el

ChV1, Br

Multi-criteria decision makingAnalytical hierarchy process

n rvidof pialvieiple

tionnaire survey of building experts is conducted to assess the relative importance of the criteria andaggregate them into six independent assessment factors. The FEAHP is used to prioritize and assign important

d criteria. A numerical example, illustrating the implementation of the model isel provides guidance to building designers in selecting sustainable building

n andan impenergindire

Automation in Construction 30 (2013) 113125

Contents lists available at SciVerse ScienceDirect

Automation in

j ourna l homepage: www.e lsto environmental performance targets that appropriate strategiesand actions are needed to make construction activities more sustain-able [13]. The pace of actions towards sustainable application de-pends on decisions taken by a number of actors in the constructionprocess: owners, managers, designers, rms, etc. [4,3]. An importantdecision is the sustainable selection of building materials to be usedin building projects. Careful selection of sustainable building mate-rials has been identied as the easiest way for designers to beginincorporating sustainable principles in building projects [5]. The se-lection of building materials is regarded as a multi-criteria decision

The earlier attempt to establish comprehensive means of simul-taneously assessing a broad range of sustainability considerations inbuilding materials was the Building Research Establishment Envi-ronmental Assessment Method (BREEAM) [9]. BREEAM known asthe rst commercially available and most widely used assessmentmethod was established in 1990 in the United Kingdom. Sincethen many different tools have been launched around the world(e.g. Building for Environmental and Economic Sustainability(BEES), Leadership in Energy and Environmental Design (LEED),Building environmental performance assessment criteria (BEPAC),problem [6], largely based on trusting exper

Corresponding author. Tel.: +44 7828049507.E-mail addresses: arcalad@yahoo.com (P.O. Akadiri)

(P.O. Olomolaiye), E.Chinyio@wlv.ac.uk (E.A. Chinyio).1 Tel.: +44 117 32 82211.2 Tel.: +44 1902 321043.

0926-5805/$ see front matter 2012 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.autcon.2012.10.004ctly (through the pres-response to these im-

rganizations committed

stage in decision-making. It should be noted that this pure form ofsustainability assessment is a challenge to develop and evidence ofachieving this in practice is yet to be seen [8].sures on often inefcient infrastructure). Inpacts, there is growing consensus among o1. Introduction

The construction, t-out, operatiobuildings are signicant factors of humboth directly (through material andconsequent pollution and waste) and 2012 Elsevier B.V. All rights reserved.

ultimate demolition ofact on the environmenty consumption and the

numerical approach, due to lack of formal and availability of measure-ment criteria or strategies. In addition, many of the current evaluationapproaches were criticized for overemphasizing the environmentalaspects [7]. Ideally, sustainability assessment would integrate social,technical, environmental and economic considerations at everymaterials.Fuzzy logic weightings for the identiegiven. The proposed modMulti-criteria evaluation model for the sbuilding projects

Peter O. Akadiri a,, Paul O. Olomolaiye b,1, Ezekiel A.a School of Technology, University of Wolverhampton, Wulfruna Street, Wolverhampton, Wb Faculty of Environment and Technology, University of West of England, Coldharbour Lane

a b s t r a c ta r t i c l e i n f o

Article history:Accepted 10 October 2012Available online 12 December 2012

Keywords:Building material selectionSustainable criteria

Sustainable material selectioselection methods fail to proprinciples, and the processproposes a building mater(FEAHP) techniques, with aed based on sustainable trience rather than using

, paul.olomolaiye@uwe.ac.uk

rights reserved.ection of sustainable materials for

inyio a,2

1LY, UKistol, BS16 1QY, UK

epresents an important strategy in building design. Current building materialse adequate solutions for two major issues: assessment based on sustainabilityrioritizing and assigning weights to relevant assessment criteria. This paperselection model based on the fuzzy extended analytical hierarchy processw to providing solutions for these two issues. Assessment criteria are identi-bottom line (TBL) approach and the need of building stakeholders. A ques-

Construction

ev ie r .com/ locate /autconEnvironmental Resource Guide (ERG), and Environmental ResourceGuide (ERG)). BREEAM, LEEDS, ENVEST and other existing methodsfor assessing buildings whose remit is largely restricted to an envi-ronmental protection and resource efciency agenda have limitedutility for assessing socio and economic factors as opposed to envi-ronmental sustainability, since they are predominantly focused onenvironment which is just one of the four principles underpinningsustainable building. Even against this single principle, they are

Saaty [17], Saaty and Shang [27].Even though AHP has been widely used to address the multi-

114 P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125only able to offer relative assessment as opposed to absolute [10].Another criticism that has been raised concerns the fact that the ma-jority of the assessment methods were designed for new construc-tion, and hence have focused on the design of the constructedbuildings. Although energy, water and occupant comfort were wellcovered in the tools, there was little focus on the effect of the build-ing system's life during operation. This is especially true for envelopeperformance. This tendency has resulted in the failure of manyassessment methods to properly consider other assessment criteriasuch as durability, lifecycle cost, and the effects of premature build-ing envelope failures. To be considered truly sustainable, assessmentmethods will have to be recast under the umbrella of sustainabilityenvironmental, social, technical and economic [11]. Broadeningthe scope of discussion beyond environmental responsibility andembracing the wider agenda of sustainability are increasingly neces-sary requirements.

Therefore there is a need for developing a systematic and holisticsustainable material selection process of identifying and prioritizingrelevant criteria and evaluating trade-offs between environmental,economic, social and technical criteria [12]. The characterization ofmaterial selection process as an essentially multifaceted probleminvolving numerous, variegated considerations, often with complextrade-offs among them, implied that a suitable solution might befound among the family of multi-criteria decision analysis (MCDA)methods [1316]. Further analysis and proling of the selection prob-lem and the identication of the solution methods' desirable capabil-ities, triggered the consideration of the Analytic Hierarchy Process(AHP) developed by Saaty [17] as a possible basis for sustainablematerial selection method envisaged.

The analytic hierarchy process (AHP) [17,18] is widely used fortackling multi-criteria decision-making problems in real situations.Bahareh et al. [19] utilized the AHP as a multi-criteria technique forsustainable assessment of ooring systems. They agree that AHP pro-vides a framework for robust decision making that is consistent withsustainable construction practices. The use of AHP for sustainableassessment has been considered in approaches developed by otherresearchers [6,7,13,1921]. In spite of its popularity and simplicityin concept, this method is often criticized for its inability to adequate-ly handle the inherent uncertainty and imprecision associated withthe mapping of the decision-maker's perception to exact (or crisp,according to the fuzzy logic terminology) numbers.

To improve the AHP method and to facilitate sustainable materials'selection process, the paper uses a fuzzy extended AHP (FEAHP) ap-proach using triangular fuzzy numbers to represent decision makers'comparison judgments and fuzzy synthetic extent analysis [22]methodto decide the nal priority of different decision criteria. The fuzzy settheory resembles human reasoning in its use of approximate informa-tion and uncertainty to generate decisions. It has the advantage ofmathematically representing uncertainty and vagueness and provideformalized tools for dealing with the imprecision intrinsic to manyproblems [22,23]. The proposed FEAHP uses the triangular fuzzynumbers as a pair-wise comparison scale for deriving the prioritiesof different selection criteria and sub-criteria. The weight vectorswith respect to each element under a certain criterion are developedusing the principle of the comparison of fuzzy numbers. As a result,the priority weights of the each material are calculated and basedon that, the most sustainable material is selected. In particular, theapproach developed can adequately handle the inherent uncertaintyand imprecision of the human decision making process and providethe exibility and robustness needed for the decision maker to un-derstand the decision problem. These merits of the approach devel-oped would facilitate its use in real-life situations for makingeffective decisions.

Based on this information and the current research deciencies,this paper proposes a multi-criteria decision-making model using

the Fussy extended analytic hierarchy process (FEAHP) approach tocriterion decision making problems, it has been generally criticized be-cause of the use of a discrete scale of one to nine which cannot handlethe uncertainty and ambiguity present in deciding the priorities of dif-ferent attributes [28]. Even though the discrete scale of AHP has the ad-vantages of simplicity and ease of use, it is not sufcient to take intoaccount the uncertainty associated with the mapping of one's percep-tion to a number [29]. The linguistic assessment of human feelingsand judgments is vague and it is not reasonable to represent it interms of precise numbers. It feels more condent to give interval judg-ments than xed value judgments [28]. In this condition, linguistic vari-ables and triangular fuzzy numbers can be used to decide the priority ofone decision variable over the other. Synthetic extent analysis methodis used to decide the nal priority weights based on triangular fuzzyevaluate building materials based on their sustainability. First,Section 2 describes the proposed FEAHP approach. The developmentof sustainable assessment criteria for building material selectionused in the FEAHP was discussed in Section 3. Section 4 discussesthe complete implementation of the FEAHP approach. The priorityweights computed for different criteria, sub-criteria and alternativesare also discussed in this section. Finally, Section 5 draws conclusionsand gives recommendations where necessary. The current study con-tributes to the building industry and sustainability research in at leasttwo aspects. First it widens the understanding of selection criteria aswell as their degree of importance. It also provides building stake-holders a newway to select materials, thereby facilitating the sustain-ability of building projects.

