Journal of Business Economics and Management 2008

9(4): 245–255

MULTI-OBJECTIVE CONTRACTOR’S RANKING

BY APPLYING THE MOORA METHOD

Willem Karel M. Brauers1, Edmundas Kazimieras Zavadskas2, Zenonas Turskis3, Tatjana Vilutienė4
1University
of Antwerpen, Birontlaan, 97, B2600 Berchem Antwerpen, Belgium
E-mail: willem.brauers@ua.ac.be
2,3,4Vilnius Gediminas Technical University, Saulėtekio al. 11, LT-10223 Vilnius, Lithuania

E-mails: 2edmundas.zavadskas@adm.vgtu.lt; 3zenonas.turskis@st.vgtu.lt; 4 tatjana.vilutiene@st.vgtu.lt

Received 18 November 2007; accepted 15 July 2008

Abstract. Construction, taking off, maintenance and facilities management of a building is a typical example of consumer
sovereignty: the new owner likes to have a reasonable price to pay, to have confidence in the contractor, to know about
the duration of the works, the service after completion and the quality of the work. On the other side the contractor has his
objectives too, like the satisfaction of the client, diminishing of external costs and annoyances and the management cost
per employee as low as possible. In other words it concerns a problem of multi-objectives. Therefore a final ranking will
show the best performing contractor from the point of view of the clients but also from the point of view of the contractors
themselves. The MOORA method based on ratio analysis and dimensionless measurement will accomplish the job of rank-
ing the contractors in a non-subjective way. As an application the largest maintenance contractors of dwellings in Vilnius,
capital of Lithuania, were approached.
Keywords: contractor, decision making, multi-objectives, alternatives, sampling, ratio analysis, dimensionless measures,
ranking and optimization, the MOORA method.

1. Introduction higher is the gain and the effectiveness of the activity.

The effectiveness of decision making will depend on
The effectiveness of construction and maintenance de- the goals of all interested groups, participating in the
pends on many micro- and macro-environmental fac- process and with regard to the impact of the micro-
tors. Therefore, planning and successful implementa- and macro-environmental factors. Contractors cannot
tion of construction and maintenance activities require correct or change aforementioned factors, but they can
the evaluation of the capabilities of the participants realize their impact and evaluate it during the imple-
of those processes and the influence of the environ- mentation of different projects, herewith successfully
ment on their effectiveness. The participants of the organizing their current and future activities. The term
process can perform their functions efficiently only ‘effectiveness’ can be interpreted differently; therefore
taking into consideration the changing environment, one has to evaluate all the needs of the participants
pursuing the best coordination of actions, raising the of the process. Modelling and multi-objective analysis
quality of services and meeting the needs of dwell- allow us to find a way to meet the goals of the par-
ing owners. Effectiveness is hereby perceived as the ticipants of different process and to choose an optimal
process of providing building maintenance services, solution as well as the ways to implement it.
which results in ultimate implementation of the goals Construction projects are one-off endeavours with
of the interested groups participating in the process. many unique features such as long period, complicat-
The effectiveness of any process is assessed in terms ed processes, changing environment. As construction
of objectives, which vary depending on the problem projects become more complex, the need for evaluating
concerned and the particular goals of the interested contractor performance becomes more crucial. Organi-
groups. The utmost effectiveness is often associated zational and technological complexity of construction
with the maximum gain from a specific activity. The projects generates enormous risks. The selection of a
more numerous and significant aims are achieved, the qualified contractor gives confidence to the stakehold-

ISSN 1611-1699 print. www.jbem.vgtu.lt, www.jbem.lt DOI: 10.3846/1611-1699.2008.9.245-255

 W. K. M. Brauers et al. Multi-objective contractor’s ranking by applying the MOORA method

er that the selected contractor can achieve the project kas (1996) selected 25 attributes of contractor selection
goals. However, the importance of contractor selection and applied COPRAS method to contractor selection.
is mostly underestimated and neglected in construc- Hatush and Skitmore (1998) have initiated the use of
tion (Ng and Wan 2005). It is hard to analyze many systematic multi-attribute decision analysis techniques
tradeoffs involved in decision making, especially in for contractor selection and bid evaluation based on ad-
times with so many uncertainties presented by environ- ditive multi-attribute utility function model. Banaitiene
mental considerations. Insufficient time for execution, and Banaitis (2006) performed an analysis of criteria
complicated procedures or poor information channels for contractors’ evaluation. Dikmen et al. (2007) after
may be the reasons of problems in the selection of conducting a thorough research, selected 44 candidate
contractors (Shiau et al. 2002). Contractor evaluation factors affecting the bid mark-up decisions as fac-
has been recognized as a particularly complex task due tors having potential impact on bid mark-up size for a
to its ambiguity and difficult formalisation (Shiau et al. project. The factors are divided into 4 groups, namely:
2002; Tseng and Lin 2002). There have been no gener- general features about company and project, risk fac-
alized sets of rules for the evaluation process. tors, opportunity factors and competition factors.
An extensive literature review by the researchers re-
Contractor selection deals with risk and risk manage-
vealed that the most acceptable contractor’s pre-qual-
ment. Companies working in construction market face
ification attributes are financial stability, management
more demanding business environment that is usu-
and technical ability, contractor’s experience, contrac-
ally non-stable. For instance, Lithuanian construction
tor’s performance, resources, quality management and
market is influenced by political, social, economic,
health and safety concerns. Therefore, the contractor’s
technological and other transformations currently un-
attributes corresponding to these attributes should be
dergoing in Eastern and Central Europe (Ginevičius evaluated.
and Korsakienė 2005). Implementing the construction
projects, it must be first of all considered what risks Many researches (Zavadskas and Kaklauskas 1996,
should be countered with measures and how costly 2007; Ginevičius and Vaitkūnaitė 2006; Zavadskas
these measures are (Schieg 2006). Zou et al. (2007) and Vilutiene 2006) have pointed out that in construc-
state that the risks in construction projects can be tion it is essential to be able to take into account the
classified as follows: cost overrun, time delay, qual- impacts of cultural, social, moral, legislative, demo-
ity, safety, environmental sustainability and funding, graphic, economic, environmental, governmental and
contractors’ poor management ability, contractors’ technological change, as well as changes in the busi-
difficulty in reimbursement, poor competency of la- ness world on international, national, regional and lo-
bourers, not buying insurance for major equipments cal real estate markets. Evaluation of contractors based
and employees, inadequate safety measures or unsafe on multi-attributes is becoming more popular and is, in
operations, lack of readily available utilities on site, essence, largely dependent on the uncertainty inherent
prosecution due to unlawful disposal of construction in the nature of construction projects and subjective
waste and serious air and water pollution due to con- judgment of decision makers.
struction activities, suppliers’ incompetency to deliver
Multi-attribute decision making is defined by processes
materials on time.
that involve designing the best alternative with the most
Many construction contracts are awarded to the low- attractive attributes, and that involves the selection of
est bidder. An offered bid price is undoubtedly an the optimal alternative, handled via preference models
important factor in choosing a contractor, but there (Sage 1977; Hwang and Yoon 1981). Multi-attribute
are many other important ones playing a vital role in decision making can be classified as follows:
project implementation that have to be incorporated a) Multi-attribute decision making (MADM) for the
in the contractor’s evaluation process. The contractor sorting or the ranking of alternatives according to
pre-qualification process involves the establishment of several attributes and
a standard for measuring and assessing the capabilities b) Multi-objective decision making (MODM), for
of potential contractors (Ng et al. 1999). The infor- driving a vector optimization-based design process
mation used for the assessment of parameters for pre- to a solution (Colson and Bruyn 1989).
qualification should fall into several groups (Hatush Multi-attribute methods can be classified by the type
and Skitmore 1997). Jaselskis and Russel (1992) have of initial information (deterministic, stochastic, fuzzy
identified commonly used attributes for pre-qualifica- set theory methods) or by the number of decision-
tion and bid evaluation and have proposed methodolo- makers (one or group). Scientists classify determin-
gies for contractor selection. Zavadskas and Kaklaus- istic MADM methods differently. Lin and Wu (2007)

