Risk assessment model selection in construction industry Expert Systems with Applications 38 (2011) 9105–9111 Contents lists available at ScienceDirect Expert Systems with Applications j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / e s w a Risk assessment model selection in construction industry AmirReza KarimiAzari a, Neda Mousavi b,⇑, S. Farid Mousavi c, SeyedBagher Hosseini a a Iran University of Science and Technology, Tehran, Iran b Science & Research Branch, Islamic Azad University, Tehran, Iran c Tarbiyat Modares University, Tehran, Iran a r t i c l e i n f o Keywords: Risk management Risk assessment model Construction industry Fuzzy TOPSIS 0957-4174/$ - see front matter � 2010 Elsevier Ltd. A doi:10.1016/j.eswa.2010.12.110 ⇑ Corresponding author. E-mail addresses: nmousavi2930@gmail.com (N. com (S.F. Mousavi). a b s t r a c t Construction industry faces a lot of inherent uncertainties and issues. As this industry is plagued by risk, risk management is an important part of the decision-making process of these companies. Risk assessment is the critical procedure of risk management. Despite many scholars and practitioners recognizing the risk assessment models in projects, insufficient attention has been paid by researchers to select the suitable risk assessment model. In general, many factors affect this problem which adheres to uncertain and imprecise data and usually several people are involved in the selection process. Using the fuzzy TOPSIS method, this study provides a rational and systematic process for developing the best model under each of the selection criteria. Decision criteria are obtained from the nominal group technique (NGT). The proposed method can discriminate successfully and clearly among risk assessment methods. The proposed approach is demonstrated using a real case involving an Iranian construction corporation. � 2010 Elsevier Ltd. All rights reserved. 1. Introduction The research on projects has expanded during the last dec- ades (Naaranoja, Haapalainen, & Lonka, 2007). A project is an organization of people dedicated to the deployment of a set of resources for a specific purpose or objective (Steiner, 1969). Pro- ject management is defined as planning, directing, and control- ling resources to achieve specific goals and objectives of the project (Fan, Lin, & Sheu, 2008). Managers need to ensure deliv- ery of projects to cost, schedule and performance requirement. To achieve this involves identifying and managing the risks to the project at all project stages from the initial assessment of strategic options through the procurement, fabrication, construc- tion and commissioning stage (Tah & Carr, 2001). The less ‘‘pre- dictable’’ nature of projects makes them riskier than day to day business activities (Elkington & Smallman, 2002). Risk is a possi- ble undesirable and unplanned event that could result in the project not meeting one or more of its objectives (Teneyuca, 2001). As the underlying concept of risk management is to man- age risks effectively, risk management is a critical part of project management (Lyons & Skitmore, 2004). Construction industries, face a lot of inherent uncertainties and issues like company’s fluctuating profit margin, competitive bid- ding process, weather change, productivity on site, the political sit- uation in a country, inflation, contractual rights, market ll rights reserved. Mousavi), mousavifd@gmail. competition, etc. Thus the construction industry, more than others, has been plagued by risk (Carr & Tah, 2001) and there is no con- struction project with risk free (Lam, Wang, Lee, & Tsang, 2007). With the rapid advancement in the construction industry, an in- creased number of uncertainties are bound to occur (Thevendran & Mawdesley, 2004). It is essential that the construction companies conquer these risks and uncertainties in order to assess the effect of these sources in order to decide which of the projects is more risky, plan for the potential sources of risk in each project and man- age each source during construction (Zayed, Amer, & Pan, 2008). Therefore it is paramount for construction companies to be sensi- tive to the issue of embracing and managing uncertainty and risk discussed above. Project related risk management has attracted steady stream of interest in the academic literature (Bannerman, 2008). One of the major steps in project risk management is to identify and assess the potential risks (El-Sayegh, 2008). Despite many scholars and practitioners recognizing the risk identification methods and assessment models in projects insufficient attention has been paid by researchers to select a suitable risk assessment model. This pa- per attempts to address this limitation and the gap in the current literature and provide a framework for determining optimal risk assessment model. In Section 2, some relevant literature is described. In Section 3, the problem of the risk assessment model selection is intro- duced. Section 4 concentrates on the proposed model. A real case study is presented to illustrate the application of the pro- posed method in Section 5. In the final section some conclusions are drawn. http://dx.doi.org/10.1016/j.eswa.2010.12.110 mailto:nmousavi2930@gmail.com mailto:mousavifd@gmail. com mailto:mousavifd@gmail. com http://dx.doi.org/10.1016/j.eswa.2010.12.110 http://www.sciencedirect.com/science/journal/09574174 http://www.elsevier.com/locate/eswa 9106 A. KarimiAzari et al. / Expert Systems with Applications 38 (2011) 9105–9111 2. Literature review 2.1. Risk The concept of risk became popular in economics during the 1920s. Since then, it has been successfully used in theories of deci- sion making in economics, finance, and the decision science (Ngai & Wat, 2005). Risk has different meaning to different people; that is, the concept of risk varies according to viewpoint, attitudes and experience. Engineers, designers and contractors view risk from the technological perspective; lenders and developers tend to view it from the economic and financial side (Baloi & Price, 2003). The traditional view of risk is negative, representing loss, haz- ard, harm and adverse consequences. But some current risk guide- lines and standards include the possibility