2. Fuzzy analytic hierarchy process methodology

2.1. Background

Singh et al. [24] describe AHP method as a multiple step analyticalprocess of judgment, which synthesizes a complex arrangement into asystematic hierarchical structure. It allows a set of complex issues thathave an impact on an overall objective to be compared with the impor-tance of each issue relative to its impact on the solution of the prob-lem [25]. It is designed to cope with the intuitive, the rational, and theirrational when making multi-objective, multi-criterion and multi-actordecisionsexactly the decision-making situation foundwithmaterial se-lection. Furthermore, it can easily be understood and applied by decisionmakers saddled with building material selection process.

The application of AHP to a decision problem involves four steps[25,26]. In structuring of the decision problem into a hierarchicalmodel, material selection problem is dened, objective is identied,criteria and attributes thatmust be satised to objective are recognized.Objective is at rst level, criteria is at second level, attributes are at thirdlevel, and decision alternatives are at fourth level in hierarchical struc-ture of the problem. In making pair-wise comparisons and obtainingthe judgment matrix, the elements at a particular level are comparedusing nine-point numerical scale to dene howmuchmore an elementis important than other. If A and B are the elements to be compared,then 1 denes that A and B are equal in importance, and 9 denesthat A is extremely more important. All pair-wise comparisons aregiven in a judgment matrix. The next step is to determine the localweights and consistency of comparisons. Local weights of the elementsare calculated from the judgment matrix using eigenvector method. Asthe comparisons in the matrix are made subjectively, consistency ratiocan be computed. If the ratio is less than 0.1 human judgments is ac-ceptable. In the last step, local weights at various levels are aggregatedto obtain nal weight of alternatives. The nal weights represent therating of the alternatives in achieving the aim of the multi-criterion de-cision making problem. Further information about AHP can be found innumbers and so-called as fuzzy extended AHP (FEAHP) [30].

Fuzzy set theory has proven advantageswithin vague, imprecise anduncertain contexts and it resembles human reasoning in its use of ap-proximate information and uncertainty to generate decisions [31]. Itwas specially designed to mathematically represent uncertainty andvagueness and provide formalized tools for dealing with the impreci-sion intrinsic tomany decision problems. The FEAHP is the fuzzy exten-sion of AHP to efciently handle the fuzziness of the data involved in thedecision of selecting building materials based on their sustainability. Itis easier to understand and it can effectively handle both qualitativeand quantitative data in the multi-attribute decision making problems.In this approach triangular fuzzy numbers are used for the preferencesof one criterion over another and then by using the extent analysismethod, the synthetic extent value of the pairwise comparison is calcu-lated. Based on this approach, the weight vectors are decided and nor-malized, thus the normalized weight vectors will be determined. As aresult, based on the different weights of criteria and attributes thenal priorityweights of the alternative sustainablematerial are decided.

Fuzzy numbers are intuitively easy to use in expressing thedecision-maker's qualitative assessments. A fuzzy number can alwaysbe given by its corresponding left and right representation of each de-gree of membership [33]:

M M1 y ;Mr y l ml y;u mu y

; y 0;1 2

where l(y) and r(y) denote the left side representation and the rightside representation of a fuzzy number, respectively. The algebraic op-erations with fuzzy numbers can be found in [30,33].

TFNs M1, M3, M5, M7 and M9 are used to represent the pairwisecomparison of decision variables from Equal to Extremelypreferred, and TFNs M2, M4, M6 and M8 represent the middle prefer-ence values between them. Fig. 2 shows the membership functions ofthe TFNs,Mi=(mi1,mi2,mi3), where i=1, 2,, 9 andmi1,mi2,mi3 arethe lower, middle and upper values of the fuzzy number Mi respec-tively. Higher the value of (mi3mi1) or (mi1mi3) signies the greaterfuzziness of the judgment.

115P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125The highest priority would be given to the material with highest sus-tainability weight.

2.2. Establishment of triangular fuzzy numbers

Saaty [17] contended that the geometric mean accurately repre-sents the consensus of experts and is the most widely used in practi-cal applications. Here, geometric mean is used as the model fortriangular fuzzy numbers. Zadeh [32] introduced the fuzzy set theoryto deal with the uncertainty due to imprecision and vagueness.A major contribution of fuzzy set theory was its capability ofrepresenting vague data. The theory also allowed mathematical oper-ators and programming to apply to the fuzzy domain. A fuzzy set is aclass of objects with a continuum of grades of membership. Such a setis characterized by a membership function, which assigns to each ob-ject a grade of membership ranging between 0 and 1. In this set thegeneral terms such as large, medium, and small will each beused to capture a range of numerical values. A triangular fuzzy num-ber (TFN), M is shown in Fig. 1. A TFN is denoted simply as (l, m, u).The parameters l, m, and u denote the smallest possible value, themost promising value and the largest possible value that describe afuzzy event [33]. When l=m=u, it is a non-fuzzy number by con-vention [30]. Each TFN has linear representations on its left andright side such that its membership function can be dened as [33];

x 0; xbl;xl = ml ; lxm;ux = um ;mxu;0; x > u:

8>>>:

9>>=>>;: 1

Ml(y) Mr(y)

M

M

1.0

0.0l m u

Fig. 1. A triangular fuzzy number, M [31].2.3. Calculation of priority weights at different level of hierarchy

In crisp AHP, a discrete scale of one to nine is used to decide thepriority of one decision variable over another whereas in fuzzy AHPfuzzy numbers or linguistic variables are used. In practice, decisionmakers usually prefer triangular or trapezoidal fuzzy numbers.

Since fuzzy numbers are used in fuzzy AHP, the solution methodsdifferentiate from crisp AHP. The most common method used in thesolution of fuzzy AHP applications is the extent analysis method pro-posed by Chang [34]. The extent analysis method is used to considerthe extent of an object to be satised for the goal, that is, satised ex-tent. In the method, the extent is quantied by using a fuzzy num-ber. On the basis of the fuzzy values for the extent analysis of eachobject, a fuzzy synthetic degree value can be obtained, which is de-ned as follows.

In a material selection problem, let p={p1, p2,, pn} represent theelements of the alternatives as an object set and let Q={q1, q2,, qm}represent the elements of the material selection criteria as a goal set.According to the concept of Chang's [34] extent analysis, each object istaken and extent analysis for each goal, Oi, is performed respectively.Therefore them extent analysis values for each object can be obtained,with the following signs:

M1oi;M2oi; :::::::;M

moi ; where i 1;2; :::::::;n 3

where all theMoij (j=1, 2,........ m) are triangular fuzzy numbers.

M1 M2 M3 M4 M5 M6 M8

equallyimportant

moderatelyimportant

stronglyimportant

very strongly important

extremelyimportant

1 3 5 7 9 X

M(x)

1

0 2 4 6 8

M7 M9Fig. 2. The membership functions of the triangular fuzzy numbers.

The steps of Chang's extent analysis can be given as in the follow-ing [33]:The value of fuzzy synthetic extent with respect to ith objectis dened as

Fi Xmj1

MjoiXni1

Xmj1

Mjoi

24

351: 4

The value ofj=1m Moij can be found by performing the fuzzy addi-tion operation of m extent analysis values from a particular matrixsuch that

116 P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125Xmj1

Mjoi Xmj1

lj;Xmj1

mj;Xmj1

uj

0@

1A 5

and the value of [i=1n j=1m Moij ]1 can be obtained by performingthe fuzzy addition operation of Moij (j=1, 2,....... m) such that

Xni1

Xmj1

Mjoi Xni1

li;Xni1

mi;Xni1

ui

!6

and then compute the inverse of the vector in equation above Eq. (6)such that

Xni1

Xmj1

Mjoi

24

351 1Xn

i1ui

;1Xn

i1mi

;1Xn

i1li

0BBBB@

1CCCCA: 7

1 The degree of possibility of M2=(l2, m2, u2)M1=(l1, m1,u1) isdened as

V M2M1 supxy

min M1 x ; M2 y h i

8

when a pair (x,y) exists such that xy and M1 x M2 y 1;then we have V(M1M2)=1. Since M1 and M2 are convex fuzzynumbers so, V(M1M2)=1 if m11m21 and

V M2M1 hgt M1M2 M2 d ; 9

where d is the ordinate of the highest intersection point D betweenM1 and M2 : Fig. 3 shows the intersection between M1 and M2. Tocompare M1 and M2, we need both the values of V(M1M2) andV(M2M1).

