Engineering, Construction and Architectural Management

Evaluating performance potential in the selection of construction contractors

GARY D. HOLT, PAUL O. OLOMOLAIYE, FRANK C. HARRIS,
Article information:
To cite this document:
GARY D. HOLT, PAUL O. OLOMOLAIYE, FRANK C. HARRIS, (1994) "Evaluating performance
potential in the selection of construction contractors", Engineering, Construction and Architectural
Management, Vol. 1 Issue: 1, pp.29-50, https://doi.org/10.1108/eb020991
Permanent link to this document:
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

https://doi.org/10.1108/eb020991
Downloaded on: 22 February 2018, At: 01:23 (PT)
References: this document contains references to 0 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 555 times since 2006*
Users who downloaded this article also downloaded:
(1996),"Tendering procedures, contractual arrangements and Latham: the contractors' view",
Engineering, Construction and Architectural Management, Vol. 3 Iss 1/2 pp. 97-115 <a href="https://
doi.org/10.1108/eb021025">https://doi.org/10.1108/eb021025</a>
(2002),"A multi-criteria approach to contractor selection", Engineering, Construction and
Architectural Management, Vol. 9 Iss 1 pp. 29-37 <a href="https://doi.org/10.1108/eb021204">https://
doi.org/10.1108/eb021204</a>

Access to this document was granted through an Emerald subscription provided by emerald-
srm:535196 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald
for Authors service information about how to choose which publication to write for and submission
guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well
as providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the
Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for
digital archive preservation.

*Related content and download information correct at time of download.

 Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

Evaluating p e r f o r m a n c e potential in t h e
s e l e c t i o n of c o n s t r u c t i o n c o n t r a c t o r s
GARY D. HOLT, PAUL O. OLOMOLAIYE & FRANK C. HARRIS
School of Construction Engineering & Technology, University of Wolverhampton, Wulfruna Street,
Wolverhampton, West Midlands WV1 1SB, UK
Paper submitted 6 October 1993; accepted for publication 6 May 1994; discussion open until
March 1995
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

Abstract A quantitative contractor selection technique which embraces the pre-

qualification, evaluation and final selection processes is being developed. The emphasis is
on evaluating contractors' performance potential in terms of their ability to achieve time,
cost and quality standards. This approach is in contrast to the majority of current selection
techniques which tend to prequalify, then discriminate predominantly on the cost com-
ponent of tenders.The conceptual model is applied to a hypothetical but realistic scenario of
a contractor competing for a small industrial contract. This illustrates the mechanics of the
new technique, emphasizing that contractor selection should include identifying the
contractor with the best performance potential and not merely the lowest bidder.

Keywords Decision making, contractor performance, contractor selection, multi-

attribute analysis

INTRODUCTION
Construction is normally 'purchased' before it is 'manufactured', therefore the
more usual methods of pre-purchase product appraisal (comparison, approval,
etc.) cannot be applied. Rather, the construction client must evaluate the
'manufacturers' of such products and repose in one (the chosen contractor) the
confidence to satisfactorily execute the construction contract. Subsequent success
or failure of the construction process (a function of contractor performance)
determines the overall level of satisfaction attained from the exercise by the client
(Mohsini & Davidson 1986). That existing contractor selection methods are
unable to predict a successful project outcome in terms of client satisfaction is
evidenced by the number of clients now utilizing alternative, predominantly
package deal, methods of procurement (Sullivan & Harris 1986). Clients' satis-
faction requirements may be broadly defined in terms of:
• Time - to utilize the product as soon as possible
• Cost - to purchase the product at the lowest possible price
• Quality - for the product to be of the highest standard (Odusote, 1990).
Cost is obviously of prime concern to the client. He/she will always be seeking to

29
 30 Holt, G.D. et al.

balance this with project time and quality to secure all-round best value for
money. Any combination of two of these superlative criteria can normally be
achieved, but rarely if ever, all three (Watkinson 1992). Therefore, an effective
contractor selection technique must necessarily evaluate candidate contractors in
the context of their optimal ability in these performance requirements. Unfortu-
nately, current methods often fail in this objective. Discrimination between
contractors tends to be based predominantly on tender sum (Baker & Orsaah
1985; Merna & Smith 1990; Griffith 1992), but the lowest bid is not necessarily
the most economic solution in the long-term. Furthermore, current selection
practices exhibit inherent weaknesses which are not conducive to comprehensive
potential performance evaluation (Holt et al. 1993a).
Hence, the purpose of this paper is to present and illustrate, with a fully worked
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

example, a technique designed to meet this objective of potential performance

evaluation. The technique is applied to a hypothetical selection exercise involving
a small industrial project, to elucidate comprehensively the mechanics of the
method. Specifically, the fortunes of one particular contractor are followed from
invitation to prequalify through to final selection.

AIM OVERVIEW OF THE MODEL

Developing a selection model was essentially a mission to create a decision tool,
the core of the decision problem being rational evaluation. Mascoll (1984)
contended that 'given perfect rationality the solution to any problem is strictly
determinate', but this may be questionable in the case of stochastic or inde-
terminate scenarios. However, in this instance rational evaluation is made difficult
by the interaction of the objectives of the client and the attributes of the
contractors (Diekmann 1979). Because the outcome of the evaluation exercise
cannot be immediately characterized in terms of a single value and because of the
need to take account of many outcome dimensions, the multi-attribute analysis
(MAA) technique has been employed (Moore & Thomas 1979; Skitmore &
Marsden 1988; Russell 1992). MAA facilitates the making of decisions in the
presence of multiple, usually conflicting criteria (Hwang & Yoon 1981), thereby
addressing the nature of real world decision-making (Mascoll 1984).

The multi-attribute analysis technique

An MAA problem may be concisely expressed in a matrix format (Hwang & Yoon
1981), the element xijs of such a matrix indicating the respective results of
evaluation. That is, the value of alternative i, Ai with respect to attribute j, X j .
Hence, A i , i = 1, 2, 3,...m may be represented horizontally within the matrix by:
xi = (xi1 x i 2 ,...x i n ) and the column vector, xj = (x1j, x 2 j ,...x m j ) shows the value of
each alternative in regard to attribute j, X j .
This concept may be illustrated by example; assume contractors A1 to A 4 (Ai)
are to be evaluated for the attributes X1 to X 6 (Xj). The attributes are: years in

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 The selection of construction contractors 31

Table 1 MAA decision matrix for alternatives Ai in respect of attributes Xj

Attributes (Xj)