246
 Journal of Business Economics and Management, 2008, 9(4): 245–255

presented classification of the methodology which can Many objectives have to be considered for selecting a
be used for qualitative and quantitative methods aimed qualified contractor on a competitive basis. Once be-
at technology management. ing convinced about multi-objectives, Multi-Objective
Decision Making research is divided into the following
To select contractors Topcu (2004) used Analytical
groups (Thousands and thousands of publications on
Hierarchy Process methods and proposed a multi-at-
each of these methods are edited. Therefore, only the
tribute decision model based on time, price and quality
attributes evaluation for eligible contractor selection. first pioneering publication of each method is men-
Skibniewski and Chan (1992) applied this method tioned):
to the selection of rational construction technology. 1) An Ordinal Preference. The method of correlation
Mitkus and Trinkuniene (2006) analyzed three mod- of ranks, consisting of totalizing ranks, is the first
els of multi-attribute systems of construction contrac- method to be considered. Rank correlation was in-
tion agreements, whereas in 2007 they (Mitkus and troduced first by psychologists such as Spearman
Trinkuniene 2007) suggested to use analytic hierarchi- (1904, 1906 and 1910) and later taken over by the
cal model (AHP) for structural evaluation of construc- statistician Kendall in 1948. The Lexicographic
tion contracts. Method belongs also to this group (Holmes 1971).
2) Additive Weighting with the sum of weights equal
Keršulienė (2007) proposed analysis model for con- to one (Churchman et al. 1954, 1957).
struction process parties during dispute settlement. She 3) Methods based on quantitative measurement but
stated that with the use of optimism and asymmetric using a few criteria to compare the alternatives
information models it is possible to determine the most pairwise. Schärlig (1985) calls them partial aggre-
economically advantageous behavioural pattern for
gation methods. This group consists of preference
both parties.
comparison methods like the group around Elec-
Selection of a contractor is an important issue in con- tre (Roy et al. 1966) and the Indifference Method
struction field (Zagorskas and Turskis 2006; Turskis et (Brauers 1977 b).
al. 2006; Zavadskas and Vilutiene 2006) for the suc- 4) Methods based on initial qualitative assessment, the
cess or failure of a project is usually influenced by the results of which take a quantitative form later. This
quality of contractor. The researches listed above had group consists mainly of analytic hierarchy meth-
significantly improved the contractor selection proc- ods (Saaty 1988), as well as the methods based on
ess in the construction industry. However, some of the fuzzy sets (Zimmermann 1978).
proposed methods and approaches could be complex 5) Methods based a) on a Reference Point (Minkowsky
and difficult to apply in practice. The construction in- 1896; Karlin and Studden 1966) in the Reference
dustry needs simple but effective methods in contractor Point Method such as used in TOPSIS (Hwang and
selection process due to the limited time intervals of Yoon 1981), COPRAS (Zavadskas and Kaklauskas
the bidding periods. For these and many other reasons, 1996) or b) on a Goal like in Goal Programming
selection of a construction contractor requires the con- (Lee 1972).
tractor selection model that should be able to meet the 6) Bridgman discussed dimensionless measures al-
critical characteristics of the pre-qualification: ready in 1922. Multi-objectives researchers discov-
• a multi-attribute problem. ered dimensionless measures as a welcome gift in
• risks inherited from different decision maker’s opin- order to whither away with the cumbersome nor-
ions. malization necessary for the different units of the
• noisy and uncertain date given by different contrac- different objectives.
tors.
• subjective judgment made by decision makers. First, the Multiplicative Form is based on dimension-
• non-linear relationships between contractor’s at- less measures. Miller and Starr (1969) demonstrated
tributes and their corresponding pre-qualification ratios that do not change if single-dollar units are trans-
decisions. formed to million-dollar units even when attributes are
• to deal with qualitative as well as quantitative data. raised to powers. The Multiplicative Form for Multiple
Objectives was further elaborated by Schärlig (1985)
It should be noted that the stakeholders must adjust the
and Brauers (1997a, 1999, 2002, 2004).
attributes depending on the demand of each project.
The critical point is that the selected attributes should Second, ratio analysis responds to dimensionless meas-
have a direct effect on performance. In addition, the ures analysis, which is applied in MOORA (Multi-Ob-
selected evaluation attributes should be also based on jective Optimization by Ratio Analysis). MOORA was
the measurement culture of the stakeholder. introduced for the first time by Brauers and Zavadskas