2 The degree of possibility for a convex fuzzy number to be greaterthan k convex fuzzy numbers Mi (i=1, 2,, k) can be dened by

V MM1;M2; :::::;Mk V MM1 and MM2 and ::::and MMk minV MMi ; i 1;2; ::::::k:

10Fig. 3. The intersection between M1 and M2 [35].If

m Pi minV FiFk ; 11

for k=1, 2,......, n; k i, then the weight vector is given by

Wp m P1 ;m P2 ; ::::;m Pn T ; 12

where Pi(i=1, 2,..... n) are n elements.3 After normalizing Wp, we get the normalized weight vectors as

W w P1 ;w P2 ; :::::;w Pn T ; 13

whereW is a non-fuzzy number and this gives the priority weightsof one alternative over the other.

3. Development of sustainable assessment criteria

One of the main objectives of this paper is to develop a holisticsustainable assessment criteria (SAC) set to assist design team mem-bers in the selection of sustainable building materials for buildingproject. A wide scope review of literature revealed that there wasno comprehensive list of assessment criteria that covers the princi-ples of sustainability, developed specically for material selection inbuilding projects. In trying to develop a set of criteria, Foxon et al.[36] proposed the consideration of two key factors. What use willbe made of this set of criteria? To what extent can any set of criteriaencompass the range of issues to be considered under the headingof sustainability? Some of these issues have been considered in ap-proaches developed by other researchers [37,12,24,7,38,39]. The fol-lowing set of guidelines has been developed to aid the choice ofcriteria to assess the options under consideration:

(1) ComprehensivenessThe criteria chosen should cover the four categories of eco-nomic, environmental, social and technical, in order to ensurethat account is being taken of progress towards sustainabilityobjectives. The criteria chosen need to have the ability to dem-onstrate movement towards or away from sustainability,according to these objectives.

(2) ApplicabilityThe criteria chosen should be applicable across the range of op-tions under consideration. This is needed to ensure the compa-rability of the options.

(3) TransparencyThe criteria should be chosen in a transparent way, so as tohelp stakeholders to identify which criteria are being consid-ered, to understand the criteria used and to propose anyother criteria for consideration.

(4) PracticabilityThe set of criteria chosenmust form a practicable set for the pur-poses of the decision to be assessed, the tools to be used and thetime and resources available for analysis and assessment. Clearly,the choice of sustainability criteria will inuence the outcome ofthe decision beingmade, aswill themethod of comparison or ag-gregation chosen. The above factors provide initial guidance inthe choice of criteria. Using a pool of existing criteria, combinedwith sustainable concerns and requirements of project stake-holders, a list of assessment criteria (see Table 1) was developed.These criteria are identied under three categories: Environmen-tal, Technical and Socio-economic.

These categories aim to encapsulate the economic, environmentaland social principles of sustainability, together with technical criteria,which relate primarily to the ability of buildings and its componentsystem to sustain and enhance the performance of the functions for

which it is designed. For any decision process, the selected criteria

must be broadly applicable to allmaterial options if comparative eval-uation is to be achieved.

3.1. Importance of derived criteria

Based on the derived criteria in Table 1, an industry questionnairesurvey was designed to investigate the perspective of constructionprofessionals on the importance of the criteria for material selection.Ninety experts comprising of architects, building designers, contractorsand structural engineers that inuence material selection were thusasked to rate the level of importance of the derived criteria based on a

index analysis was selected in this study to rank the criteria according totheir relative importance. The following formula is used to determine therelative index (RI) [5962]:

RI w=AxN 14

where w, is the weighting as assigned by each respondent on a scale ofone to ve with one implying the least and ve the highest. A is thehighestweight (i.e. 5 in our case) andN is the total number of the sample.Five important levels are transformed from Relative Index values: High(H) (0.8RI1), HighMedium (HM) (0.6RIb0.8), Medium (M)

Table 1Sustainable assessment criteria for building material selection.

Source Environmental criteria Social-economic criteria Technical criteria

Literature review, existing assessmentmethods and focus ofconstruction stakeholders

E1: Potential for recycling and reuse [40]E2: Availability of environmentally sounddisposal options [41]E3: Impact of material on air quality [19]E4: Ozone depletion potential [42]E5: Environmental Impact duringmaterial harvest [43]E6: Zero or low toxicity [42]E7: Environmental statutory compliance [44]E8: Minimize pollution e.g. air, land [19]E9: Amount of likely wastage in useof material [45,42]E10: Method of raw material extraction [43]E11: Embodied energy within material [46,47]

S1: Disposal cost [39,48]S2: Health and safety [42]S3: Maintenance cost [7,49]S4: Esthetics [50]S5: Use of local material [44]S6: Initial-acquisition cost [51]S7: Labor availability [52,21]

T1: Maintainability [53,42]T2: Ease of Construction (buildability) [52,54]T3: Resistance to decay [55]T4: Fire resistance [42,56]T5: Life expectancy of material(e.g. strength, durability etc.)[7,57]T6: Energy saving and thermalinsulation [58]

1.63.73.38.63.1

117P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125scale of 15,where 1 is least important, 2 fairly important, 3 important,4 very important, and 5 extremely important. Respondents were alsoencouraged to provide supplementary criteria that they consider to inu-encebuildingmaterial selectionbutwerenot listed in the survey. Relative

Table 2Rank of sustainable assessment criteria for building material selection.

Sustainable performance criteria Valid percentage of score of (%)

1 2 3 4 5

Environmental criteriaE7: Environmental statutory compliance 4.4 1.1 13.2 29.7 5E8: Minimize pollution 1.1 1.1 18.0 46.1 3E6: Zero/low toxicity 3.3 2.2 22.2 38.9 3E4: Ozone depletion potential 3.3 8.8 19.8 39.6 2E1: Recyclable/reusable material 1.1 7.7 29.7 38.5 2

E9: Amount of likely wastage in use 3.3 7.7 29.7 39.6 19.8E11: Embodied energy in material 1.1 9.9 28.6 47.3 13.2E2: Environmental sound disposal options 1.1 10.1 36.0 34.8 18.0E3: Impact on air quality 4.4 8.8 35.2 39.6 12.1E5: Impact during harvest 4.4 15.4 31.9 37.4 11.0E10: Methods of extraction of raw materials 5.5 19.8 45.1 20.9 8.8

Technical criteriaT1: Maintainability 0.0 0.0 3.3 47.3 49.5T6: Energy saving and thermal insulation 0.0 0.0 3.2 50.4 46.2T5: Life expectancy (e.g. durability) 0.0 0.0 4.4 50.5 45.1T4: Fire resistance 0.0 0.0 13.2 44.0 42.9T3: Ease of construction/buildability 0.0 0.0 9.9 53.8 36.3T2: Resistance to decay 1.1 1.1 28.6 48.4 20.9

Socio-economic criteriaS4: Esthetics 0.0 0.0 10.1 30.3 59.6S3: Maintenance cost 0.0 0.0 12.1 56.0 31.9S2: Health and safety 1.1 3.4 15.9 40.9 38.6S6: First cost 0.0 5.5 14.3 49.5 30.8S1: Disposal cost 1.1 0.0 22.0 47.3 29.7S5: Use of local materials 3.3 5.5 23.1 48.4 19.8S7: Labor availability 5.5 16.5 39.6 29.7 8.8(0.4RIb0.6), MediumLow (ML) (0.2RIb0.4), and Low (L)(0RIb0.2). A cut-off value of 0.4 is used and identied those criteriaas relevant for which the values are greater than or equal to 0.4. Fromthe results in Table 2, an interesting observation is that none of the

Relative index Ranking by category Overall ranking Importance level

0.846 1 7 H0.820 2 10 H0.793 3 13 MH0.763 4 15 MH0.749 5 17 MH

0.729 6 18 MH0.723 7 19 MH0.717 8 20 MH0.692 9 21 MH0.670 10 22 MH0.615 11 24 MH

0.892 1 2 H0.886 2 3 H0.881 3 4 H0.859 4 5 H0.853 5 6 H0.774 6 14 MH

0.898 1 1 H0.839 2 8 H0.825 3 9 H0.810 4 11 H0.808 5 12 H0.752 6 16 MH0.639 7 23 MH

with rotated factor loading matrix, the percentage of variance attrib-utable to each factor and the cumulative variance values are shown inTable 5. From the table, it can be seen that the three factors accountedfor 71.3% of the total variance of the eleven environmental criteria.

Following the results of the survey, the 24 criteria identied asbeing important components of sustainable material selection are

Table 3Factor loadings for socio-economic criteria after varimax rotation.