Alternatives (Ai) X1 X2 x3 X4 x5 X6

A1 2 £0.9m Yes Good No 3 months

A2 4 £1.2m Yes Poor No Instantly
A3 7 £3.0m Yes Good Yes Instantly
A4 6 £1.7m No Very good No 2 months

business (X 1 ); turnover (X 2 ); experience of the required work type (X 3 ); health

and safety record (X 4 ); failure to have completed a contract (X 5 ); and availability
(X 6 ). Then, based on the evaluation of contractors Ai for each of the attributes X j
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

a decision matrix may be compiled as shown in Table 1. It should be noted at this

point that the matrix shows a fundamental problem of MAA - that of incom-
mensurable units. However, it will become clear that within the selection model
this has been overcome by a series of independent evaluations, facilitating con-
version of a contractor's attribute scores (in terms of natural unit values) into a
commensurable scale of measurement (between zero and 1.0).
Having evaluated the decision alternatives to attain values for each of the
attributes, the optimal solution (decision alternative) to the selection problem
(Ai⋆) would be that which resulted in the maximum value for each of its objective
functions within the matrix (xij⋆) simultaneously. That is, xij⋆ is an element of
attribute X j in respect of alternative Ai⋆ and all functions of Xij⋆ ≥ all functions of
xij for remaining alternatives A i .
Hence, Ai⋆ is the optimal solution ifxij⋆εX j and fxij⋆ ≥ fxij for all remaining x
εXj.
Unfortunately, such a solution is only achievable by chance. The matter is
further complicated by conflict between attributes often necessitating a 'trade off'.
For example, the objective adequate resources may conflict with the attribute
financial stability — the former normally placated when a contractor has spare
workload capacity, the latter often generated from a full order book! Hence,xij⋆is
seldom achievable for all x ε Ai⋆. Therefore, the ideal solution (SAi⋆) will consist
of objective functions (sij⋆) where sij⋆= (f1⋆,f2⋆,.. .f3⋆), when fj⋆ is an optimal but
feasible value. However, even this may be hard to achieve due to the conflict
mentioned.
This results in the best solution being a subjective one in that it is composed of
the most preferable values within the matrix, i.e.
SA i * = Σs ij *
where sij* = (f1*, f2*,...fn*) when fn* = max Uj(fij) (i = 1,2,...m) (Eq.1)
when U j indicates the value/utility function of the jth attribute. These concepts are
applicable to MAA generally.
In designing the selection model, attention was directed towards addressing the
prime weaknesses of current UK selection practices. These may be summarized as:

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 32 Holt,G.D.et al.

1 Lack of a universal (standard) approach

2 Long-term confidence placed in contractors based solely on prequalification
3 Over-reliance on subjective data/analysis
4 Reliance on tender sum as the predominant discriminatory factor.
A more comprehensive description of these weaknesses maybe found in Holt et
al. (1993a, b).
In contrast to the majority of current selection practices and research to date
(which tend to rely upon prequalification as a means of substantiating acceptance
of the lowest bid), a 'stepwise logic' is inherent within the model to accommodate
the entire selection process - from prequalification of contractors' desirous to
tender, through to optimum selection choice (Fig. 1).
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 The selection of construction contractors 33

The model furnishes objective, relative, numeric values for each contractor
evaluated to serve as unambiguous references in the decision process. These
values are in the form of combined (P) scores, calculated at both prequalification
and tenderer evaluation stages. A final ranking (P3) identifies optimal choice (Fig.
1). P scores are achieved by expressing the sum of a contractor's attribute values,
i.e. variable scores (the Z score) as a percentage of the maximum attainable sum of
scores (ZMax), i.e. P score = Zscore/ZMax. Final ranking is a numerical amalgam
of P2 score and bid value (£). Because the model evaluates time, cost and quality
contractor attributes, combined scores mirror performance potential in the
context of a contractor achieving client satisfaction, by way of satisfactory project
performance for each of these performance standards - time, cost and quality.
The three basic functions of the model are:
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

1 The prequalification component

2 The tenderer evaluation component
3 The final selection component.
We may summarize these functions as in Table 2. Each function is now
elucidated in turn.

Prequalification component - P1 analysis

The potential performance score (P1) is designed to facilitate prequalification of
contractors desirous to tender. It investigates the more general organizational
attributes of a contracting company in light of several discriminating criteria.
More specific P2 criteria (see later) are not considered during prequalification,
because if a contractor were below par in such fundamentally important areas as
'financial stability' or 'past experience' (i.e. P1 criteria) the firm's potential in
project specific variables such as 'plant resource available for the project' (P2
criterion) would be irrelevant.
P1 replaces the traditional pre-selection task of establishing a 'select list', and
instead incorporates prequalification as an essential and integral part of every
selection exercise, thereby furnishing the practitioner with up-to-date perfor-
mance potential of a contractor. This contrasts with current trends which place
long-term confidence in prequalification. (Should a contractor remain on a select

Table 2 Descriptive summary of the model

Identification within
'Logical' stage model Component purpose Component output

Prequalification P1 analysis Identify contractors %/Rank

component worthy of invitation to
tender

Tender evaluation P2 analysis Furnish extra dimension %/Rank

component in tender evaluation

Final selection P3 score Identify best contractor %/Rank

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 34 Holt, G.D. et al.

list for an appreciable amount of time it is conceivable that the firm may witness
negative changes within its corporate structure - liquidity, resources, etc., parti-
cularly in view of prevalent macroeconomic and market forces).
Theoretically, any number of contractors can be assessed and their respective
P1 score derived. Furthermore, there would appear scope for the 'experienced'
client with an on-going construction programme to develop a history of P1 scores,
to serve as a retrospective indicator of a contractor's past performance and
corporate stability trends. Mathematically P1 may be expressed as:

where Z1 j = score achieved under P1 analysis for contractor j and is determined

via the formula:
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

when:
• 2l = the twenty-one discriminating criteria attributed to P1 analysis
• Vij = variable scores achieved by contractor j for each variable
• Wi = importance weights attached to Vi.
Z1 Max is the maximum attainable Z score under P1 analysis. Maximum Vi
value is 1.0, therefore

It can be seen from the formula that a contractor who achieves a maximum
score in all variables Vi(i = 21) would have a P1 score of 1.0. Accordingly, a
contractor obtaining an average score of half would exhibit a potential perfor-
mance score of 50%, and so on. Be that as it may, one must not overlook that a
'respectable' combined score may be achieved by a contractor 'compensating'
poor scores for certain variables with 'good' scores in others. Hence, factor scores
(score achieved for a given set of variables - see later) may be derived and checked
for signs of specific contractor weakness. Factor scores are achieved via the
formula:

where rationalized variable scores:

1 — V score x respective weighting index for any given P1 variable
2 = V score x respective weighting index x respective utility weight for any given
P2 variable.
Derivation and checking of factor scores also addresses the problem that will
arise where any overlap of clusters ('selectable' and 'unselectable' contractors)
occurs. At this stage of the research no 'pass marks' have been established for
variable/factor/P scores, but this is currently being investigated.