247
 W. K. M. Brauers et al. Multi-objective contractor’s ranking by applying the MOORA method

in “Control and Cybernetics” in 2006, after introducing 2.1. The Ratio System as a part of MOORA
its first part, the ratio system, in Brauers (2004).
We go for a ratio system in which each response of an
Given these different approaches preference is given to alternative on an objective is compared to a denomina-
an approach satisfying the following conditions: tor which is a representative for all alternatives con-
• Cardinal and not ordinal, given the weak points of an cerning that objective (Brauers and Zavadskas 2006).
ordinal approach as demonstrated by Arrow (1974) Authors proved that for this denominator the best
and Brauers (2007); choice is the square root of the sum of squares of each
• All objectives should be considered and respected as alternative per objective:
much as possible;
• All stakeholders are involved and not only the de- (2)
cision maker or a group of decision makers. All
stakeholders mean everybody interested in a certain
issue;
where: xij – response of alternative j on objective i; j =
• All interrelations between objectives and alternatives
1, 2, ..., m; m – the number of alternatives; i = 1, 2,
are looked upon at the same time instead of pairwise
..., n; n – the number of objectives; – a dimension-
considerations;
less number representing the normalized response of
• Only discrete cases facing a set of a limited number
alternative j on objective i.
of alternatives are considered, whereas continuous
cases concern alternatives generated out of a set of Dimensionless Numbers, having no specific unit of
continuous and numerous alternatives; measurement, are obtained for instance by deduction,
• As non-subjective as possible: no normalization by multiplication or division. The normalized responses of
subjective weights but by non-subjective dimension- the alternatives on the objectives belong to the inter-
less measures. val [0; 1]. However, sometimes the interval could be
For all these reasons we selected MOORA. [–1; 1]. Indeed, for instance, in the case of productivity
growth some sectors, regions or countries may show
a decrease instead of increase in productivity, i.e. a
2. The MOORA method negative dimensionless number.
The method starts with a matrix of responses of differ- For example, instead of a normal increase in productiv-
ent alternatives on different objectives: ity growth a decrease remains possible. At this moment,
the interval becomes [–1, 1]. Let us consider an ex-
ample of productivity which has to increase (positive).
Consequently, we look after productivity maximiza-
tion, e.g. in European and American countries. What
(1)
if the opposite does occur? For instance, let us analyse
the original transition from the USSR to Russia.
Contrary to the other European countries, its productiv-
where: xij – the response of alternative j on objective or ity decreased. It means that in formula (2) the numera-
attribute i; i = 1, 2, ..., n – is the number of the objec- tor for Russia would be negative with the whole ratio
tives or the attributes; j = 1, 2, ..., m – is the number of becoming negative. Consequently, we get the interval
the alternatives. [–1; +1] instead of [0; 1]. For optimization, these re-
sponses are added in case of maximization and sub-
In order to define objectives, we have to focus on the tracted in case of minimization:
notion of Attribute. Keeney and Raiffa (1993) present
the example of the objective ‘reduce sulfur dioxide
(3)
emissions’ to be measured by the attribute ‘tons of sul-
fur dioxide emitted per year’. An objective and a cor- where: i = 1, 2, ... , g – the objectives to be maximized;
respondent attribute always go together. Consequently, i = g + 1, g + 2, ... , n – the objectives to be minimized;
when the text mentions objective, the correspondent – the normalized assessment of alternative j with
attribute is also meant. respect to all objectives.
The MOORA Method consists of two components: (a) An ordinal ranking of the shows the final preference.
the ratio system and (b) the reference point approach. Indeed, cardinal scales can be compared in an ordi-

248
 Journal of Business Economics and Management, 2008, 9(4): 245–255

nal ranking, according to Arrow (1974): ‘Obviously, assessment of alternative j with respect to all objec-
a cardinal utility implies an ordinal preference but not tives with significance coefficients; – a dimension-
vice versa’. less number representing the normalized response of
alternative j on objective i.
2.2. The Reference Point Approach
as a part of MOORA The Attribution of Sub-Objectives represents still an-
other solution. Let us consider an example of purchas-
Reference Point Theory is based on the ratios found ing fighter planes (Brauers 2002). From an economic
in formula (2) whereby a Maximal Objective Refer- point of view, apart from military effectiveness the ob-
ence Point is also deduced. The Maximal Objective jectives concerning the fighter planes are threefold –
Reference Point approach is called realistic and non- price, employment and balance of payments. In order
subjective when the co-ordinates (ri) selected for the to give more importance to military defence, effective-
reference point are realized in one of the candidate ness is broken down in, for instance, the maximum
alternatives. For example, we have three alternatives. speed, the power of the engines and the maximum
The alternatives are described as follows: A (10;100), range of the plane. Anyway, the Attribution Method
B (100;20) and C (50;50). In this case the maximal is more refined than a significance coefficient method.
objective reference point Rm results in (100;100). The The attribution method succeeds in characterizing an
Maximal Objective Vector is self-evident if the alterna- objective better.
tives are well defined as for the projects in the area of
Project Analysis and Planning.
3. Application of the proposed method
Having given the dimensionless number representing for evaluating contractor’s alternatives
the normalized response of alternative j on objective i,
i.e. in formula (2) , we come to: Objectives are determined on the basis of the oppor-
tunities for the contractors and of the wishes of the
(ri – ), (4)
customers.
where: j = 1, 2, ..., m; m – the number of alternatives;
i = 1, 2,… n; n – the number of objectives; ri – the i th The largest maintenance contractors of dwellings in
co-ordinate of the reference point; – the normalized Vilnius, the capital of Lithuania, were approached, of
attribute i of alternative j. which 15 agreed to fix and estimate their main ob-
jectives, namely 9 objectives as given in Table 1 (Dr
This matrix is subject to the Min-Max Metric of Tatjana Vilutienė took care of the field work). The full
Tchebycheff (Karlin and Studden 1966; Brauers and names of the contractors are not provided for the sake
Zavadskas 2006): of confidentiality.
(5) The nature of the construction industry involves that
the total number of the minima is mostly larger than
Bauers and Zavadskas (2006) showed Min-Max metric
the total number of the maxima.
is the best choice between all the possible metrics of
reference point theory. Contractor’s rating is performed according to the at-
tributes (Table 2).
2.3. The Importance given to an Objective
From information of the Dwelling Owners Association,
The normalized responses of the alternatives on the a panel of 30 owners of dwellings chosen at random
objectives belong to the interval [0; 1] (see formula 2). agreed with these 9 objectives, but they increased the ob-
Nevertheless, it may turn out to be necessary to stress jectives with 11 other ones (These additional objectives
that some objectives are more important than the oth- were: quality standard of management services, quality
ers. In order to give more importance to an objective, it of maintenance of common property, work organiza-
could be multiplied with a Significance Coefficient. tion, the effectiveness of information use, certification
of company, range of services, reliability of company,
(6) company reputation, staff qualification and past experi-
ence, communication skills, geographical market restric-
where: j = 1, 2, ..., m; m – the number of alterna- tions.). However these additional objectives were only
tives; i = 1, 2,… n; n – the number of objectives; g – expressed in qualitative points, showed some overlap-
the number of objectives to be minimized; n–g – the ping and after their rating represented only 25.9% im-
number of objectives to be maximized; si – the signifi- portance of the total. If these opinions are only taken as
cance coefficient of objective i; – the normalized indicative these qualitative objectives can be dropped.