Observed socio-economic variable Latent socio-economic factors

Life cycle cost Social benet

S3: Maintenance cost 0.757S6: First cost 0.693S1: Disposal cost 0.576S4: Esthetics 0.830S5: Use of local material 0.759S2: Health and safety 0.579S7: Labor availability 0.556Eigenvalues 1.556 2.205Percentage of variance (%) 22.234 31.502

Table 5Factor loadings for environmental criteria after varimax rotation.

Observed environmental variable Latent environmental factors

Environmentalimpact

Resourceefciency

Wasteminimization

E7: Environmental statutorycompliance

0.882

E6: Zero or low toxicity 0.824E4: Ozone depletion potential 0.719E8 : Minimize pollution(e.g. water, land)

0.586

E3: Impact of material on air quality 0.557E10: Method of rawmaterial extraction

0.893

E9: Amount of likely wastage in useof material

0.773

E11: Embodied energywithin material

0.588

E5: Environmental Impactduring material harvest

0.546

E2: Availability of environmentally 0.912

Table 6Summary of roong options for the proposed project.

Description Option A Option B Option C

118 P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125criteria fall under the medium and other lower importance level. Thisclearly shows how important the criteria are to building designers inevaluating sustainable building materials. All criteria were rated withHigh or HighMedium importance levels, and all were used in theassessment process.

3.2. Factor analysis

Factor analysis was employed to analyze the structure of interrela-tionships among the criteria. Although the most signicant criteriawere identied using ranking analysis, some of them are likely to beinter-related with each other through an underlying structure of pri-mary factors. Factor analysis was used to obtain a concise list of SACs.It is conducted through a two-stage process: factor extraction and factorrotation. Before the factor analysis, validity test for factors is conductedaccording to themethod by Kaiser [63]. By KaiserMethod, a value calledeigenvalue under 1 is perceived as being inadequate and therefore un-acceptable for factor analysis. For the socio-economic criteria, the anal-ysis results showed that the KaiserMeyerOlkin [KMO] measure ofsampling adequacy was 0.606, larger than 0.5, suggesting that the sam-plewas acceptable for factor analysis. The Bartlett Test of Sphericitywas96.100 and the associated signicance level was 0.000, indicating thatthe population correlation matrix was not an identity matrix. Both ofthe tests showed that the obtained data in socio-economic categorysupported the use of factor analysis and that these could be groupedinto a smaller set of underlying factors. Using principal component anal-ysis, the factor analysis extracted two latent factors with eigenvaluesgreater than 1.0 for the 7 socio-economic criteria, explaining 53.7% ofthe variance. The rotated factor loadingmatrix based on the varimax ro-tation for the two latent factors is shown in Table 3.

The component matrix identies the relationship between the ob-served variables and the latent factors. The relationships are referredto as factor loadings. The higher the absolute value of the loading, themore the latent factor contributes to the observed variable. Small fac-tor loadings with absolute values less than 0.5 were suppressed to

Cumulative of variance (%) 22.234 53.736help simplify Table 3. For further interpretation, the two latent factorsunder the socio-economic category (shown in Table 3) are givennames as: Factor 1: life cycle cost; and Factor 2: social benet. Similar

Table 4Factor loadings for technical criteria after varimax rotation.

Observed technical variable Latent technical factors

Performance capability

T4: Fire resistance 0.799T3: Resistance to decay 0.740T6: Energy saving and thermal insulation 0.724T5: Life expectancy of material 0.712T2: Ease of construction 0.658T1: Maintainability 0.604Eigenvalues 3.016Percentage of variance (%) 50.264factor analyses were performed to identify the underlying structuresfor technical and environmental categories. In the technical category,the results for the factor analysis showed that the KMO measure was0.804 and the Bartlett's test (p=0.000) was also signicant, whichindicated that the factor analysis was also appropriate in identifyingthe underlying structure of the technical category. The results of theanalysis are presented in Table 4 Just one factor named Factor 6: per-formance capability was extracted, explaining 50.3% of the total vari-ance of the six technical criteria.

For Environmental category, both the KMO measure of samplingadequacy test (0.801) and Bartlett's sphericity (p=0.000) were sig-nicant, which indicated that factor analysis was also appropriate.Three factors under environmental category were extracted fromthe factor analysis, namely, Factor 3: environmental impact; Factor4: Resource efciency; and Factor 5: waste minimization. Along

sound disposal optionsE1: Potential for recycling and reuse 0.871Eigenvalues 5.505 1.216 1.116Percentage of variance (%) 50.048 11.057 10.149Cumulative of variance (%) 50.048 61.105 71.254Elementtype

Pitched Roof TimberConstruction

Pitched Roof TimberConstruction

Pitched Roof TimberConstruction

Buildingtype

Residential Residential Residential

Element Timber trussedrafters and joists withinsulation, roongunderlay, counterbattens, battens andUK produced con-crete interlockingtiles

Structurally insulatedtimber panel systemwith OSB/3 each side,roong underlay,counter battens,battens and UK pro-duced reclaimed claytiles

Structurallyinsulated timberpanel system withplywood (temperateEN 6362) deckingeach side, roongunderlay, counterbattens, battens andUK produced Fibercement slates

Size of tileor slate

420 mm330 mm 420 mm330 mm 420 mm330 mm

Pitch ofroof

22.5 22.5 22.5

analyzed and ranked according to expert's opinions. The 24 criteriawere further compressed into 6 assessment criteria factors of envi-ronmental impact, resource efciency, waste minimization, life cyclecost, performance capability and social benet for easy evaluation.Since these criteria are derived from the survey through expertopinion, they symbolize the sustainable criteria that promote socio-economic, technical and environmental consideration in building ma-terial assessment and selection. Consideration of these six criteria inmaterial selection will ensure sustainable development in buildingdesign and construction. In the next section, the mathematicalmulti-criteria decision-making model used for this study is brieypresented.

4. Implementation of the FEAHP selection model

The worked example for elucidating the application of the modelin practice involves the application to a realistic scenario of a buildingmaterial selection problem. The case study used intends to provide anindication of the use of the FEAHP multi-criteria decision-makingmodel for the problem analyzed (i.e., the selection of sustainablebuilding materials). The case study involves the design of a singlefamily home located in a light residential area of Wolverhampton,United Kingdom. An architectural rm is working with a client to se-lect materials (in this case roong elements) for a residential build-ing. The rm wants to take into account all the possible importantcriteria which determine sustainability of roong elements. A

decision-making group is formed which consists of ve constructionexperts from each strategic decision area. Detailed discussion onevery criteria, sub-criteria and alternative materials has beenconducted and based on expert rating and opinion, six sustainable as-sessment criteria have been identied. The discussion has been fur-ther prolonged to decide the twenty four sub-criteria with threepotential roong elements.

Table 6 summarizes the details for the three options of roong el-ements for the proposed project. The description of the three optionsis based on the standard practices and construction details commonlyused in the United Kingdom.

These 3 roong elements described above will be analyzed for theselection of sustainable option among alternatives. In other words,this section will analyze, through the use of the mathematicalmulti-criteria decision-making model described in Section 2, whichone is the most sustainable roong material for this scenario.

4.1. Fuzzy AHP procedure for the sustainable building materialsselection problem

In sustainable material selection problem, the relative importanceof different decision criteria involves a high degree of subjective judg-ment and individual preferences. The linguistic assessment of humanfeelings and judgments are vague and it is not reasonable to representthem in terms of precise numbers. It feels more condent to give in-terval judgments. Therefore triangular fuzzy numbers were used in

SELECTING SUSTAINABLE

MATERIAL

Environmental impact

Life cycle cost Resource efficiency

Waste minimization

Performance capability

Social benefit

Environmental statutory

Initial cost Method of raw material

ex

Amwas

Em

Envimp

h

Environmental sound disposal

Fire resistance Use of local material

Goal

Criteria

119P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125compliance

Zero /low toxicity

Ozone depletion

Minimize pollution

Impact on air quality

Maintenance cost

Disposal costSub criteria

Alternatives Option A Fig. 4. Hierarchy of thetraction

ount of tage in use

bodied energy

ironmental act during arvest

option

Recycling and Reuse

Resistance to decay

Energy saving and thermal insulation

Life expectancy

Ease of construction

Maintainability

Aesthetics

Health and safety

Material availability

Option B Option Cdecision problem.

this problem to decide the priority of one decision variable over an-other. The triangular fuzzy numbers were determined from reviewingliterature [33]. Then synthetic extent analysis method was used to de-cide the nal priority weights based on triangular fuzzy numbers andso-called as fuzzy extended AHP. In the following sections, the mainsteps of the method will be explained in detail.

Step 1. Dene the main criteria and sub-criteria for material selectionto design the fuzzy analytical hierarchy process tree structure.