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 The selection of construction contractors 35

Evaluating tenderers - P2 analysis

P2 evaluates further those contractors invited to tender because of their satis-
factory P1 scores. It is proposed that P2 analysis be performed whilst contractors
are compiling their tenders, thereby furnishing the practitioner with additional
information regarding potential performance by the time formal submissions are
received. This secondary investigative element has been influenced in part by the
more comprehensive existing selection regimes. Such regimes are predominantly
a feature of the civil engineering sector where qualified (and subsequently more
'complex') bids are allowed (National Joint Consultative Committee for Building
Codes, prohibiting such qualification in the building sector).
As per P 1 , P2 also employs a number of discriminating criteria, but these are
much more project specific in nature, the aim being to evaluate tenderers'
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

potential in relation to the proposed project. 'Project specific criteria can evaluate
contractors in terms of unusual expertise or specialist facilities required by the
project' (Russell 1992). Table 3 exhibits all P1 and P2 selection criteria included
in the model.
It should also be noted that P2 incorporates a utility weighting making use of
multi-attribute utility theory (Moore & Thomas 1979; Skitmore & Marsden 1988;
Moselhi & Martinelli 1990). If the practitioner/client is indifferent to all possible
outcomes and is willing to accept their effect regardless, then weighted criteria
alone are an appropriate decision technique. However, 'utility' allows the client's
perception of performance dimensions to be attached to each discriminating
criterion (Fig. 2), thereby increasing the probability that the selected contractor
will mirror the client's preferences as much as possible. This simultaneously
renders the model dynamic by responding to the specific circumstances pertaining
to a given project, and the owner's perception of a successful outcome. Mathe-
matically P2 score may be expressed as:

where Z2j = score achieved under P2 analysis for contractor j and is determined
via the formula:

when:
• 22...29 = the eight criteria attributable to P2 analysis
• V kj = variable scores achieved by contractor j under P2 analysis
• Wk = importance weights attached to Vk
• Uk = utility weights attached to Vk by practitioner/client.
Z2Max = maximum attainable Z score under P2 analysis. Because the max-
imum Vk and U k values are 1.0, then:

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 36 Holt,G.D.et al.

Table 3 Selection criteria and respective weighting indices

Weighting*
Factors/variables index (W1) Factor rank

P1 factor: Contractors organization

Variables V1 Size of company 0.501 4
V2 Age of company 0.435 5
v3 Company image 0.408 6
v4 Quality control policy 0.529 3
v5 Health and safety policy 0.583 1
v6 Litigation tendency 0.545 2

P1 factor: Financial considerations

Variables V7 Ratio analysis of financial accounts 0.631 4
v8 Bank reference 0.669 1
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

v9 Credit reference(s) 0.634 3

V10 Turnover history 0.667 2

P1 factor: Management resource

Variables V11 Qualification of company owners 0.676 3
v12 Qualification of key persons 0.648 4
V 13 Key persons - years with company 0.695 2
V14 Extent of formal training regime 0.814 1

P1 factor: Past experience

Variables V15 Type of past projects completed 0.735 3
V16 Size of past projects completed 0.851 1
V17 National or local catchment 0.748 2

P1 factor: Past perfor mance

Variables V18 Failure to have completed a contract 0.679 1
V 19 Contract overruns: time 0.541 4
v20 Contract overruns: cost 0.576 3
v21 Actual quality achieved 0.667 2

P2 factor: Project specific

Variables V 22 Geographic experience in location of project 0.409 5
v23 Experience similar construction to proposed 0.564 2
project
Plant resource available for project 0.486 4
v24
v25 Key persons available for project 0.547 3
v26 Qualification of these key persons 0.673 1

P2 factor: Other specific

Variables V27 Calculated workload for project duration 0.862 1
v28 Prior relationship with client/consultant 0.651 2
v29 Home office location in regard to proposed 0.642 3
project

* Derived from Holt etal.(1993c).

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 The selection of construction contractors 37
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

Calculation of P1 and P2 variable scores

Within the model, selection variables of like nature are grouped under generic
heads; 'factors' and each group of 'variables' attributed to each factor are eval-
uated by way of several 'subvariables'. Figure 3 graphically illustrates this. P1
analysis consists of five factors, 21 variables and a possible 65 subvariables. P2
analysis evaluates two factors, eight variables and 20 possible subvariables. There
is future scope for the number/nature of selection variables to be varied and/or
broadened, to placate the requirements of alternative selection situations, e.g.
differing procurement forms. This potential is discussed in the summary.
Discriminating variables and respective importance weights (Wi) included in
the model were established via a nationwide survey of contractor selection prac-
titioners and client groups. The fundamental aim of the survey was to confirm
those variables to include in the model and their overall significance in the context
of selecting a contractor. Therefore, Wi consists of two elements:
1 Practitioners' importance response (IR)
2 Problem response (PR).

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

38 Holt, G.D. et al.
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

IR represents perceived importance and was measured by asking respondents to

score each variable 1 to 3 where: 1 = no importance; 2 = some importance; and 3
= critical importance. Aggregate response was converted to relative importance
using the relative ranking technique (Olomolaiye et al. 1987). PR considers how
often each variable was considered to be the cause of clients' dissatisfaction with
contractors' performance. Therefore, PR is expressed as a problem frequently in
relation to the number of contracts awarded by each respondent. The mean of all
such values for a given variable is its PR.
As IR ≤ 1.0 and PR ≤ 1.0 then W = 0.5 (IR + PR) and therefore Wi ≤ 1.0.
The basic philosophy behind consolidation of these two elements being that a
variable's significance in the selection process is a product of its intrinsic impor­
tance (IR) and potential to create problems for the client (PR). A more complete
description of the survey containing all these data can be found in Holt et al.
(1993c). (Refer to Table 3 for a summary of the weighting indices.)
During both P1 and P2 analysis a score between zero and 1.0 is ascertained in
respect of each variable by measuring the contractor against each variable's sub-
variables. Mathematically variable scores are calculated by the formula:
V X j = Sv x1j + Sv x2j + ... Sv xnj
where VXj = variable x score for contractorj (Eq.9)

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 The selection of construction contractors 39

Svx1j ... xnj = subvarifble scores attributable to variable x for contractor j

In any event:Σ S v x 1 j +... Svxnj ≤ 1.0. (Eq.10)
At this time subvariables are equally weighted, i.e.