249
 W. K. M. Brauers et al. Multi-objective contractor’s ranking by applying the MOORA method

Table 1. Main attributes and objectives of maintenance contractors of dwellings in Vilnius

No. Attributes Units of measurement max/min

6. Length of time in maintenance business (experience) years max x1
7. Market share for each contractor % max x2
8. Number of projects per executive units/person max x3
9. Evaluation of management cost (Cmin / Cp)* – max x4
1. Cost of building management Lt*/m2 min x5
2. Cost of common assets management Lt/m2 min x6
3. HVAC system maintenance cost (mean) Lt/m2 min x7
4. Courtyard territory cleaning (in summer) Lt/m2 min x8
5. Total service cost Lt/m2 min x9
*The management cost evaluates the cost competitiveness of a contractor:
where: Cmin – the minimal cost in all offerings, Cp – the cost offered by the contractor considered.
Appendix A provides more details on the attributes.

Table2. Initial decision making matrix of 15 contractors of dwellings in Vilnius

Objectives ↔
Alternatives ↓
x1 x2 x3 x4 x5 x6 x7 x8 x9
a1 12 11.75 4.6 0.83 0.064 0.11 0.18 0.31 0.67
a2 3 0.39 0.33 0.885 0.06 0.14 0.37 0.12 0.5
a3 12 5.25 1.47 0.935 0.057 0.11 0.18 0.15 0.69
a4 12 7.09 2.78 0.912 0.058 0.12 0.09 0.15 0.57
a5 12 5.56 1.39 0.912 0.058 0.1 0.18 0.2 0.45
a6 13 26.62 5.67 0.746 0.071 0.3 0.18 0.26 0.82
a7 5 2.82 1.2 0.483 0.11 0.14 0.18 0.12 0.55
a8 11 9.48 3.03 0.916 0.058 0.18 0.37 0.19 0.61
a9 11 2.23 0.76 1 0.053 0.14 0.16 0.23 0.8
a10 11 13.47 9.05 0.746 0.071 0.26 0.29 0.23 0.73
a11 4 4.7 1.5 0.443 0.12 0.2 0.09 0.2 0.81
a12 12 2.35 0.86 0.746 0.071 0.28 0.18 0.28 0.73
a13 8 5.6 3.25 0.681 0.078 0.2 0.18 0.3 0.76
a14 11 2.66 1.7 0.948 0.056 0.14 0.18 0.12 0.5
a15 3 0.04 0.03 0.531 0.12 0.14 0.09 0.21 0.56

Is the owner’s information useful to allot significance Instead of attributing significance coefficients the con-
coefficients? Therefore the sample is not enough rep- tractors and the small group of owners preferred the
resentative. Attribution of Sub-Objectives. Indeed, five objectives
concern minimization of costs (Effectiveness). Even,
Indeed, a significance coefficient of importance was
the last maximization forms a cost consideration. Fur-
not possible to give to the 9 objectives as 30 interviews
thermore, one objective is related to Experience, as
even chosen at random mean a confidence level of only:
measured by length of time in maintenance business,
standard error which means one to Size as measured by market share and finally
one objective measures Effectiveness as expressed by
9% under or 9% above the real percentage. Economics
number of projects per executive.
generally accept 100 interviews with a standard error
Table 3 presents the results of the calculation proc-
of: which means 5% under ess of MOORA. Appendix B gives the details of this
or 5% above the real percentage (p = expected prob- calculation.
ability; q = 1 – p; in a symmetric distribution: q = p.