First the overall objective of the material selection problem has

0:28 0:41 0:33 0:23 V F2F3 1;V F23F4 1;V F2F6 1:

efciency Waste minimization Performance capability Social benet Wo

4 1/6,1/5,1/4 1/8,1/7,1/6 1/6,1/5,1/4 0.1552 1,1,2 1/4,1/3,1/2 1/4,1/3,1/2 0.310

2,3,4 1/4,1/3,1/2 1,1,2 0.0792 1,1,1 1/6,1/5,1/4 1/4,1/3,1/2 0.058

4,5,6 1,1,1 1,1,2 0.3542,3,4 1/2,1,1 1 0.044

Table 7The linguistic variables and their corresponding fuzzy numbers.

Intensity ofimportance

Fuzzynumber

Denition Membershipfunction

9 ~9 Extremely more importance (EMI) (8, 9,10)7 ~7 Very strong importance (VSI) (6, 7, 8)5 ~5 Strong importance (SI) (4, 5, 6)3 ~3 Moderate importance (MI) (2, 3, 4)1 ~1 Equal importance (EI) (1, 1, 2)

120 P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125been identied which was selection of sustainable materials forbuilding project. In selecting sustainable materials, a lot of criteriashould be taken into account. All of the possible important criteriawhich could affect the sustainability of building materials have beendiscussed with experts in the Construction sector. Also other materialselection studies in the literature were reviewed with the expert. Bycombining the criteria determined by the expert and the criteriaused in the literature, the main criteria and the sub-criteria in thestudy were determined and validated.

After the main criteria, sub-criteria and alternatives were deter-mined, the hierarchy of the material selection problem was struc-tured. Fig. 4 shows the structuring of the material selection problemhierarchy of four levels. The top level of the hierarchy representsthe ultimate goal of the problem which is to choose a sustainableroong element among options for the project described previously.The goal is placed at the top of the hierarchy. The hierarchy descendsfrom the more general criteria in the second level to sub-criteria inthe third level to the alternatives at the bottom or fourth level. Thegeneral criteria level involved six major criteria: environmental im-pact, life cycle cost, resource efciency, waste minimization, perfor-mance capability and social benet. The decision-making teamconsidered three roong elements for the decision alternatives, andlocated them on the bottom level of the hierarchy.

Step 2. Calculate the weights of the main attributes, sub-attributesand alternatives.

After the construction of the hierarchy, the different priorityweights of each main criteria, sub-criteria and alternatives were cal-culated using the fuzzy extended AHP approach. The comparison ofthe importance of one main criteria, sub-criteria and alternativeover another were achieved by the help of the questionnaire. Thequestionnaires facilitate the answering of pair-wise comparison ques-tions. The preference of one measure over another was decided by theavailable research and the experience of the experts.

First the expert compared themain criteria with respect to themaingoal; then the expert compared the sub-criteria with respect to themain criteria. At the end, the expert compared the roong material op-tions with respect to each sub-criteria. The expert used the linguisticvariables to make the pair-wise comparisons. Then the linguistic vari-ables were converted to triangular fuzzy numbers. Table 7 shows thelinguistic variables and their corresponding triangular fuzzy numbers.Owing to the limited space, the pair-wise comparison matrix of themain criteria with respect to the goal will be presented here.

After the pair-wise comparison matrices were formed, the consis-tency of the pair-wise judgment of each comparison matrix waschecked using the calculation method of consistency index and

Table 8The Fuzzy evaluation of criteria with respect to the overall objective.

Main criteria Environmental impact Life cycle cost Resource

C1: Environmental impact 1,1,1 1/6,1/5,1/4 1/6,1/5,1/C2: Life cycle cost 4,5,6 1,1,1 1/4,1/3,1/C3: Resource efciency 4,5,6 2,3,4 1,1,1C4: Waste minimization 4,5,6 1/2,1,1 1/4,1/3,1/C5: Performance capability 6,7,8 2,3,4 2,3,4C6: Social benet 4,5,6 2,3,4 1/2,1,1consistency ratios in crisp AHP discussed in [64]. Each triangular fuzzynumber, M=(l, m, u) in the pair-wise comparison matrix wasconverted to a crisp number using M_crisp=(4m+ l+u)/6. Afterthe fuzzy comparison matrices were converted into crisp matrices, theconsistency of each matrix was checked by the method in crisp AHP[64]. After calculation, the consistency ratio of each comparison matrixwas found to be under 0.10. So we can conclude that the consistency ofthe pair-wise judgments in all matrices is acceptable. Then the priorityweights of eachmain criteria, sub-criteria and alternativewere calculat-ed using FAHP method. As an example the calculation of the priorityweights of the main criteria will be explained in detail below.

By using the values in Table 7, the linguistic variables in a compar-ison matrix can be converted to triangular fuzzy numbers. The fuzzyevaluation matrix with respect to the goal with triangular fuzzy num-bers can be seen in Table 8. In order to nd the priority weights of themain criteria, rst the fuzzy synthetic extent values of the attributeswere calculated by using Eq. (4). The different values of fuzzy syn-thetic extent with respect to the six different criteria are denoted byF1, F2, F3, F4, F5, and F6, respectively.

F1 9;11;13 1=39:62;1=33:32;1=27:52 0:23;0:33;0:47 ;

F2 7:9;9:5;11:17 1=39:62;1=33:32;1=27:52 0:20;0:28;0:41 ;

F3 4:8;6;7:34 1=39:62;1=33:32;1=27:52 0:12;0:18;0:27 ;

F4 3:46;4:16;4:97 1=39:62;1=33:32;1=27:52 0:09;0:12;0:18 ;

F5 2:36;2:66;3:14 1=39:62;1=33:32;1=27:52 0:06;0:08;0:11

F6 2:34;3;4 1=39:62;1=33:32;1=27:52 0:05;0:10;0:15 :

The degree of possibility of Fi over Fj (I j) can be determined byEqs. (8) and (9).

V F1F2 1;V F1F3 1;V F1F4 1V F1F5 1;V F1F6 1

V F2F1 0:23 0:41 0:78

Similarly,

V F3F1 0:21;V F3F2 0:41;V F3F4 1;V F3F5 1;V F3F6 1;

V F4F1 0:31;V F4F2 0:14;V F4F3 0:5;V F4F5 1;V F4F6 1;

V F5F1 0:92;V F5F2 0:82;V F5F3 0:11;V F5F4 0:33;V F5F6 1

V F6F1 0:82;V F6F2 0:71;V F6F3 0:34;V F6F4 0:44;V F6F5 0:31:

With the help of Eq. (11), the minimum degree of possibility canbe stated as below:

M C1 minV F1F2; F3; F4; F5; F6 min 1;1;1;1;1 1:

Similarly,m (C2)=0.78,m (C3)=0.21,m (C4)=0.14,m (C5)=0.11,and m (C6)=0.44.

Therefore theweight vector is given asWc=(1, 0.78, 0.21.0.14,0.11,0.44)T and after the normalization process, the weight vector with re-spect to decision criteria C1, C2, C3, C4, C5, and C6 can be expressed asfollows:Wo=(0.155 environmental impact, 0.310 life cycle cost, 0.079

resource efciency, 0.058wasteminimization, 0.354 performance capability, 0.044 social benet)T. We can conclude that the most importancriteria in the sustainable material selection process for the projecunder consideration is Performance capability because it has thehighest priority weight. Life cycle cost is the next preferred criteria.

The complete result is shown in Table 8.Now the different sub-criteria are compared under each of the cri

terion separately by following the same procedure as discussedabove. Whenever the value of (n11n23)>0, the elements of the matrix must be normalized and then do the same process to nd theweight vector of each criteria. The fuzzy comparison matrices of thesub-criteria and the weight vectors of each sub-criterion are shownin Table 9.

Similarly the fuzzy evaluation matrices of decision alternativeand corresponding weight vector of each alternative with respect tocorresponding sub-criteria are determined. The priority weights omaterials with respect to each criterion are given by adding theweights per materials multiplied by weights of the correspondingcriteria. The priority weights of each main criteria, sub-criteria, andalternative can be found in Table 10.

Step 3. Synthesizing the results.After computing the normalized priority weights for each pair

wise comparison judgment matrices (PCJM) of the AHP hierarchythe next phase is to synthesize the rating for each criterion. The nor

Table 9Fuzzy evaluation of the attributes with respect to sub-criteria.