The possible need to apply weightings to subvariables has not been discounted.
This will be confirmed over time by a detailed comparison of actual contractor
performance with the model prediction.
Space does not permit full explanation of how each individual variable is
quantified. This is fully elucidated, including pro-forma evaluation documenta­
tion, in sister publications (Holt et al. 1993d; 1994). However, an example is given
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

to outline the principle. Consider variable V 4 (quality control policy). Quality

control may be measured via three types of assessment:
1 'First party' assessment which involves the setting up of a quality system fol­
lowed by internal evaluation or quality assurance audit. Guidelines for such
have been produced by the Chartered Institute of Building (CIOB 1989)
2 'Second party' assessment which facilitates an input from the client to meet his/
her specific (product) needs
3 'Third party' assessment which is a comprehensive independent evaluation/
registration under the guidance of British Standard 5750 (BS 5750) (for
international equivalent see ISO, 9000) and which eliminates the need for
multiple assessments (Griffith 1990).
Adoption of quality assurance is at present voluntary and therefore, where taken
up by a company, would seem to indicate a genuine desire to attain consistent
quality (Griffith 1990). Therefore, where BS 5750 certification exists, full third
party assessment has been successfully achieved, indicating that satisfactory
quality assurance procedures are being implemented. Such conclusions mean that
certification may be adopted as the yardstick by which this criterion is measured.
Full credit is given to a company where BS 5750 certification exists and is duly
awarded a maximum score (1.0). Confirmed intention to register is an indicator of
a positive company attitude towards quality and is subsequently scored [0.5].
Where neither attribute exists the firm is scored zero.

Identifying optimum choice - P3 score

P3 is a numerical amalgam, designed to combine the results of a contractor's
'project specific' evaluation with the firm's tender sum to arrive at a final selection
ranking. Previous works have demonstrated that both clients and contractors rate
cost as the most important factor in the winning and awarding of contracts (Baker
& Orsaah 1985; Merna & Smith 1990). Nonetheless, as highlighted earlier, total
discrimination on cost is foolhardy. It would seem reasonable, therefore, to assign
greater weight to tender sum than P2 score. This is proposed as 60% cost, 40%
P2, proportions in line with Hawwash (1991). However, individual clients may
wish to determine this balance for themselves.

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

40 Holt, G.D. et al.

At present, the authors have an open mind regarding the integration of P1

scores with P3. Tests are currently being conducted to determine the accuracy of
P3 with or without P 1 . Mathematically P3 score may be expressed as:
P3 score = 0.6 (BSj) + 0.4 (P2j) (Eq. 12)
where P2j = P2 score for contractor j expressed as a decimal
BSj = bid score for contractor j and is determined via the formula:

Calculating bid scores in this manner is based on the sole criterion of net value.
However, for large projects or where contract duration is at the contractor's
discretion, the client/practitioner may wish to broaden the analysis by considering
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

the time value of capital. In such instances the lowest bid may not necessarily give
the best return on the investment since it represents a series of payments over time
(Hardy et al. 1981). Further variables may be considered including:
1 Tender cash flow - by plotting the cumulative cash flow curve
2 Price adjustment - under fluctuation contracts the anticipated escalation rate
may be applied to 1
3 Positive cash flows - stemming from early completion, e.g. generation of
revenue.
Finally, mobilization payments, site establishment costs and retention monies
may all have some influence on financial analysis of the bid. By plotting all the
above and performing a discounted cash flow analysis (Smit 1978; Hardy et al.
1981; Mott 1992) then the smallest Net Present Value (NPV) will identify the
most financially attractive bid. NPV values may be converted to a bid score for
inclusion in the model via the formula:

for all contractors under review.

Having explained the model, we may now apply the concept in a hypothetical
scenario to elucidate its workings.

THE S C E N A R I O

For the purpose of this example a small industrial development is assumed. The
client estimates that the contract value will be in the region of £484 000 made up
as follows:
Building cost including light, power and heating services but
ignoring any effect of VAT/loose or special equipment @ £267/M2* 440000
Add 10% for external works/sundry items 44 000
£484 000
* mean value from Spons (1992).

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 The selection of construction contractors 41

It is important to appreciate that in practice, a more comprehensive cost

analysis is preferred because a realistic estimate by which tender submissions may
be compared is a vital component in the quality of selection decision made
(Hawwash 1991).

PREQUALIFICATION - P1 ANALYSIS
See Appendix A - P1 Summary Analysis Sheet. As shown, the variable scores
(V 1 -V 21 ) have been summarized and multiplied by their relevant weighting
indices (W 1 -W 2 1 ), producing a rationalized score for each attribute.
En route to calculating P 1 , we may divide the sum of rationalized scores for all
variables under each P1 factor, by the sum of weighting indices for the same
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

variables, to produce factor scores. This allows the practitioner to check for signs
of specific weakness in the contractor company (refer 'pass marks' earlier). Table
4 exhibits factor scores achieved by the contractor undergoing prequalification
(the maximum possible factor score is 1.0). For our contractor the factor financial
standing (consisting variables: ratio analysis, bank reference, credit references,
turnover history) ranks lowest with a factor score of 0.76. This does not seem a
particularly low score and therefore does not give cause for concern.

Table 4 Contractor undergoing prequalification - P1 factor scores

Factor Factor score Score as a % Factor score rank

Organizational attributes 0.887 88 1

Financial standing 0.767 76 5
Management resource 0.818 81 3
Past experience 0.792 79 4
Past performance 0.856 85 2

By dividing the sum of rationalized variable scores under each factor (2.662 +
1.995 + 2.318 + 1.850 + 2.110 = Z1score = 10.935) by the sum of weighting
indices for each factor (3.001 + 2.601 + 2.833 + 2.334 + 2.463 = Z1Max =
13.232), the contractor achieves a P1 score of 0.826. This combined score maybe
expressed as 8 3 % potential performance in terms of being an all round good
prospect for the job. We will assume for the purpose of this example that this score
leads to the contractor being invited to tender and, therefore, subsequently sub-
jected to P2 analysis. Obviously, where several contractors have undergone P1
analysis, selection of tenderers will primarily be dependent upon their scores
relative to each other.

TENDER EVALUATION - P2 ANALYSIS

Having been invited to tender, evaluation of the contractor in relation to project
specific factors may commence. See Appendix B - Tender Evaluation (P2)

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 42 Holt,G.D.et al.

