 13.6 90 12.2 110 14.9 110 14.9 Certainty 12 0.15 30 4.4 30 4.4 70 10.4 100 14.8 100 14.8 10 1.5 110 16.3 Flexibility 5 0.06 110 6.8 110 6.8 40 2.5 40 2.5 40 2.5 90 5.6 10 0.6 Quality level 14 0.17 110 19.0 110 19.0 40 13.8 40 6.9 40 6.9 90 15.6 20 3.5 Complexity 17 0.21 100 21.0 100 21.0 50 14.7 50 10.5 50 10.5 110 23.1 20 4.2 Responsibility 12 0.15 30 4.4 30 4.4 100 10.4 100 14.8 100 14.8 10 1.5 110 16.3 Price competition 10 0.12 20 2.5 110 13.6 10 9.9 10 1.2 80 9.9 40 4.9 30 3.7 Totals 81 1.00 63.6 70.6 69.8 64.3 71.6 67.0 59.5 Rank order 6 2 3 5 1 4 7 Table 7: Result of procurement selection when priority rating for “high complexity” is 20 Procurement paths A B C D E F G Client's priority criteria Client's priority rating Rationalise priority rating Utility result factor Utility result factor Utility result factor Utility result factor Utility result factor Utility result factor Utility result factor Speed 11 0.13 40 5.2 10 1.3 60 7.9 100 13.1 90 11.8 110 14.4 110 14.4 Certainty 12 0.14 30 4.3 30 4.3 70 10.0 100 14.3 100 14.3 10 1.4 110 15.7 Flexibility 5 0.06 110 6.5 110 6.5 40 2.4 40 2.4 40 2.4 90 5.4 10 0.6 Quality level 14 0.17 110 18.3 110 18.30 80 13.3 40 6.7 40 6.7 90 15.0 20 3.3 Complexity 20 0.24 100 23.8 100 23.8 70 16.7 50 11.9 50 11.9 110 26.2 20 4.8 Responsibility 12 0.14 30 4.3 30 4.3 70 10.0 100 14.3 100 14.3 10 1.4 110 15.7 Price competition 10 0.12 20 2.4 110 13.1 80 9.5 10 1.2 80 9.5 40 4.8 30 3.6 Totals 84 1.00 64.9 71.7 69.8 63.8 70.8 68.6 58.1 Rank order 5 1 3 6 2 4 7 DECISION CRITERIA AND THEIR SUBJECTIVITY IN CONSTRUCTION PROCUREMENT SELECTION THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 79 CONCLUSION The selection and use of an appropriate pro- curement system is crucial to project suc- cess. This paper aims to improve our understanding of the commonly used pro- curement selection criteria and the objec- tiveness of those criteria. The results indicate that there are nine procurement selection criteria commonly used by Austra- lian clients: speed, time certainty, price certainty, complexity, flexibility, responsibil- ity, quality level, risk allocation and price competition. Only time certainty and price certainty were seen by the respondents as unambiguous criteria, as the completion date and price can be objectively predicted by the client beforehand. However, the other seven were regarded by the experts as subjective. An example has been presented to illustrate the effects of misinterpreting “high complexity” in a multi-attribute procurement selection model. The results indicate that different perceptions, as reflected by various priority ratings, would yield different recommendations for procurement system. This clearly does not improve the objectiveness of procurement system selection. The requirements for linguistic input for some criteria justify the use of the fuzzy set theory (Zadeh, 1965). Fuzzy set theory is the key to decision-making when encountering vague conceptual phenomena. It has been applied to various construction manage- ment decision models involving the use of vague input variables, such as project scheduling (Ayyub and Haldar, 1984; Lorterapong and Moselhi, 1996), tender evaluation (Nguyen, 1984), contractor evaluation (Russell, 1992), and prediction of contractor failure (Russell and Jaselskis, 1993). Research into the application of fuzzy set theory to construction procurement se- lection is being conducted by the authors, and the results of this study will be reported when they become available. REFERENCES Alhazmi, T. and McCaffer, R. (2000) Project procurement system selection model. Jour- nal of Construction Engineering and Man- agement, ASCE, 126 (3), 176–184. Ambrose, M.D. and Tucker, S.N. (1999) Matching a procurement system to client and project needs: a procurement system evaluator. In: Bowen, P.A. and Hindle, R.D (eds.) Proceedings: Customer Satisfaction: A Focus for Research and Practice in Con- struction, University of Cape Town, South Africa,. 280–288. Ayyub, B.M. and Haldar, A. (1984) Project Scheduling Using Set Concepts. Journal of Construction Engineering and Management, ASCE, 110 (2), 189–204. Bennett, J. and Grice, A. (1990) Procurement systems for building, Quantity Surveying Techniques — New Directions, Brandon, P.S.