Environmental impact Environmental statutory compliance Zero/low toxicity Ozone depletion Minimize pollution Impact on air quality Priority vector

Environmental statutory compliance 1,1,1 3/2,2,5/2 3/2,2,5/2 5/2,3,7/2 5/2,3,7/2 0.517Zero/low toxicity 2/5,1/2,2/3 1,1,1 3/2,2,5/2 5/2,3,7/2 5/2,3,7/2 0.137Ozone depletion 2/5,1/2,2/3 2/5,1/2,2/3 1,1,1 3/2,2,5/2 3/2,2,5/2 0.219Minimize pollution 2/7,1/3,2/5 2/7,1/3,2/5 2/5,1/2,2/3 1,1,1 3/2,2,5/2 0.68Impact on air quality 2/7,1/3,2/5 2/7,1/3,2/5 2/5,1/2,2/3 2/5,1/2,2/3 1,1,1 0.60C.I.=0.01, R.I.=1.12, C.R.=0.009

Life cycle cost Purchase cost Disposal cost Maintenance cost Priority vector

Purchase cost 1,1,1 1,1,2 4,5,6 0.69Disposal cost 1/2,1,1 1,1,1 2,3,4 0.22Maintenance cost 1/6,1/5,1/4 1/4,1/3,1/2 1,1,1 0.09C.I.=0.02, R.I.=0.58, C.R.=0.03

Lifeexp

5/23/23/21,12/3

2/5

121P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125Resource efciency Embodied energy Amount of wastage

Embodied energy 1,1,1 1,1,2Amount of wastage 1/2,1,1 1,1,1Method of extraction 1/2,1,1 1/2,1,1Impact during harvest 1/2,1,1 1/2,1,1C.I.=0.025, R.I.=0.90, C.R.=0.028

Waste minimization Recycling and reuse

Recycling and reuse 1,1,1Environmentally sound disposal 1/4,1/3,1/2C.I.=0.00, R.I.=0.00, C.R.=0.00

Performancecapability

Fireresistance

Maintainability Resistance todecay

Fire resistance 1,1,1 3/2,2,5/2 2/3,1,3/2Maintainability 2/5,1/2,2/3 1,1,1 2/3,1,3/2Resistance to decay 2/3,1,3/2 2/3,1,3/2 1,1,1Life expectancy 2/7,1/3,2/5 2/5,1/2,2/3 2/5,1/2,2/3Energy saving andthermal insulation

2/5,1/2,2/3 3/2,2,5/2 3/2,2,5/2

Buildability 2/5,1/2,2/3 2/5,1/2,2/3 3/2,2,5/2C.I.=0.00821, R.I.=1.24, C.R.=0.007

Social benet Local material Esthetics

Local material 1,1,1 2/3,1,3/2Esthetics 2/3,1,3/2 1,1,1Health and safety 2/3,1,3/2 2/5,1/2,2/3Material availability 2/5,1/2,2/3 2/5,1/2,2/3C.I.=0.025, R.I.=0.90, C.R.=0.03Method of extraction Impact during harvest Priority vector

1,1,2 1,1,2 0.2891,1,2 1,1,2 0.4751,1,1 1/4,1/3,1/2 0.0812,3,4 1,1,1 0.155

Environmental sound disposal Priority vector

2,3,4 0.671,1,1 0.33

ectancyEnergy saving and thermalinsulation

Buildability Priorityvector

,3,7/2 3/2,2,5/2 2/3,1,3/2 0.183,2,5/2 2/7,1/3,2/5 3/2,2,5/2 0.258,2,5/2 2/7,1/3,2/5 2/5,1/2,2/3 0.47,1 2/3,1,3/2 7/2,4,9/2 0.204,1,3/2 1,1,1 2/7,1/3,2/5 0.212

,1/2,2/3 2/5,1/2,2/3 1,1,1 0.095

Health and Safety Material availability Priority vector

2/3,1,3/2 2/3,1,3/2 0.1933/2,2,5/2 3/2,2,5/2 0.3681,1,1 3/2,2,5/2 0.3682/5,1/2,2/3 1,1,1 0.070-tt

-

-

s

f

-,-

Table 10Priority weights for sustainability criteria and sub criteria used in the case study.

Variables in level 1 Local weight (1)a Variables in level 2 Local weight (2) Global weight (3)b Variables in level 3 Local weight (4)

Environmental impact 0.155 Environmental statutory compliance 0.517 0.07962 Material AMaterial BMaterial C

0.0720.6500.278

Zero/low toxicity 0.137 0.02109c Material AMaterial BMaterial C

0.2000.4000.400

Ozone depletion 0.219 0.03373 Material AMaterial BMaterial C

0.7470.0600.193

Minimize pollution 0.68 0.10472 Material AMaterial BMaterial C

0.6740.1010.226

Impact on air quality 0.60 0.0924 Material AMaterial BMaterial C

0.7960.1250.079

Life cycle cost 0.310 Maintenance cost 0.22 0.07062 Material AMaterial BMaterial C

0.6910.2180.091

Initial cost 0.69 0.22149 Material AMaterial BMaterial C

0.7700.0680.162

Disposal cost 0.09 0.02889 Material AMaterial BMaterial C

0.7310.0810.188

Resource efciency 0.079 Method of raw material extraction 0.081 0.00624 Material AMaterial BMaterial C

0.4000.2000.400

Amount of wastage 0.475 0.03658 Material AMaterial BMaterial C

0.1260.4160.458

Embodied energy 0.289 0.02225 Material AMaterial BMaterial C

0.6910.0910.218

Environmental impact during harvest 0.155 0.01194 Material AMaterial BMaterial C

0.7540.1810.065

Performance capability 0.354 Fire resistance 0.183 0.06405 Material AMaterial BMaterial C

0.8040.0740.122

Resistance to decay 0.47 0.1645 Material AMaterial BMaterial C

0.4720.0840.444

Energy saving and thermal insulation 0.212 0.0742 Material AMaterial BMaterial C

0.8020.0750.211

Life expectancy 0.204 0.0714 Material AMaterial BMaterial C

0.1840.5840.232

Ease of construction 0.095 0.03325 Material AMaterial BMaterial C

0.6910.0910.218

Maintainability 0.258 0.0903 Material AMaterial BMaterial C

0.450.090.46

Social benet 0.044 Use of Local material 0.193 0.00791 Material AMaterial BMaterial C

0.3000.6000.600

Esthetics 0.368 0.01508 Material AMaterial BMaterial C

0.0820.7000.378

Health and safety 0.368 0.01508 Material AMaterial BMaterial C

0.7700.0580.162

Material availability 0.070 0.00287 Material AMaterial BMaterial C

0.7700.0500.170

Waste minimization 0.058 Environmental sound disposal option 0.33 0.01881 Material AMaterial BMaterial C

0.690.090.22

Recycling and reuse 0.67 0.03819 Material AMaterial BMaterial C

0.450.090.46

1.000 1.000a Local weight is derived from judgment with respect to a single criterion.b Global weight is derived from multiplication by the priority of the criterion.c This entry is obtained as follows: 0.1540.137=0.02109. The global weight of the sub-criterion is obtained by multiplying the local weight of the sub-criterion by the weight of

the criterion.

122 P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125

st

n

n

123P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125malized local priority weights of dimensions of sustainability and var-ious SC were obtained and were combined together in order to obtainthe global composite priority weights of all SC used in the third levelof the AHP model. In order to shorten the solution process for the ma-terial selection problem, Expert Choice 11.5 was used to determinethe global priority weights of material alternatives based on the

Table 11Overall rating of the three assessed roong material using AHP technique.

Criterion Local weight (1) Sub-criterion

Environmental impact 0.155 Environmental statutory complianceZero/low toxicityOzone depletionMinimize pollutionImpact on air quality

Life cycle cost 0.310 Maintenance costInitial costDisposal cost

Resource efciency 0.079 Method of raw material extractionAmount of wastageEmbodied energyEnvironmental impact during harve

Performance capability 0.354 Fire resistanceResistance to decayEnergy saving and thermal insulatioLife expectancyEase of constructionMaintainability

Social benet 0.044 Use of local materialEstheticsHealth and safetyMaterial availability

Waste minimization 0.058 Environmental sound disposal optioRecycling and reuse

Total 1.000questionnaire used to facilitate comparisons of main attributes, sub-attributes and alternatives. It provides two ways of synthesizing thelocal priorities of the alternatives using the global priorities of theirparent criteria: the distributive mode and the ideal mode. In the dis-tributive mode the weight of criteria reects the importance that thedecision maker attaches to the dominance of each alternative relativeto all other alternatives under that criterion. In this case, the distribu-tive mode would be the way to synthesize the results. After derivingthe local priorities for the criteria and the alternatives throughpair-wise comparisons, the priorities of the criteria are synthesizedto calculate the overall priorities for the decision alternatives. Asshown in Table 11, the materials are ranked according to their overallpriorities. Material option (A) turns out to be the most preferablematerial among the three materials, with an overall priority scoreof 0.453.

5. Conclusion

This paper discussed the development of assessment criteria,computational methods, and analytical models for sustainable mate-rial selection. The complex tasks of comparing building materials op-tions based on their sustainability, using multi-criteria considerationsare regarded as daunting challenges by many material speciers.Hence developing suitable systematic approaches and appropriatestructured decision-making frameworks for sustainable building ma-terial selection was considered in this research. Decision making for asustainable material alternative, while considering various criteriathat inuence selection, is difcult and this difculty is further com-plicated not only when conicting relationships exist between thecriteria considered, but also when qualitative criteria are included.To deal with this difculty effectively, this research proposed a meth-od that focused on aggregation of criteria that inuence sustainablematerial selection.