Summary Analysis Sheet. As per P1 analysis, all variable scores (V 22 -V 29 ) have

been summarized and multiplied by their relevant weighting indices (W 2 2 -W 2 9 ).
In addition we now witness the introduction of utility weights (U 2 2 -U 2 9 ).
Utility weights mirror the practitioner/client's perception of the importance of
each variable in the context of the proposed project, such utility weights having
been extrapolated as in Fig. 2. In this example it can be seen that all variables
maintain a utility of 1.0 except plant resource (0.9), prior relationship (0.7) and
office location (0.6).
Similarly, as per P1 analysis, factor scores may be checked for signs of weakness
but none seem apparent; project specific criteria score is 0.81 and other specific
criteria score is 0.70. The sum of rationalized variable scores under each factor
(2.180 + 1.520 = Z2 score = 3.7) divided by the sum of weighting indices for each
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

factor (2.679 + 2.155 = Z2Max = 4.834) achieves a P2 score of 0.765, which may
be expressed as 77% potential performance in terms of the contractor being a
good prospect for this particular contract.
The resultant effect of the utility coefficients on the P2 score is worthy of more
detailed examination. Assume a contractor has been evaluated and has achieved
P2 variable scores as shown in column 1 of Table 5. We may now observe two
scenarios:

Scenario 1
Assumes that the contractor has scored well in those variables perceived as
important by the client, i.e. there is strong correlation (coefficient = 0.78) between
high variable scores (column 1) and high utility weights (column 2).

Scenario 2
Assumes that the contractor has not scored so well in those variables perceived as
important by the client, i.e. there is strong correlation (coefficient = -0.96)
between low variable scores (column 1) and high utility weights (column 3).

Table 5 P2 variable scores/utility weights

Contractors' Utility weights Utility weights

Variables V scores high correlation low correlation

V 22 Geographic experience in relation to project 0.95 1.00 0.75

V23 Experience similar construction to project 0.60 0.70 1.00
V24 Plant resource available for project 0.50 0.75 1.00
V 25 Key persons available for project 1.00 0.90 0.60
V 26 Qualification of these key persons 1.00 1.00 0.40
V 27 Calculated workload for project duration 0.70 0.60 0.90
v28 Prior relationship with client/consultant 0.50 0.40 1.00
v29 Home office location in regard to project 1.00 1.00 0.70

Between columns 1 and 2 there is a high degree of correlation between selection variables in which the contractor has
scored well and has high utility weights (coefficient 0.78). Between columns 1 and 3. the same utility weights are utilized but
their redistribution means that the correlation is much lower (coefficient-0.96).

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 The selection of construction contractors 43

P2 score calculations for both the above scenarios are shown in Tables 6 and 7,
respectively. It can be seen that a higher P2 score (64%) is achieved by the con-
tractor who has high variable scores for those criteria perceived as important by the
client. Conversely, a lower P2 score (58%) is yielded for the contractor who does
not score well in these same criteria. In short, a contractor exhibits greater
potential for achieving client satisfaction (i.e. a higher P2 score) where high
variable scores are achieved in those selection criteria that the client perceives as
important to his/her particular project.

Table 6 P2 score calculation. High correlation between high variable and high utility scores

Contractors' Importance Rationalized

Variable V score x weight x Utility weight = score
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

V 22 Experience geographic 0.95 0.409 1.00 0.388

V 23 Experience work elements 0.60 0.564 0.70 0.236
V 24 Plant policy 0.50 0.486 0.75 0.182
V 25 Key persons available 1.00 0.547 0.90 0.492
V 26 Qualification key persons 1.00 0.673 1.00 0.673
2.679 1.971

Factor score = 1.971/2.679 = 0.735

V 27 Workload capacity 0.70 0.862 0.60 0.362

V28 Prior relationship 0.50 0.651 0.40 0.130
V 29 Office location 1.00 0.642 1.00 0.642
2.155 1.134

Factor score = 1.134/2.155 = 0.526

Z2 score = (1.971 + 1.1341 = 3.105 Z2 Max = (2.679 + 2.155) = 4.834

Therefore, P2 score = 3.105/4.834 = 0.642 or 64%

Table 7 P2 score calculation. Low correlation between high variable and high utility scores

Contractors' Importance Rationalized

Variable V score x weight x Utility weight = score

V 22 Experience geographic 0.95 0.409 0.75 0.291

V 23 Experience work elements 0.60 0.564 1.00 0.338
V 24 Plant policy 0.50 0.486 1.00 0.243
V 25 Key persons available 1.00 0.547 0.60 0.328
V 26 Qualification key persons 1.00 0.673 0.40 0.269
2.679 1.469

Factor score = 1.469/2.679 = 0.548

V 27 Workload capacity 0.70 0.862 0.90 0.543

V 28 Prior relationship 0.50 0.651 1.00 0.325
V29 Office location 1.00 0.642 0.70 0.449
2.155 1.317

Factor score = 1.317/2.155 = 0.611

Z2 score = (1.469 + 1.317) = 2.786 Z2 Max = (2.679 + 2.155) = 4.834

Therefore. P2 score = 2.786/4.834 = 0.576 or 58%

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 44 Holt, G.D. et al.

FINAL SELECTION CHOICE - P3 SCORE

The formula for P3 score is 0.6 (bid score) + 0.4 (P2 score) where bid score =
lowest bid for project (£)/this contractor's bid for project (£). The lowest bid
submitted for this project is £475K and our contractor's bid is £486K, so bid
score is calculated by £475K/486K = 0.977.
Therefore, the contractor's P3 score is (0.6 x 0.97) (bid component) + (0.4 x
0.76) (P2 component) = 0.886. This final score may be expressed as 89%
potential performance in terms of our contractor being a good prospect for
meeting, time, cost and quality standards, based on the thorough three-stage
process of elimination/evaluation demonstrated.
Having followed the fortunes of one particular contractor, from invitation to
prequalify through to final P3 score, it is interesting to now explore how that
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

contractor compared in relation to the other five who were invited to tender. 'Our'
contractor is identified as contractor Cr1 in the following text. Table 8 exhibits P2
and tender sums submitted by all contractors competing for the award.
It is worth noting at uiis point that under current selection practice, contractor
3, i.e. Cr 3 , would most probably be awarded the contract having submitted the
lowest bid. However, having used this alternative selection technique to evaluate
the firm's performance potential, he/she may not be the best contractor for the job
- as evidenced by his/her P2 score (0.380). 'Our' contractor Cr1 submitted the
second highest bid and would therefore stand little chance of the award under the
traditional tendering approach.
However, if we apply the alternative selection technique to all contractors, first
by calculating bid scores (Table 9) and subsequently determining respective P3
scores (Table 10), then our contractor now ranks highest overall and would,
under this method, be most eligible for the contract. Indeed, the lowest bidder

Table 8 P2 scores/tender sums for all contractors tendering

Contractor P2 score P2 rank Bid £K Bid rank

Cr1 0.765 1 486 5

Cr2 0.612 5 478 2
Cr3 0.380 6 475 1
Cr4 0.720 2 487 6
Cr5 0.637 4 479 3
Cr6 0.704 3 480 4

Table 9 Contractors' bid scores

Contractor Lowest bid/this bid Bid score

Cr1 475/486 0.977

Cr2 475/478 0.993
Cr3 475/475 1.000
Cr4 475/487 0.975
Cr5 475/479 0.991
Cr6 475/480 0.989

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 The selection of construction contractors 45