250
 Journal of Business Economics and Management, 2008, 9(4): 245–255

Table 3. Ranking of the 15 contractors (costs), which seems rather unusual. On the other side,
by the two parts of MOORA size of the enterprise seems to be very important. In this
way the comments that from the beginning no small
Contractors MOORA firms were considered are without any value.
Square root part Reference Point part
Concerning the use of the Maximal Objective Ref-
a1 4 3 erence Point Approach as a part of MOORA some
a2 12 14 reserves can be made in connection with consumer
a3 6 8 sovereignty of the house owners. Consumer sover-
a4 3 5 eignty is measured with the community indifference
a5 5 7 manifold map of these house owners (Brauers 2004).
a6 1 1 Given its definition the Maximal Objective Reference
a7
Point is pushed in the non-allowed non-convex zone
11 10
of the highest community indifference manifold (Brau-
a8 8 4
ers and Zavadskas 2006). Therefore an aspiration ob-
a9 9 13 jective vector can be preferred, which moderates the
a10 2 2 aspirations by choosing smaller co-ordinates than in
a11 14 9 the maximal objective vector and consequently can be
a12 13 12 situated in the convex zone of the highest community
a13 10 6 indifference manifold. Indeed stakeholders may be
a14 7 11 more moderate in their expectations. The co-ordinates
a15 15 15 qi of an Aspiration Objective Vector are formed as: qi ≤
ri. (ri – qi) being a subjective element we do not like
Both parts of MOORA method rank in the same way to introduce subjectivity in that way again. Instead:
1) a test shows that the Min-Max Metric of Tchebycheff
the first three positions after the quality of the contrac-
delivers points inside the convex zone of the highest
tors. In this way a double check is made on the results
community indifference manifold (Brauers 2008).
as shown in the following Table 4.
2) Given the higher number of minimizations com-
pared to the number of maximizations in the example
Table 4. No discussion on the ranking
a second argument exists to assume that the reference
of the first three positions
point is located in the convex zone.
MOORA
Contractors
Square root part Reference Point part 4. Conclusions
a6 1 1
Different approaches exist in Multi -Objective Deci-
a10 2 2 sion Making research. Given these different approach-
a1 4 3 es preference is given to an approach satisfying the
following conditions:
According to the results of Table 3, we can find the 1) cardinal and not ordinal, given the weak points of
priority between the contractors (P- preferred to): an ordinal approach;
a6 P a10 P a4 P a1 P a5 P a3 P a14 P a8 P a9 P a13 P a7 2) all objectives should be considered as much as pos-
P a2 P a12 P a11 P a15 (Square root part) and sible;
3) all stakeholders are involved and not only the decision
a6 P a10 P a1 P a8 P a4 P a13 P a5 P a3 P a11 P a7 P a14
maker or a group of decision makers. All stakehold-
P a12 P a9 P a2 P a15 (Reference Point part). ers mean everybody interested in a certain issue;
In other words three contractors are classified in a good 4) all interrelations between objectives and alternatives
order, a4 has still to be mentioned with its 3rd and 5th are looked upon at the same moment instead of pair-
ranking position, whereas contractor a15 is the very last wise considerations;
one. The other 10 contractors are ranked low but it is 5) only discrete cases facing a set of a limited number
unclear in what position. of alternatives are considered, whereas continuous
cases concern alternatives generated out of a set of
Contractor 6 is ranked first for size and experience and continuous and numerous alternatives;
second for effectiveness. Contractor 10 is ranked first for 6) as non-subjective as possible: no normalization by
effectiveness and second for size. Contractor 1 together subjective weights but by non-subjective dimen-
with contractor 4 are ranked second for experience. All sionless measures.
these strong contractors are not so good in effectiveness For all these reasons we selected MOORA.

251
 W. K. M. Brauers et al. Multi-objective contractor’s ranking by applying the MOORA method

The MOORA Method, based on dimensionless meas- Total service cost – covers the cost of building man-
ures, consists of two parts: the ratio system and the agement, cost of common assets’ management, HVAC
reference point approach, each controlling each other. system maintenance cost, courtyard territory cleaning
MOORA was applied for the choice between the 15 (in summer) cost, cost of territory cleaning in winter,
main contractors of dwellings to satisfy the wishes of cost of cleaning of internal surfaces, cost of repairs and
the owners of dwellings in the city of Vilnius, the capi- cost of other necessary services.
tal of Lithuania. Length of time in maintenance business (experi-
ence) – the lifetime of company in facilities manage-
The MOORA method came to the following results:
ment field from establishment.
three contractors take the first three positions. A fourth
one has still to be mentioned with its favourable rank- Market share for each contractor (in Vilnius) – in
ing position whereas one contractor is classified as the strategic management and marketing, is the percentage
very last one. The other 10 contractors are ranked low or proportion of the total available market or market
but it is unclear in what position. The best contractors segment that is being serviced by a company. It can
are not the best in effectiveness (costs), which seems be expressed as a company’s sales revenue (from that
market) divided by the total sales revenue available in
rather unusual. On the other side, the size of the en-
that market. It can also be expressed as a company’s
terprise seems to be very important. In this way the
unit sales volume (in a market) divided by the total
comments that from the beginning no small firms were
volume of units sold in that market. In this study this
considered are without any value. criterion shows the extent of work for each contractor
Even more, for contractors and their clients’ firm qual- and expressed as a company’s client number (only in
ity as measured by size, experience and effectiveness Vilnius) divided by the total number of clients for all
seems to dominate the cost price of a dwelling. A new companies (also only in Vilnius):
research based on the newest data, larger samples and a
larger number of quantified objectives may verify this
rather unexpected outcome.

where: MC – market share for each contractor; Cn –

Appendix A
number of clients of company n; – the total
Cost of building management – calculated average
value of all buildings’ management cost. Building number of clients for r companies; R – total number
management activities cover all administrative activi- of maintenance contractors involved in research (in our
ties necessary to manage building maintenance opera- case r = 15).
tions: monthly reports, budgeting, cost calculations, Number of projects per executive – this criterion eval-
hiring the subcontractors, control and coordination of uates the effectiveness of contractor’s personnel (labour
all subcontractors involved in building maintenance productivity). Higher value means effective organiza-
process, procurement, repair organization, preparation tion of work, presence of special procedures, tools, in-
of procedures, manuals and instructions, etc. structions, forms and other means for work coordination
and performance in time ensuring proper quality.
Cost of common assets’ management – common
assets mean all building equipment, structures, land Value is calculated according to the formula:
and internal spaces, which are the common property
of building owners.
HVAC system maintenance cost (average value) – where: Np – number of projects implemented per year;
HVAC mean heating, ventilating, and air conditioning Ne – average number of employees directly involved
systems, which are installed in building. In question- in projects being implemented per year.
naires this criterion was marked as one of the most Evaluation of management cost (Cmin/Cp) – this cri-
important criteria for building users. terion evaluates the cost competitiveness of contractor
(could be noted as Cec). Cmin is minimal cost in all
Courtyard territory cleaning (in summer) – an
offerings, Cp (could be noted anew as Cc – cost of
environmental issue is signifying the importance of
contractor) is the cost offered by contractor:
healthy surroundings. In questionnaires this criterion
was marked as one of the most important criteria for
building users.