Sustainable assessment criteria (SAC) used in study were identi-ed based on sustainability triple bottom (TBL) approach, review ofthe literature in the eld of material selection, combined with re-

Local weight (2) Local weight (3) Global weight (4)

M (A) M (B) MI M (A) M (B) MI

0.517 0.072 0.650 0.278 0.0057 0.0504 0.07250.137 0.200 0.400 0.400 0.0042 0.0084 0.00840.219 0.747 0.060 0.193 0.0252 0.0020 0.00650.68 0.674 0.101 0.226 0.0306 0.0106 0.02370.60 0.796 0.125 0.079 0.0736 0.0116 0.00730.22 0.691 0.218 0.091 0.0088 0.0054 0.00640.69 0.770 0.068 0.162 0.0705 0.0151 0.03590.09 0.731 0.081 0.188 0.0216 0.0023 0.00540.081 0.400 0.200 0.400 0.0025 0.0012 0.00250.475 0.126 0.416 0.458 0.0046 0.0152 0.01680.289 0.691 0.091 0.218 0.0154 0.0020 0.00490.155 0.754 0.181 0.065 0.0090 0.0027 0.00080.183 0.804 0.074 0.122 0.0515 0.0047 0.00780.47 0.472 0.084 0.444 0.0376 0.0138 0.03300.212 0.802 0.075 0.211 0.0295 0.0056 0.01570.204 0.184 0.584 0.232 0.0131 0.0417 0.01660.095 0.691 0.091 0.218 0.0229 0.0030 0.00720.258 0.45 0.90 0.46 0.0206 0.0313 0.02150.193 0.300 0.600 0.600 0.0024 0.0048 0.00480.368 0.082 0.700 0.378 0.0012 0.0106 0.00570.368 0.770 0.058 0.162 0.0116 0.0009 0.00240.070 0.770 0.050 0.170 0.0022 0.0001 0.00050.33 0.69 0.09 0.22 0.0130 0.0017 0.00410.67 0.45 0.09 0.46 0.0172 0.0034 0.0176Overall priority 0.453 0.249 0.324Rank 1 3 2quirement of project stakeholders. A questionnaire survey wasemployed to obtain the perceived importance of the criteria. Follow-ing the results of the survey, the twenty four criteria identied asbeing important components of sustainable material selection werefurther compressed into six assessment criteria factors of environ-mental impact, resource efciency, waste minimization, life cycle cost,performance capability and social benet for modeling and evaluat-ing sustainable performance of building materials. A Fuzzy extendedanalytical hierarchical process (FEAHP) was used for aggregatingand assigning numerical values (i.e., weights) to measure the relativeimportance of these criteria for a given material alternative. For thispurpose, FEAHP uses simple pairwise comparisons to determineweights and ratings so that the analyst can concentrate on just twofactors at one time. This process enables decision makers to formalizeand effectively solve the complicated, multi-criteria and vague per-ception problem of building material alternative selection. An empir-ical case study of three proposed roong element alternatives for anew building project was used to exemplify the approach. From theresult of the case study, it can be concluded that the application ofthe FEAHP in incorporating both objective and subjective criteriainto the assessment process and improving the team decision processis desirable. The underlying concepts applied were intelligible to thedecision maker groups, and the computation required is straightfor-ward and simple. A comparison of the result of the traditional deci-sion method and that of the Fuzzy extended AHP method clearlyshows that qualitative criteria have a signicant impact on sustain-ability of building materials. In the case study project, it was discov-ered that some materials selected by the traditional methods weredropped when the qualitative criteria were introduced using AHP.However, Fuzzy AHP is a highly complex methodology and requires

124 P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125more numerical calculations in assessing composite priorities thanthe traditional method and hence it increases the effort. The fuzzymethodology could also be extended with the other multi-criteriadecision-making (MCDM) methods such as Analytical NetworkProcess (ANP), TOPSIS, ELECTRE and DEA techniques in solving mate-rial selection problems. Finally, the evaluation model and results canprovide a valuable reference for building professionals seeking to en-hance the sustainability of construction projects.

The approach suggested in this research can allow for assessmentof the other civil infrastructures taking into consideration a large cri-terion set and an extensive number of alternatives, especially whereconicting objectives exist. The authors are convinced that this re-search can assist building stakeholders in making critical decisionsduring the selection of sustainable material alternatives.

References

[1] S. Halliday, Sustainable Construction Butterworth Heinemann. (London) , 2008.[2] P.S. Barrett, M.G. Sexton, L. Green, Integrated delivery systems for sustainable

construction, Building Research and Information 27 (6) (1999) 397404.[3] N.Z. Abidin, Investigating the awareness and application of sustainable construc-

tion concept by Malaysian developers, Habitat International 34 (4) (2010)421426.

[4] L. Braganca, R. Mateus, H. Koukkari, Assessment of building sustainability, in:L. Braganca, et al., (Eds.), Sustainability of Constructions, Integrated Approach toLife-time Structural Engineering, Proceedings of the rst COST C25 Workshop,Lisbon, Selected papersMulticomp, Lisbon, Sept 1315 2007, pp. 312.

[5] J. Godfaurd, D. Clements-Croome, G. Jeronimidis, Sustainable building solutions: areview of lessons from the natural world, Building and Environment 40 (3)(2005) 319328.

[6] K. Nassar, W. Thabet, Y. Beliveau, A procedure for multi-criteria selection of build-ing assemblies, Automation in Construction 12 (5) (2003) 543560.

[7] J.K.W. Wong, H. LI, Application of the analytic hierarchy process (AHP) inmulti-criteria analysis of the selection of intelligent building systems, Buildingand Environment 43 (1) (2008) 108125.

[8] R. Gibson, S. Hassan, S. Holtz, J. Tansey, G. Whitelaw, Sustainability AssessmentCriteria, Processes and Application, Earthscan Publications Limited, London, UK,2005.

[9] D. Crawley, I. Aho, Building environmental assessment methods: environmentalperformance or sustainability? Building Research and Information 27 (4/5)(1999) 300308.

[10] M. Sinou, S. Kyvelou, Present and future of building performance assessmenttools, Management of Environmental Quality: An International Journal 17 (5)(2006) 570586.

[11] O. Ortiz, J.C. Pasqualino, F. Castells, Environmental performance of constructionwaste: comparing three scenarios from a case study in Catalonia, Spain, WasteManagement 30 (2010) 646654.

[12] P.O. Akadiri, P.O. Olomolaiye, Development of sustainable assessment criteria forbuilding materials selection, Journal of Engineering Construction and Architectur-al Management 19 (6) (2012) 1837.

[13] A. Shapira, M. Goldenberg, AHP-based equipment selection model for construc-tion projects, Journal of Construction Engineering and Management 131 (12)(2005) 12631273.

[14] T.J. Barker, Z.B. Zabinsky, A multi-criteria decision making model for reverse lo-gistics using analytical hierarchy process, Omega 39 (5) (2011) 558573.

[15] E.K. Zavadskas, A. Kaklauskas, Z. Turskis, J. Tamosaitiene, Selection of the effectivedwelling house walls by applying attributes values determined at intervals,Journal of Civil Engineering and Management 14 (2) (2008) 8593.

[16] H.Z. Alibaba, M.B. Ozdeniz, A building elements selection system for architects,Building and Environment 39 (3) (2004) 307316.

[17] T.L. Saaty, The Analytic Hierarchy Process, McGraw-Hill, New York, 1980.[18] E.K. Zavadskas, Z. Turskis, J. Tamosaitiene, Selection of construction enterprises

management strategy based on the SWOT and multi-criteria analysis, Archivesof Civil and Mechanical Engineering 4 (11) (2011) 10631082.

[19] R. Bahareh, R. Rehan, H. Kasun, Sustainability assessment of ooring systems inthe city of Tehran: an AHP-based life cycle analysis, Construction and BuildingMaterials 25 (4) (2011) 20532066.

[20] H.Wang,H. Bai, J. Liu, H. Xu,Measurement indicators and an evaluation approach forassessing strategic environmental assessment effectiveness, Ecological Indicators 23(2012) 413420.

[21] O.O. Ugwu, M.M. Kumaraswamy, A. Wong, S.T. Ng, Sustainability appraisal in in-frastructure projects (SUSAIP): Part 1. Development of indicators and computa-tional methods, Automation in Construction 15 (2) (2006) 239251.

[22] E.K. Zavadskas, J. Antucheviciene, Multiple criteria evaluation of rural building'sregeneration alternatives, Building and Environment 42 (1) (2007) 436451.