Table 10 P3 calculation for all tenderers

Contractor P2 element Bid score element P3 score Overall rank

Cr1 (0.765 x 0.4) + (0.977 x 0.6) = 0.904 1

Cr2 (0.612 x 0.4) + (0.993 x 0.6) = 0.840 5
Cr3 (0.380 x 0.4) + (1.000 x 0.6) = 0.752 6
Cr4 (0.720 x 0.4) + (0.975 x 0.6) = 0.873 3
Cr5 (0.637 x 0.4) + (0.991 x 0.6) = 0.849 4
Cr6 (0.704 x 0.4) + (0.989 x 0.6) = 0.875 2

(Cr 3 ) now achieves the lowest rank due to a very poor P2 score. Therefore, the
model has identified the best all-round contractor for the project along with the
contractor who (albeit submitted the lowest bid) in this instance exhibits a greater
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

probability of generating problems for the client en route to completion of the

contract.
When validation of the model in its present form is nearing completion, further
analysis of die data (prediction with actual) will permit sensitivity analysis of the
P3 formula coefficients. This will confirm whether benefit can accrue to those
clients who wish to place greater/lesser emphasis on the cost component of tenders
(i.e. adjust the 60/40 P2/cost ratio).
Notwithstanding its being simplistic in nature, the example has highlighted the
salient aspects of the new technique. Selection should consider a cocktail of
discriminating criteria to evaluate contractor performance potential. Hence, the
best prospect, not necessarily the lowest tenderer, will be identified.

SUMMARY

This research is based on the premise that selection should concentrate on

determining contractor potential for achieving client satisfaction in terms of time,
cost and quality standards. This approach contrasts with the majority of current
selection regimes which tend to prequalify (using bespoke methods - see Holt et
ah (1993a), then discriminate predominantly on cost. With this in mind, it maybe
that the new technique will yield greatest potential for the public sector who
indeed are showing much interest in the research. This is because public sector
clients presently find it difficult to accept other than the lowest bid (from
whomsoever received) on the grounds of public accountability. However, the
objective values furnished by the model will identify those contractors with least
performance potential and would therefore allow public sector clients to con-
fidently reject such low, but higher risk, tenders. As Birrell (1988) pointed out, 'If
you deal with the lowest bidder it's as well to add something for the risk you run, in
which case you will have enough to pay for something better'.
To date, the technique has been moulded on the 'traditional' lump sum con-
tract but, as demonstrated, a broader evaluation embracing financial ramifications
of tenders is possible. There is scope for future inclusion of other discriminating
criteria such as suitability of design, predicted costs in use, contract period, etc., to

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 46 Holt, G.D. et al.

mould the new technique to alternative procurement forms, e.g. design and build
or, by including alternative project specific criteria such as safety record, internal
security regime, etc., to adapt to specific project types. Indeed, survey of selection
practitioners/client groups for any particular country would identify essential
national selection criteria (e.g. bonding capacity - USA) for wider application. In
contrast, a more compact version may be of interest to contractors themselves for
selecting sub-contractors, or to management contractors for selecting trade-
packages.
Finally, the significance of the research is confirmed when one observes the
interim findings of the government instigated review of me construction industry,
being conducted by Sir Michael Latham. 'Choice of consultant or contractor
should be made on a value for money basis, with proper weighting of criteria for
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

skill, experience and previous performance, rather than automatically accepting

the lowest tender in all cases' (Latham 1993).

ACKNOWLEDGEMENTS

The authors wish to thank the Leverhulme Trust for funding this research, and
numerous selection practitioners and client groups for their collaboration and
invaluable ongoing input. The anonymous reviewers of the paper are also
acknowledged for their constructive comments.

References
Baker, M. & Orsaah, S. (1985) How do the customers choose a contractor? Building Magazine, May,
30-31.
Birrell, G.S. (1988) Bid Appraisal Incorporating Past Performances By Contractors. American Association
of Cost Engineers. Trans. D 1.1-D 1.6.
C.I.O.B. (1989) Quality Assurance in Building, 2nd edn, Chartered Institute of Building, London.
Diekmann, J.E. (1979) Selection of cost plus contractors using normative decision methodologies. PhD thesis,
University of Washington.
Griffith, A. (1990) Quality Assurance in Building. MacMillan, London.
Griffith, A. (1992) Small Building Works Management. Macmillan, London.
Hardy, S.C., Norman, A. & Perry, J.G. (1981) Evaluation of bids for construction contracts using discounted
cashflow techniques. Proceedings of the Institute of Civil Engineers, 70 (1), 91-111.
Hawwash, K. (1991) Bid Evaluation By Points: Review. Management of Contracts & Projects - Project
Management Group, Institute of Science and Technology, University of Manchester.
Holt, G.D., Olomolaiye, P.O. & Harris, F.C. (1993a) A conceptual alternative to current tendering
practice. Building Research & Information, 21 (3), 167-72.
Holt, G.D., Olomolaiye, P.O. & Harris, F.C. (1993b). Tendering practice - exploring alternatives.
Faculty of Building Journal, Autumn edn, 28-30.
Holt, G.D., Olomolaiye, P.O. & Harris, F.C. (1993c) Factors influencing UK construction clients'
choice of contractor. Building and Environment, 241-248.
Holt, G.D., Olomolaiye, P.O. & Harris, F.C. (1993d) Evaluating prequalification criteria in contractor
selection. Building and Environment, Pergamon Press, Oxford.
Holt, G.D., Olomolaiye, P.O. & Harris, F.C. (1994) Incorporating project specific criteria and client
utility into the evaluation of construction tenderers. Building Research & Information.
Hwang C. & Yoon K. (1981) Multi-Attribute Decision Making- a state of the art survey. Verlag, Berlin.
ISO (9000) Standard 9000, Quality Systems. International Standards Organization, Switzerland.