252
 Journal of Business Economics and Management, 2008, 9(4): 245–255

Appendix B

MOORA: square root method (1a until 1c) and MOORA reference point theory (1d-1e)
Table 1 – MOORA applied on 9 objectives for Lithuanian contractors
1a – Matrix of Responses of Alternatives on Objectives: (xij)
1 2 3 4 5 6 7 8 9
max max max max min min min min min
a1 12 11.75 4.6 0.83 0.064 0.11 0.18 0.31 0.67
a2 3 0.39 0.33 0.885 0.06 0.14 0.37 0.12 0.5
a3 12 5.25 1.47 0.935 0.057 0.11 0.18 0.15 0.69
a4 12 7.09 2.78 0.912 0.058 0.12 0.09 0.15 0.57
a5 12 5.56 1.39 0.912 0.058 0.1 0.18 0.2 0.45
a6 13 26.62 5.67 0.746 0.071 0.3 0.18 0.26 0.82
a7 5 2.82 1.2 0.483 0.11 0.14 0.18 0.12 0.55
a8 11 9.48 3.03 0.916 0.058 0.18 0.37 0.19 0.61
a9 11 2.23 0.76 1 0.053 0.14 0.16 0.23 0.8
a10 11 13.47 9.05 0.746 0.071 0.26 0.29 0.23 0.73
a11 4 4.7 1.5 0.443 0.12 0.2 0.09 0.2 0.81
a12 12 2.35 0.86 0.746 0.071 0.28 0.18 0.28 0.73
a13 8 5.6 3.25 0.681 0.078 0.2 0.18 0.3 0.76
a14 11 2.66 1.7 0.948 0.056 0.14 0.18 0.12 0.5
a15 3 0.04 0.03 0.531 0.12 0.14 0.09 0.21 0.56
1b – Sum of squares and their square roots
a1 0.4489 144 138.0625 21.16 0.6889 0.0041 0.0121 0.0324 0.0961
a2 0.25 9 0.1521 0.1089 0.7832 0.0036 0.0196 0.1369 0.0144
a3 0.4761 144 27.5625 2.1609 0.8742 0.0033 0.0121 0.0324 0.0225
a4 0.3249 144 50.41 7.7284 0.81 0.0036 0.0144 0.01 0.0225
a5 0.2025 144 30.9136 1.9321 0.81 0.0034 0.01 0.0324 0.04
a6 0.6724 169 708.6244 32.1489 0.5565 0.0050 0.09 0.0324 0.0676
a7 0.3025 25 7.9524 1.44 0.2333 0.0121 0.0196 0.0324 0.0144
a8 0.3721 121 89.8704 9.1809 0.8390 0.0034 0.0324 0.1369 0.0361
a9 0.64 121 4.9729 0.64 1 0.0028 0.0196 0.0256 0.0529
a10 0.49 121 182.25 81.9025 0.5625 0.0049 0.0676 0.0841 0.04
a11 0.6561 16 22.09 2.25 0.1962 0.0144 0.04 0.0081 0.04
a12 0.5329 144 5.5225 0.7396 0.5565 0.0050 0.0784 0.0324 0.0784
a13 0.5776 64 31.36 10.5625 0.4638 0.0061 0.04 0.0324 0.09
a14 0.25 121 7.0756 2.89 0.8987 0.0031 0.0196 0.0324 0.0144
a15 0.3136 9 0.0016 0.0009 0.2820 0.0144 0.0196 0.0081 0.0441
Σ 6.5096 1496 1306.8205 174.8456 9.5549 0.0892 0.4950 0.6689 0.6734
root 2.5514 38.6782 36.1500 13.2229 3.0911 0.2986 0.7036 0.8177 0.8206
1c – Objectives divided by their square roots and MOORA *
sum 1.9336 rank
a1 0.2626 0.3102 0.3250 0.3480 0.2690 0.2143 0.1563 0.2201 0.3778 0.0206 1.9541 4
a2 0.1960 0.0776 0.0108 0.0250 0.2863 0.2009 0.1990 0.4524 0.1462 -0.7949 1.1386 12
a3 0.2704 0.3102 0.1452 0.1112 0.3025 0.1907 0.1563 0.2201 0.1828 -0.1514 1.7821 6
a4 0.2234 0.3102 0.1964 0.2102 0.2912 0.2009 0.1706 0.1223 0.1828 0.1081 2.0416 3
a5 0.1764 0.3102 0.1538 0.1051 0.2912 0.1942 0.1421 0.2201 0.2437 -0.1162 1.8173 5
a6 0.3214 0.3361 0.7364 0.4288 0.2413 0.2377 0.4264 0.2201 0.3168 0.2202 2.1537 1
a7 0.2156 0.1293 0.0780 0.0908 0.1562 0.3683 0.1990 0.2201 0.1462 -0.6949 1.2386 11
a8 0.2391 0.2844 0.2622 0.2291 0.2963 0.1942 0.2558 0.4524 0.2315 -0.3010 1.6326 8
a9 0.3136 0.2844 0.0617 0.0605 0.3235 0.1775 0.1990 0.1956 0.2803 -0.4358 1.4977 9
a10 0.2744 0.2844 0.3734 0.6844 0.2426 0.2344 0.3695 0.3546 0.2437 0.1083 2.0418 2
a11 0.3175 0.1034 0.1300 0.1134 0.1433 0.4018 0.2843 0.1100 0.2437 -0.8671 1.0664 14
a12 0.2861 0.3102 0.0650 0.0650 0.2413 0.2377 0.3980 0.2201 0.3412 -0.8015 1.1320 13
a13 0.2979 0.2068 0.1549 0.2458 0.2203 0.2612 0.2843 0.2201 0.3656 -0.6012 1.3324 10
a14 0.1959 0.2844 0.0736 0.1286 0.3067 0.1875 0.1990 0.2201 0.1462 -0.1556 1.7780 7
a15 0.2195 0.0776 0.0011 0.0023 0.1718 0.4018 0.1990 0.1100 0.2559 -0.9335 1.0000 15
* The nature of the construction industry involves that the total number of the minima is mostly larger than the total number
of the maxima with very often negative sums as a result, which is the case here. Therefore, to make the ranking more com-
prehensive, the supplement to make the smallest sum, here for a15, equal to one, is added to all the sums.