[23] M.M. Fouladgar, A. Yazdani-Chamzini, A. Lashgari, E.K. Zavadskas, Z. Turskis,Maintenance strategy selection using AHP and COPRAS under fuzzy environment,International Journal of Strategic Property Management 16 (1) (2012) 85104.

[24] R.K. Singh, H.R. Murty, S.K. Gupta, A.K. Dikshit, Development of composite sus-tainability performance index for steel industry, Ecological Indicators 7 (3)(2007) 565588.[25] T.L. Saaty, Relative measurement and its generalization in decision making whypairwise comparisons are central inmathematics for themeasurement of intangiblefactors, The Analytic Hierarchy/Network Process. RACSAM 102 (2) (2008) 251318.

[26] F. Zahedi, The analytic hierarchy process: a survey of the method and its applica-tions, Interfaces 16 (4) (1986) 96108.

[27] T.L. Saaty, J.S. Shang, An innovative orders-of-magnitude approach to AHP-basedmulti-criteria decision making: prioritizing divergent intangible humane acts,European Journal of Operational Research 214 (3) (2011) 703715.

[28] M. Medineckiene, E.K. Zavadskas, Z. Turskis, Dwelling selection by applying fuzzygame theory, Archives of Civil and Mechanical Engineering 11 (3) (2011)681697.

[29] P. Jaskowski, S. Biruk, R. Bucon, Assessing contractor selection criteria weightswith fuzzy AHP method application in group decision environment, Automationin Construction 19 (2) (2010) 120126.

[30] F.T.S. Chan, N. Kumar, Global supplier development considering risk factors usingfuzzy extended AHP-based approach, Omega 35 (4) (2007) 417431.

[31] A. Nieto-Morote, F. Ruz-Vila, A fuzzy multi-criteria decision-making model forconstruction contractor prequalication, Automation in Construction 25 (2012)819.

[32] L.A. Zadeh, Fuzzy Sets Information and Control 8 (1965) 338353.[33] P. Nang-Fei, Fuzzy AHP approach for selecting the suitable bridge construction

method, Automation in Construction 17 (8) (2008) 958965.[34] D.Y. Chang, Applications of the extent analysis method on Fuzzy-AHP, European

Journal of Operational Research 95 (1996) 649655.[35] C.K. Kwong, H. Bai, Determining the importance weights for the customer re-

quirements in QFD using a fuzzy AHP with an extent analysis approach, IIE Trans-actions 35 (7) (2003) 619626.

[36] T.J. Foxon, G. Mcilkenny, D. Gilmour, C. ltean-dumbrava, N. Souter, R. Ashley, D. Butler,P. Pearson, P. Jowitt, J. Moir, Sustainability criteria for decision support in theUKwaterindustry, Journal of Environmental Planning and Management 45 (2) (2002)285301.

[37] A.J. Balkema, H.A. Preisig, R. Otterpohl, A.J.D. Lambert, S.R. Weijers, Developing amodel based decision support tool for the identication of sustainable treatmentoptions for domestic wastewater, Water Science and Technology 43 (7) (2001)265269.

[38] T. Buchholz, V.A. Luzadis, T.A. Volk, Sustainability criteria for bioenergy systems:results from expert survey, Journal of Cleaner Production 17 (1) (2009) S86S98.

[39] Y. Chen, G.E. Okudan, D.R. Riley, Sustainable performance criteria for constructionmethod selection in concrete buildings, Automation in Construction 19 (2)(2010) 235244.

[40] P. Asokan, M. Osmani, A.D.F. Price, Assessing the recycling potential of glass brereinforced plastic waste in concrete and cement composites, Cleaner Production17 (9) (2009) 821829.

[41] S. Beder, Environmental Principles and Policies: An Interdisciplinary Approach,UNSW Press, Sydney and Earthscan, London, 2006.

[42] R. Spiegel, D. Meadows, A Guide to Product Selection and Specications GreenBuilding Materials, third ed. Canada: John Wiley and Sons, Inc., Hoboken, NewJersey, 2010.

[43] J. Kim, B. Rigdon, Sustainable Architecture Module: Qualities, Use, and Examples ofSustainable Building Materials,Sustainable Architecture Compendium, NationalPollution Prevention Center, University of Michigan, 1998.

[44] K.R. Bunz, G.P. Henze, D.K. Tiller, Survey of sustainable building design practices inNorth America, Europe and Asia, Architectural Engineering 12 (1) (2006) 3362.

[45] J. Anderson, D. Shiers, K. Steele, The Green Guide to Specication: An Environ-mental Proling System for Building Material and Components, BR 501, BuildingResearch Establishment, UK, 2009.

[46] L.C. Bank, B.P. Thompson, M. McCarthy, Decision-making tools for evaluating theimpact of materials selection on the carbon footprint of buildings, CarbonManagement 2 (4) (2011) 431441.

[47] T. Crosbie, N. Dawood, S. Dawood, Improving the energy performance of the builtenvironment: the potential of virtual collaborative life cycle tools, Automation inConstruction 20 (2) (2011) 205216.

[48] H. Yao, L. Shen, Y. Tan, J. Hao, Simulating the impacts of policy scenarios on thesustainability performance of infrastructure projects, Automation in Construction20 (8) (2011) 10601069.

[49] J. Wong, H. Li, J. Lai, Evaluating the system intelligence of the intelligent buildingsystems: Part 1: development of key intelligent indicators and conceptual analyt-ical framework, Automation in Construction 17 (3) (2008) 284302.

[50] M. Ashby, K. Johnson,Materials andDesign: theArt and Science ofMaterial Selectionin Product Design, Butterworth-Heinemann, Oxford, 2002.

[51] S. Emmitt, D.T. Yeomans, Specifying Buildings: a Design Management Perspective.(261 pp) 2nd ed. Elsevier, Amsterdam, 2008.

[52] M. Calkins, Materials for Sustainable Sites: a Complete Guide to the Evaluation, Se-lection, and Use of Sustainable Construction Materials, Wiley, Hoboken, N.J., 2009.

[53] P. Sirisalee, M.F. Ashby, G.T. Parks, P.J. Clarkson, Multi-criteria material selectionin engineering design, Advanced Engineering Materials 6 (12) (2004) 8492.

[54] C.J. Kibert, Sustainable Construction: Green Building Design and Delivery, seconded. John Wiley and Sons, Inc., Hoboken, New jersey, USA, 2008.

[55] P. Joseph, S. Tretsiakova-McNally, Sustainable non-metallic building materials,Sustainability 2 (2) (2010) 400427.

[56] S.M. Lo, C.M. Zhao, M. Liu, A. Coping, A simulation model for studying the imple-mentation of performance-based re safety design in buildings, Automation inConstruction 17 (7) (2008) 852863.

[57] C.C. Zhou, G.F. Yin, X.B. Hu, Multi-objective optimization of material selection forsustainable products: articial neural networks and genetic algorithm approach,Materials and Design 30 (4) (2009) 12091215.

[58] J. Goggins, T. Keane, A. Kelly, The assessment of embodied energy in typicalreinforced concrete building structures in Ireland, Energy and Buildings 42 (5)(2010) 735744.

[59] P.O. Olomolaiye, K.A. Wahab, A.D.F. Price, Problems inuencing craftsmen's pro-ductivity in Nigeria, Building and Environment 22 (4) (1987) 317323.

[60] E.A. Chinyio, P.O. Olomolaiye, P. Corbett, Quantication of construction clients'needs through paired comparisons, Journal of Management in Engineering 14(1) (1998) 8792.

[61] N. Braimah, I. Ndekugri, Consultants' perceptions on construction delay analysismethodologies, Journal of Construction Engineering and Management 135 (12)(2009) 12791288.

[62] P.O. Akadiri, Development of a multi-criteria approach for the selection of sustain-ablematerials for building projects. PhD thesis, SEBE, University ofWolverhampton,Wolverhampton, 2011.

[63] H.F. Kaiser, An index of factorial simplicity, Psychometrika 39 (1) (1974) 3136.[64] W. Ying-Ming, L. Ying, H. Zhongsheng, On the extent analysis method for fuzzy

AHP and its applications, European Journal of Operational Research 186 (2)(2008) 735747.

125P.O. Akadiri et al. / Automation in Construction 30 (2013) 113125

Multi-criteria evaluation model for the selection of sustainable materials for building projects1. Introduction2. Fuzzy analytic hierarchy process methodology2.1. Background2.2. Establishment of triangular fuzzy numbers2.3. Calculation of priority weights at different level of hierarchy

3. Development of sustainable assessment criteria3.1. Importance of derived criteria3.2. Factor analysis

4. Implementation of the FEAHP selection model4.1. Fuzzy AHP procedure for the sustainable building materials selection problem

5. ConclusionReferences