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 The selection of construction contractors 47

Latham, M. (1993) Trust and Money. Interim report of the joint Government/industry review of
procurement and contractual arrangements in the UK construction industry. Department of the
Environment.
Mascoll, S.G. (1984) Organisational structures and contracts for construction projects and selection. PhD
thesis, University of Leeds.
Merna, A. & Smith, N . (1990) Bid Evaluation for Public Sector ConstructionContracts. Proceedings of the
Institution of Civil Engineers, 88 (1), 91-105.
Mohsini, R. & Davidson, C.H. (1986) Procurement Organisation Design and Building Team Performance-
A Study of Inter-firm Conflict. CIB Proc., 8, Washington DC.
Moselhi, O. & Martinelli, A. (1990) Analysis of Bids Using Multi Attribute Utility Tlieory. Proceedings of
the International Symposium on Building Economics & Construction Management. Sydney,
Australia, 335-345.
Moore, P.G. & Thomas H. (1979) The Anatomy of Decisions. Penguin, Middlesex.
Mott, G. (1992) Investment Appraisal. Pitman, London.
Odusote, O.O. (1990) An examination of the importance of resource considerations when contractors make
project selection decisions. MSc Dissertation, University of Bath.
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

Olomolaiye, P.O., Wahab, K. & Price, A. (1987) Problems facing craftsmen's productivity in Nigeria.
Building & Environment, 22 (4), 317-23.
Russell, J. (1992). Decision models for analysis and evaluation of construction contractors. Con-
struction Management & Economics, 10, 185-202.
Skitmore, R.M. & Marsden, D.E. (1988) Which procurement system? Towards a universal
procurement selection technique. Construction Management & Economics, 6, 71-89.
Smit, J.J. (1978) Tender evaluation, the whole concept. The Civil Engineering Contractor (South Africa),
13 (10), 39-19.
Spons (1992) Spoils' Architects and Builders Price Book. E & FN Spon, London.
Sullivan, A. & Harris, F.C. (1986) Delays on large construction projects. International Journal of
Operational and Production Management 6 ( 1 ) .
Watkinson, M. (1992) Procurement alternatives. Faculty of Building Journal Autumn, Winter edns.

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 48 Holt, G.D. et al.

APPENDIX A
PREQUALIFICATION (P1) SUMMARY ANALYSIS SHEET
Factor: Organizational structure

Score Importance Rationalized

Variables achieved x weight = score

V1 Size of contractor company 1.0 0.501 0.501

V2 Age of contractor company 1.0 0.435 0.435
V3 Company image 1.0 0.408 0.408
V4 Quality control policy 1.0 0.529 0.529
V5 Health & safety policy 0.7 0.583 0.408
V6 Litigation tendency 0.7 0.545 0.381

Totals 3.001 a 2.662 b

Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

b
divided by a = Organizational structure. Factor score = 0.887c

Factor: Financial stability

Score Importance Rationalized

Variables achieved x weight = score

V7 Ratio analysis of financial accounts 0.67 0.631 0.422

V8 Bank reference 0.80 0.669 0.535
V9 Creditor reference(s) 0.85 0.634 0.538
V10 Turnover history 0.75 0.667 0.500

Totals 2.601 d 1.995e

e f
divided by d = Financial stability. Factor score = 0.767

Factor: Management resource

Score Importance Rationalized

Variables achieved x weight = score

V11 Qualification of company owners 0.78 0.676 0.527

V 12 Qualification of key persons 0.65 0.648 0.421
V 13 Key persons years with company 0.80 0.695 0.556
V 14 Extent of company training regime 1.00 0.814 0.814

Totals 2.833g 2.318 h

h
divided by g = Management resource. Factor score = 0.818i

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

 The selection of construction contractors 49

Factor: Past experience

Score Importance Rationalized

Variables achieved x weight =: score

V 15 Type of projects completed 0.75 0.735 0.551

V 16 Size of projects completed 1.00 0.851 0.851
V 17 Area of catchment 0.60 0.748 0.448

Totals 2.334 j 1.850k

k m
divided byj = Past experience. Factor score == 0.792

Factor: Past performance

Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

Score Importance Rationalized

Variables achieved x weight = score

V 18 Failure to have completed a contract 1.0 0.679 0.679

V 19 Contract overruns-time 0.75 0.541 0.405
V 20 Contract overruns-cost 0.80 0.576 0.460
V21 Actual quality achieved 0.85 0.667 0.566

Totals 2.463 n 2.110°

° divided by n = Part performance. Factor score = 0.856°

Prequalification (P1) final calculation

Factor Σ weighting indices per factor Σ rationalized scores per factor

Organizational structure 3.001 a 2.662 b

Financial stability 2.601d 1.995 e
Management resource 2.833 g 2.318 h
Past experience 2.334 i 1.850 k
Past performance 2.463 n 2.110°

Totals 13.232 q (Z1 Max) 10.935 r (Z1 score)

Z1 scorer divided by Z1 Maxq = P1 score == 0.826 which may be expressed as 83% performance potential.

Engineering, Construction and Architectural Management 1994 1 | 1, 29-50

50 Holt, G.D. et al.

APPENDIX B
TENDER EVALUATION (P2) SUMMARY ANALYSIS SHEET
Factor: Project specific

Importance Utility Score Rationalized

Variable weight x weight x achieved = score

V22 Geographic experience (to project) 0.409 1.0 1.00 0.409

V23 Experience (project construction) 0.564 1.0 0.80 0.451
V 24 Plant resource available for project 0.486 0.9 0.75 0.328
V25 Key persons available for project 0.547 1.0 0.83 0.454
V26 Qualification of t h e s e key persons 0.673 1.0 0.80 0.538

Totals 2.679 s 2.180 t

Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

t
divided by s = Project specific. Factor score == 0.813u

Factor: Other specific

Importance Utility Score Rationalized

Variable weight x weight x achieved = score

V 2 7 Workload (project duration) 0.862 1.0 1.00 0.862

V 2S Prior relationship with client 0.651 0.7 0.60 0.273
V 2 9 Home office location (regard project) 0.642 0.6 1.00 0.385

Totals 2.155 v 1.520 w

w
divided by v = other specific. Factor score = 0.705x

T e n d e r e v a l u a t i o n ( P s ) final c a l c u l a t i o n

Factor Σ weighting indices per factor Σ rationalized scores per factor

Project specific 2.679 s 2.180 t

O t h e r specific 2.155 r 1.520 w

Totals 4.834 y (Z2 Max) 3.700 z <Z2 score)

Z2 score z divided by Z2Max y == P2 score = 0.765 which may be expressed as 77 performance potential.

Engineering, Construction and Architectural Management 1994 1 | 1, 2 9 - 5 0

 This article has been cited by:

1. Olugbenga Jide Olaniran. 2015. The effects of cost-based contractor selection on construction
project performance. Journal of Financial Management of Property and Construction 20:3, 235-251.
[Abstract] [Full Text] [PDF]
2. Amirhossein Movahedian Attar, Mostafa Khanzadi, Shahin Dabirian, Elmira Kalhor. 2013.
Forecasting contractor's deviation from the client objectives in prequalification model using support
vector regression. International Journal of Project Management 31:6, 924-936. [Crossref]
3. Amirhosein Jafari. 2013. A contractor pre-qualification model based on the quality function
deployment method. Construction Management and Economics 31:7, 746-760. [Crossref]
4. Pertti Lahdenperä. 2013. Determining ‘the most economically advantageous tender’ based on
capability and fee-percentage criteria. Journal of Public Procurement 13:4, 409-446. [Abstract] [PDF]
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