of upside risk or opportunity, i.e. uncertainties that could have a beneficial effect on achieving objectives (Hillson, 2002). Project risk is defined by Project Management Body of Knowledge (PMBOK) published by the Project Management Institute (PMI) as an uncertain event or condition that, if it occurs, has a positive or a time, cost, span or quality, which implies an uncertainty about identified events and conditions. PMBOK describes risk through the notion of uncer- tainty; however, these two phenomena are not synonymous (Perminova, Gustafsson, & Wikstrom, 2008). According to the Olsson (2007) and Hillson (2004) attempts to link risk with uncer- tainty based on the distinction between aleatory and epistemic uncertainty in the following couplet: � Risk is measurable uncertainty. � Uncertainty is immeasurable risk. This implies that, when measurable, an uncertainty is to be con- sidered a risk. PMBOK’s definition of risk and uncertainty is the considered definition through the entire paper because this defini- tion implies that risk is quantifiable and lends itself to assessment. 2.2. Risk management If a risk is not identified it cannot be controlled, transferred or otherwise managed (Bajaj, 1997) and trying to eliminate all risks in projects is impossible. Thus, there is need for a formal risk man- agement process to manage all types of risks. The project success usually depends on the combination of all risks, response strategies used to mitigate risks and a company’s ability to manage those (Dikmen, Birgonul, & Han, 2007). Hence, the underlying concept of risk management is to manage risks effectively (Thevendran & Mawdesley, 2004). Risk management can lead to a range of project and organizational benefits including: (Bannerman, 2008) � Identification of favorable alternative courses of action. � Increased confidence in achieving project objectives. � Improved chances of success. � Reduced surprises. � More precise estimates (through reduced uncertainty). � Reduced duplication of effort (through team awareness of risk control actions). PMBOK included risk management as one of the nine focuses in project management and described it as the process concerned with conducting risk management planning, identification, analy- sis, responses, and monitoring and control on a project (Zou, Zhang, & Wang, 2007). Risk management in construction is a tedious task as the objective functions tend to change during the project life cycle, and the scenarios are numerous due to sensitivity of projects to uncontrollable risks stemming from the changes in the macro-environment, existence of high number of parties involved in the project value chain, and one-off nature of the construction process (Dikmen, Birgonul, Anac, Tah, & Aouad, 2008). Project risk management is an integrated process which in- cludes activities to identify project uncertainty, estimate their im- pact, analyze their interactions, control them in the execution stage, and even provide feedback to the maintenance of collective knowledge asset (Williams, 1995). Risk management based on con- sensus in the literatures, used the following three-step approach (Zayed et al., 2008): � Risk identification. � Risk assessment. � Risk mitigation. The first step in risk management is risk identification. Before risks can be managed, they must be identified. Identification sur- faces risks before they become problems and adversely affect a project. It refers to the evidences from previous experience or sim- ilar cases which would apply to the current project, in order to avoid or ameliorate the probability of compromising the project’s success. Construction risks can be categorized in a number of ways based on the source of risk, impact of risk or by project phase (Klemetti, 2006). In the most reference one, project risks are di- vided into two groups, according to their source, into internal and external. Internal risks are initiated inside the project while external risks originate due to the project environment (El-Sayegh, 2008). In risk identification step all internal and external risks must be identified. After the establishment of a list of risk events that had actually occurred in the process of project performance, these risks must be assessed. The primary objective of risk assessment is to estimate risk by identifying the undesired event, the likelihood of occurrence of the unwanted event, and the consequence of such event. Risk assessment involves measures, either conducted quantitatively or qualitatively, to produce the estimation of the significance level of the individual risk factors to the project, so as to produce the estimation of the risk of the potential factors to project success. However, this step results will become the input to the determina- tion of the optimum decision. With a better quantification measur- ing result, the managers can recognize which risks are more important and then deploy more resources on it to eliminate or mitigate the expected consequences. The identification and assessment of project risk are the critical procedures for projecting success, and they usually become the essential factors in the decision-making process (Williams, 1995). Most authors refer to the processes which include risk identifica- tion and assessment, as the stage called ‘‘risk analysis’’. Risk anal- ysis can provide insight to the specific sources of project risk and enable management to devise targeted remedial action. Several methods have been proposed and utilized thorough re- search by a lot of scholars to help contractors and subcontractors to evaluate and select the best projects in order to decide which pro- jects are more risky. And so these models help to plan for the po- tential sources of risk in each project and manage each source during construction. Currently project management teams have more options from which to choose. Risk assessment methods have ranged from simple classical methods to fuzzy approach mathematical models. Many construc- tion project risk assessment techniques currently used are compar- atively mature tools (Zeng, An, & Smith, 2007). Monte Carlo Simulation (White, 1995), Sensitivity Analysis (White, 1995), Critical path method (Kaufmann & Gupta, 