253
 W. K. M. Brauers et al. Multi-objective contractor’s ranking by applying the MOORA method

Table 1 – MOORA applied on 9 objectives for Lithuanian contractors. Reference Point Part
1d – Reference Point Theory with Ratios: co-ordinates of the reference point equal to the maximal objective values
ri 0.1764 0.3361 0.7364 0.6844 0.3235 0.1775 0.1421 0.1100 0.1462
1e – Reference Point Theory: Deviations from the reference point max. rank min.
a1 0.0862 0.0258 0.4113 0.3365 0.0550 0.0368 0.0142 0.1100 0.2315 0.4113 3
a2 0.0196 0.2585 0.7256 0.6595 0.0372 0.0234 0.0569 0.3424 0.0000 0.7256 14
a3 0.0941 0.0258 0.5911 0.5732 0.0210 0.0134 0.0142 0.1100 0.0366 0.5911 8
a4 0.047 0.0258 0.5400 0.4742 0.0324 0.023 0.028 0.012 0.037 0.5400 5
a5 0.0000 0.0258 0.5826 0.5793 0.0324 0.0167 0.0000 0.1100 0.0975 0.5826 7
a6 0.1450 0 0 0.2556 0.0822 0.0603 0.2843 0.1100 0.1706 0.2843 1
a7 0.0392 0.2068 0.6584 0.5937 0.1672 0.1909 0.0569 0.1100 0 0.6584 10
a8 0.063 0.0517 0.4741 0.4553 0.0272 0.017 0.114 0.342 0.085 0.4741 4
a9 0.1372 0.0517 0.6747 0.6239 0 0.0000 0.0569 0.0856 0.1340 0.6747 13
a10 0.0980 0.0517 0.3629 0 0.0809 0.0569 0.2274 0.2445 0.0975 0.3629 2
a11 0.1411 0.2327 0.6064 0.5710 0.1802 0.2244 0.1421 0.0000 0.0975 0.6064 9
a12 0.1097 0.0258 0.6714 0.6194 0.0822 0.0603 0.2558 0.1100 0.1950 0.6714 12
a13 0.1215 0.1293 0.5815 0.4386 0.1032 0.0837 0.1421 0.1100 0.2193 0.5815 6
a14 0.0196 0.0517 0.6628 0.5558 0.0168 0.0100 0.0569 0.1100 0.0000 0.6628 11
a15 0.0431 0.2585 0.7353 0.6821 0.1517 0.2244 0.0569 0.0000 0.1097 0.7353 15

References Churchman, C. W.; Ackoff, R. L.; Arnoff, E. L. 1957. Intro-

duction to operations research. New York: Wiley.
Arrow, K. J. 1974. General economic equilibrium: purpose, Churchman, C. W.; Ackoff, R. L. 1954. An approximate
analytic techniques, collective choice, American Economic measure of value, Operations Research 2: 172–180.
Review 64(3): 253–272.
Colson, G. and Bruyn, C. D. 1989. Models and methods in
Banaitiene, N.; Banaitis, A. 2006. Analysis of criteria for multiple criteria decision making. New York: Pergamon.
contractors‘ evaluation, Technological and Economic De-
Dikmen, I.; Birgonul, M. T.; Gur, A. K. 2007. A case-based
velopment of Economy 12(4): 276–282.
decision support tool for bid mark-up estimation of inter-
Brauers, W. K. 2008. Multi-objective decision making by national construction projects, Automation in Construction
reference point theory for a wellbeing economy, Operations 17(1): 30–44.
Research International Journal 8: 89–104.
Ginevičius, R. and Korsakienė, R. 2005. Exploration of
Brauers, W. K. 2007. What is meant by normalization in strategy: objectives, competencies and competitive advan-
Decision Making? International Journal of Management tage, Journal of Business Economics and Management 6(1):
and Decision Making 8(5/6): 445–460. 13–22.
Brauers, W. K. 2004. Optimization methods for a stakehold- Ginevičius, R. and Vaitkūnaitė, V. 2006. Analysis of organi-
er society. Boston: Kluwer Academic Publishers. zational culture dimensions impacting performance, Journal
Brauers, W. K. 2002. The multiplicative representation for of Business Economics and Management 7(4): 201–211.
multiple objective optimization with an application for arms Hatush, Z. and Skitmore, R. M. 1997. Assessment and eval-
procurement, Naval Research Logistics 49: 327–340. uation of contractor data against client goals using PERT
Brauers, W. K. 1999. An optimal economic policy for Bel- approach, Construction Management and Economics 15(4):
gium and its regions (in Dutch), Documentatieblad. Depart- 327–340.
ment of Finance of the Belgian Government, Brussels. Hatush, Z. and Skitmore, R. M. 1998. Contractor selection
Brauers, W. K. 1997a. The multiplicative representation using multi-criteria utility theory: an additive model, Build-
in multiple objective utility theory, Working Paper 97/06, ing and Environment 33(2/3): 105–115.
Department of Applied Economics, University of Antwerp. Holmes, J. C. 1971. An ordinal method of evaluation, Urban
Brauers, W. K. 1977b. Multiple criteria decision making and Studies 9: 179–191.
the input-output model for national defense. Belgian Depart- Hwang, C. L. and Yoon, K. S. 1981. Multiple attribute deci-
ment of National Defense, Brussels. sion making/methods and applications. Berlin, Heidelberg,
Brauers, W. K.; Zavadskas, E. K. 2006. The MOORA New York: Springer-Verlag.
method and its application to privatization in a transition Jaselskis, E. J.; Russell, J. S. 1992. Risk analysis approach
economy, Control and Cybernetics 35(2): 445–469. to selection of contractor evaluation method, ASCE Jour-
Bridgman, P. W. 1922. Dimensional analysis. New Haven, nal of Construction Engineering and Management 118(4):
CT: Yale University Press (new ed. 1963). 814–821.