(ed.). BSP Professional Books, Oxford. Bennett, J. and Flanagan, R. (1983) For the good of the client. Building, 1st April, 26–27. Brandon, P., Basden, A. and Hamilton, I.W. (1988) Expert System: The Strategic Plan- ning of Construction Projects. Royal Institu- tion of Chartered Surveyors and University of Salford, Salford University Press, Salford. Burgess, R.G. (1989) In the Field: An Intro- duction to Field Research. Onwn Hyman, London. Cheung, S.O., Lam, T.I., Wan, Y.W. and Lam, K.C. (2001) Improving objectivity in procure- ment selection, Journal of Management in Engineering, ASCE, 17 (3), 132–139. Chua, D.K.H., Kog, Y.C. and Loh, P.K. (1999) Critical success factors for different project objectives. Journal of Construction Engi- neering and Management, ASCE, 125 (3), 142–150. Fellows, R.F. (1993) Contracts for Refur- bishment. School of Architecture and Build- ing Engineering, University of Bath Press, Bath. Franks, J. (1990) Building Procurement Sys- tems — A Guide to Building Project Man- agement. Chartered Institute of Building, Ascot. Griffith, A. and Headley, J.D. (1997) Using a weighted score model as an aid to selecting procurement methods for small building projects. Construction Management and Economics, 15 (4), 341–348. Hewitt, R.A. (1985) The procurement of buildings: proposals to improve the per- formance of industry. Report to the College of Estate Management, UK. Hibberd, P.R. and Djebarni, R. (1996) Criteria of choice for procurement system. In: Pro- ceedings: COBRA ’96, 19–20 September, S. THOMAS NG, DUC THANH LUU AND SWEE ENG CHEN 80 THE AUSTRALIAN JOURNAL OF CONSTRUCTION ECONOMICS AND BUILDING VOL.2 NO.1 University of the West of England, Royal In- stitution of Chartered Surveyors. Kolmogoroff, A. (1956) Foundation of the Theory of Probability. Chelsea, New York. Lorterapong, P. and Moselhi, A. (1996) Pro- ject network analysis using fuzzy sets the- ory, Journal of Construction Engineering and Management, ASCE, 122 (4), 308–320. Masterman, J.W.E. (1992) An Introduction to Building Procurement Systems. E and FN Spon, London. Masterman, J.W.E. and Duff, A.R. (1994) The selection of building procurement systems by client organizations. In: Skitmore, R.M. and Betts, M. (eds) Proceedings: 10th An- nual ARCOM Conference, Vol. 2. Loughbor- ough University of Technology, Association of Researchers in Construction Manage- ment, Leicestershire, 14–16 September, 650–659. Masterman, J.W.E. and Gameson, R. (1994) Client characteristics and needs in relation to their selection of procurement systems. In: Rowlinson, S. (ed.) Proceedings: “East Meets West” Procurement Systems Sympo- sium, CIB Publication 175, 4–7 December, Hong Kong, 79–87. Nahapiet, H. and Nahapiet, J. (1985) The Management of Construction Projects, Case Studies from the UK and USA. Chartered Institute of Building, Ascot. NEDO (1985) Thinking About Building, Na- tional Economic Development Office. HMSO, London. Nguyen, V.U. (1985) Tender Evaluation by Fuzzy Sets. Journal of Construction Engi- neering and Management, ASCE, 111 (3), 231–243. Rowlinson, S. and McDermott, P. (1999) Procurement systems: A guide to best prac- tice in construction. E and FN Spon, London. Russell, J.S. (1992) Decision models for analysis and evaluation of construction con- tractors. Construction Management and Economics, 10 (3), 185–202. Russell, J.S. and Jaselskis, E.J. (1993) Pre- dicting construction contractor failure prior to contract award. Journal of Construction Engineering and Management, ASCE, 118 (3), 612–624. Sharif, A. and Morledge, R. (1994) A func- tional approach to modelling procurement systems internationally and the identifica- tion of necessary support frameworks. In: Rowlinson, S. (ed) Proceedings: “East Meets West” Procurement Systems Symposium, CIB Publication 175, 4–7 December, Hong Kong, 79–87. Skitmore, M. and Marsden, D.E. (1988) Which procurement system? Towards a uni- versal procurement selection technique. Construction Management and Economics, 6 (1), 71–89. Singh, S. (1990) Selection of appropriate project delivery system for building con- struction projects In: Proceedings: CIB-90 Building Economics and Construction Man- agement, University of Technology Sydney, 469–480. Turner, A. (1990) Building Procurement. Macmillan, London. Zadeh, L.A. (1965) Fuzzy sets. Information and Control, 8, 338–353. Zimmermann, H. J. (1991) Fuzzy Set Theory and Its Applications, 2nd edition. Kluwer Academic Publishers, USA.