5. Firuzan Yasamis-Speroni, Dong-Eun Lee, David Arditi. 2012. Evaluating the Quality Performance
of Pavement Contractors. Journal of Construction Engineering and Management 138:10, 1114-1124.
[Crossref]
6. Lijuan Tao, Mohan Kumaraswamy. 2012. Unveiling relationships between contractor inputs and
performance outputs. Construction Innovation 12:1, 86-98. [Abstract] [Full Text] [PDF]
7. Gary Holt. 2010. Contractor selection innovation: examination of two decades' published research.
Construction Innovation 10:3, 304-328. [Abstract] [Full Text] [PDF]
8. Ka-Chi Lam, Mike Chun-Kit Lam, Dan Wang. 2010. Efficacy of Using Support Vector Machine
in a Contractor Prequalification Decision Model. Journal of Computing in Civil Engineering 24:3,
273-280. [Crossref]
9. Ka Chi Lam, Ekambaram Palaneeswaran, Chen-yun Yu. 2009. A support vector machine model for
contractor prequalification. Automation in Construction 18:3, 321-329. [Crossref]
10. T.H. Nguyen, T. Shehab, Z. Gao. 2008. Selecting an architecture‐engineering team by using fuzzy
set theory. Engineering, Construction and Architectural Management 15:3, 282-298. [Abstract] [Full
Text] [PDF]
11. Luciana Hazin Alencar, Adiel Teixeira de Almeida, Caroline Maria de Miranda Mota. 2007.
Sistemática proposta para seleção de fornecedores em gestão de projetos. Gestão & Produção 14:3,
477-487. [Crossref]
12. Florence Yean Yng Ling. 2004. Consultancy Fees: Dichotomy between A/E’s Need to Maximize
Profit and Employers’ Need to Minimize Cost. Journal of Professional Issues in Engineering Education
and Practice 130:2, 120-123. [Crossref]
13. Ekambaram Palaneeswaran, Mohan Kumaraswamy, Thomas Ng. 2003. Targeting optimum value in
public sector projects through “best value”‐focused contractor selection. Engineering, Construction
and Architectural Management 10:6, 418-431. [Abstract] [Full Text] [PDF]
14. . References and Bibliography 409-447. [Crossref]
15. Ekambaram Palaneeswaran, Mohan Kumaraswamy, Motiar Rahman, Thomas Ng. 2003. Curing
congenital construction industry disorders through relationally integrated supply chains. Building
and Environment 38:4, 571-582. [Crossref]
16. Jakrapong Pongpeng, John Liston. 2003. TenSeM: a multicriteria and multidecision-makers' model
in tender evaluation. Construction Management and Economics 21:1, 21-30. [Crossref]
17. Stephen C. Ngai, Derek S. Drew, H. P. Lo, Martin Skitmore. 2002. A theoretical framework for
determining the minimum number of bidders in construction bidding competitions. Construction
Management and Economics 20:6, 473-482. [Crossref]
 18. Ekambaram Palaneeswaran, Mohan M. Kumaraswamy, Xue Qing Zhang. 2001. Reforging
construction supply chains:. European Journal of Purchasing & Supply Management 7:3, 165-178.
[Crossref]
19. S.Thomas Ng. 2001. EQUAL: a case-based contractor prequalifier. Automation in Construction 10:4,
443-457. [Crossref]
20. Ekambaram Palaneeswaran, Mohan Kumaraswamy. 2001. Recent advances and proposed
improvements in contractor prequalification methodologies. Building and Environment 36:1, 73-87.
[Crossref]
21. Prasanta Kumar Dey. 2000. Managing projects in fast track – A case of public sector organization
in India. International Journal of Public Sector Management 13:7, 588-609. [Abstract] [Full Text]
[PDF]
22. Ekambaram Palaneeswaran, Mohan M. Kumaraswamy. 2000. Contractor Selection for Design/Build
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)

Projects. Journal of Construction Engineering and Management 126:5, 331-339. [Crossref]

23. E. Palaneeswaran, M.M. Kumaraswamy. 2000. Benchmarking contractor selection practices
in public-sector construction-a proposed model. Engineering Construction and Architectural
Management 7:3, 285-299. [Crossref]
24. K.C. LAM, S. THOMAS NG, TIESONG HU, MARTIN SKITMORE, S.O. CHEUNG.
2000. Decision support system for contractor pre‐qualification—artificial neural network model.
Engineering, Construction and Architectural Management 7:3, 251-266. [Abstract] [PDF]
25. E. PALANEESWARAN, M.M. KUMARASWAMY. 2000. Benchmarking contractor selection
practices in public‐sector construction—a proposed model. Engineering, Construction and
Architectural Management 7:3, 285-299. [Abstract] [PDF]
26. Paul Jennings, Gary D. Holt. 1998. Prequalification and multi-criteria selection: a measure of
contractors' opinions. Construction Management and Economics 16:6, 651-660. [Crossref]
27. Gary D Holt. 1998. Which contractor selection methodology?. International Journal of Project
Management 16:3, 153-164. [Crossref]
28. Ezekiel A. Chinyio, Paul O. Olomolaiye, Simon T. Kometa, Frank C. Harris. 1998. A needs-based
methodology for classifying construction clients and selecting contractors. Construction Management
and Economics 16:1, 91-98. [Crossref]
29. Gary D. Holt. 1997. Classifying construction contractors. Building Research & Information 25:6,
374-382. [Crossref]
30. D. G. Proverbs, G. D. Holt, P. O. Olomolaiye. 1997. Factors influencing the choice of concrete
supply methods. Building Research & Information 25:3, 176-184. [Crossref]
31. Gary D. Holt. 1996. Applying cluster analysis to construction contractor classification. Building and
Environment 31:6, 557-568. [Crossref]
32. Mohan M. Kumaraswamy. 1996. Contractor evaluation and selection: a Hong Kong perspective.
Building and Environment 31:3, 273-282. [Crossref]
33. GARY D. HOLT, PAUL O. OLOMOLAIYE, FRANK C. HARRIS. 1996. Tendering procedures,
contractual arrangements and Latham: the contractors' view. Engineering, Construction and
Architectural Management 3:1/2, 97-115. [Abstract] [PDF]
34. Gary D. Holt, Paul O. Olomolaiye, Frank C. Harris. 1995. A review of contractor selection practice
in the U.K. construction industry. Building and Environment 30:4, 553-561. [Crossref]
35. Gary D. Holt, Paul O. Olomolaiye, Frank C. Harris. 1995. Application of an alternative contractor
selection model. Building Research & Information 23:5, 255-264. [Crossref]
 36. Derek H. T. Walker. 1995. An investigation into construction time performance. Construction
Management and Economics 13:3, 263-274. [Crossref]
Downloaded by International Islamic University Malaysia At 01:23 22 February 2018 (PT)