254
 Journal of Business Economics and Management, 2008, 9(4): 245–255

Karlin, S. and Studden, W. J. 1966. Tchebycheff systems: selection system, in International Symposium on Automation
with applications in analysis and statistics. Interscience and Robotics in Construction, 19th (ISARC) Proceedings,
Publishers, New York. National Institute of Standards and Technology, Gaithers-
Keeney, R. L. and Raiffa, H. 1993. Decisions with multi- burg, Maryland , 35–40.
ple objectives, preferences and value tradeoffs. Cambridge Skibniewski, M. J. and Chan, L. C. 1992. Evaluation of
[England]; New York, NY, USA: Cambridge University advanced construction technology with AHP method, Jour-
Press. nal of Construction Engineering and Management 118(3):
Kendall, M. G. 1948. Rank correlation methods. London: 577–593.
Charles Griffin & Company, Ltd. Spearman, C. 1904. The proof and measurement of associa-
Kersuliene, V. 2007. Possibilities of clients and contractor’s tion between two things, American Journal of Psychology
disputes settlement at the pretrial stage, Technological and 15: 88–93.
Economic Development of Economy 13(2): 139–143. Spearman, C. 1906. A footrule for measuring correlation,
Lee, S. M. 1972. Goal programming for decision analysis. British Journal of Psychology 2: 89–108.
Philadelphia: Auerbach Publishers. Spearman, C. 1910. Correlation calculated from faulty data,
Lin, C. T. and Wu, C. R. 2007. Editorial: qualitative and British Journal of Psychology 3: 271–295.
quantitative methods aimed at technology management,
Topcu, Y. I. 2004. A decision model proposal for construc-
International Journal of Technology Management (IJTM)
tion contractor selection in Turkey, Building and Environ-
40(1/2/3): 1–6.
ment 39: 469–481.
Miller, D. W. and Starr, M. K. 1969. Executive decisions and
Tserng, H. P.; Lin, P. H. 2002. An accelerated subcontracting
operations research, 2nd edition, Englewood Cliffs, N. J.:
and procuring model for construction projects, Automation
Prentice-Hall Inc., 237–239.
in Construction 11(1): 105–125.
Minkowsky, H. 1896. Geometrie der Zahlen. Teubner, Leipzig.
Turskis, Z.; Zavadskas, E. K; Zagorskas, J. 2006. Sustain-
Mitkus, S. and Trinkuniene, E. 2007. Analysis of criteria able city compactness evaluation on the basis of GIS and
system model for construction contract evaluation, Tech- Bayes rule, International Journal of Strategic Property
nological and Economic Development of Economy 13 (3): Management 10: 185–207.
244–252.
Zagorskas, J. and Turskis, Z. 2006. Multi-attribute model
Mitkus, S. and Trinkūnienė, E. 2006. Models of indicator for estimation of retail centres influence on the city struc-
systems of construction contraction agreements, Journal of
ture. Kaunas city case study, Technological and Economic
Civil Engineering and Management 12(4): 327–335.
Development of Economy 12(4): 347–352.
Ng, S. T.; Skitmore, R. M.; Smith, N. J. 1999. Decision-
Zavadskas, E. K. and Kaklauskas, A. 1996. Determination
makers perceptions in the formulation of prequalification
of an efficient contractor by using the new method of mul-
criteria, Engineering Construction & Architectural Manage-
ticriteria assessment, in International Symposium for the
ment 6(2): 155–165.
Organisation and Management of Construction. Shaping
Ng, S. T. and Wan, W. Y. 2005. Appraisal of subcontrac- Theory and Practice. Vol. 2: Managing the Construction
tor performance — criteria and their importance, in A. S. Project and Managing Risk. Ed. by Langford, D. A. and
Kazi (eds.). Proceedings of CIB2005— Advancing Facilities Retik, A. CIB W 65. London, Weinheim, New York, Tokyo,
Management and Construction through Innovation, Finlan- Melbourne, Madras. London: E and FN SPON 2: 94–104.
dia Hall, Helsinki, Finland , 305–314.
Zavadskas, E. K. and Vilutienė, T. 2006. A multiple criteria
Roy, B.; Benayoun, R.; Sussman, B. 1966. ELECTRE, So- evaluation of multi-family apartment block’s maintenance
ciété d’Economie et de Mathématique Appliquées. contractors: I—Model for maintenance contractor evaluation
Saaty, T. L. 1988. The analytic hierarchy process. New York: and the determination of its selection criteria, Building and
McGraw-Hill. Environment 41(5): 621–632.
Sage, A. P. 1977. Methodology for large scale systems. New Zavadskas, E. K. and Kaklauskas, A. 2007. Mehrzielsele-
York: McGraw-Hill. ktion für Entscheidungen im Bauwesen. Fraunhofer IRB
Schärlig, A. 1985. Décider sur plusieurs critères. Lausanne Verlag.
Presses Polytechniques Romandes. Zimmermann, H. J. 1978. Fuzzy Programming and Linear
Schieg, M. 2006. Risk management in construction project Programming with several Objective Functions, Journal of
management, Journal of Business Economics and Manage- Fuzzy Sets and Systems 1: 45–55.
ment 7(2): 77–83. Zou, P. X. W.; Zhang, G.; Wang, J. 2007. Understanding the
Shiau, Y. C.; Tsai, T. P.; Wang W. C.; Huang, M. L. 2002. Use key risks in construction projects in China, International
questionnaire and AHP techniques to develop subcontractor Journal of Project Management 25(6): 601–614.

255