1988), Fault tree analysis (Terano, Asai, & Sugeno, 1992), Event tree analysis (Huang, Chen, & Wang, 2001), Failure mode, effects and A. KarimiAzari et al. / Expert Systems with Applications 38 (2011) 9105–9111 9107 criticality analysis (Bowles & Pelaez, 1995) are the classical quan- titative methods, used in construction industry for risk assessment. These methods only use data that are quantitative so, for effective application of these sophisticated quantitative techniques high quality data are a prerequisite (Zeng et al., 2007). Only on a few projects and contracts are risk considered in a consistent and log- ical manner; much assessment is too subjective (Mills, 2001). So, some other models suggested, involve both quantitative and qual- itative ones. Fuzzy risk assessment methods have also been deployed with some scholars too. Mustafa and Al-Bahar (1991) investigated the subject of risk assessment and developed a scheme of classifying the various sources of risk in construction projects. They applied Analytical Hierarchical Process (AHP) in assessing the riskiness of a real-life constructing project (Mohammad & Al-Bahar, 1991). Sadiq and Husain (2005) developed a three-stage hierarchical structure aggregative risk model for grouping of risk items. For this grouping, an analytical hierarchy process was used for assessment. Another hierarchical risk breakdown structure is described to represent a formal model for qualitative risk assessment by Carr and Tah (2001). In their paper, using fuzzy approximation and composition, the relationships between risk sources and the consequences on project performance measures were identified and quantified. Cho, Choi, and Kim (2002) proposes another methodology for incor- porating uncertainties using fuzzy concepts into conventional risk assessment frameworks in construction industry. Choi, Cho, and Seo (2004) presents fuzzy risk assessment methodology for under- ground construction projects. A formalized procedure and associ- ated tools were developed to assess and manage the risks involved in underground construction. The suggested risk assess- ment procedure is composed of four steps of identifying, analyzing, evaluating, and managing the risks inherent in construction pro- jects. Zeng et al. (2007) developed a methodology to deal with risks associated with the construction projects in the complicated situa- tions. This model can handle with the expert knowledge, engineer- ing judgment and the historical data for risk assessment and in this model the risk can be evaluated directly using linguistic terms which are employed in risk assessment. Zayed et al. (2008) intro- duces a model, based on AHP to help practitioners to assess Chinese highway risk projects and prioritize them. This methodology quan- tifies the qualitative effect of subjective factors of risk. These methods differ in a variety of ways and they have their own advantages and disadvantages. So an ideal risk assessment method which would suit all organizations does not exist, as each of the organizations and projects possesses its own unique charac- teristics (Lichtenstein, 1996), so, an organization and project man- agement team need to select the most appropriate methodology on its specific. This problem labeled as the risk assessment model selection. 3. Risk assessment model According to Lichtenstein (1996) selecting which model is suit- able for the organization or project is affected by many factors. The cost of employing the technique, the level of external party’s approval, Organizational structure, Agreement, Adaptabil- ity, Complexity, Completeness, Level of risk, Organizational size, Organizational security philosophy, Consistency, Usability, Feasi- bility, Validity, Credibility and Automation are factors to be consid- ered in the selection of a risk assessment method (Lichtenstein, 1996). Owning to several quantitative and qualitative factors, some of which them may be in conflict with the others, the risk assessment technique selection is complicated. Risk assessment model involves two other key modeling as- pects: First, construction is described as a collaborative teamwork process where parties with different interests, functions, and objectives, share a common goal, which is successful completion of a project. Thus in this problem solving it is vital to involve sev- eral people from different parties. A second important consideration of the risk assessment model selection is that much knowledge in the real world is imprecise than precise (Olcer & Odabasi, 2005), thus the preference informa- tion provided to model selection may be imprecise or incomplete. As a result, multiple factors, which are either quantitative or qualitative and may be in conflict with each other, impact the risk assessment model selection problem and problem arises in group setting with incomplete, vague and uncertain information. In line with the multidimensional characteristics of the risk assessment model selection, the problem is a kind of multi-criteria decision-making (MCDM) problem, which requires MCDM meth- ods for an effective problem solving. MCDM refers to screening, prioritizing, ranking or selecting a set of alternatives under usually independent, incommensurate or conflicting attributes (Hwang & Yoon, 1981). It can rank differ- ent methods when they are compared in terms of their overall performance. Over the years, a variety of MCDM theories and techniques have been proposed by different behavioral scientists, operational researchers and decision theorists. The methods differ in many areas of theoretical background, type of questions asked and the type of results given (Hobbs & Meier, 1994). The availability and selection of such a method depends on the structure of the model and the information that can be collected. For brevity, a short introduction to TOPSIS is provided and the reader is referred to Saremi, Mousavi, and Sanayei (2009) and Shih, Syur, and Lee (2007) for a more in-depth treatment. The Technique for Order Preference by Similarity to Ideal Solu- tion (TOPSIS), proposed by Hwang and Yoon (1981), referring to the positive and negative ideal solutions as the ideal and anti-ideal solutions is a widely used MCDM method. It based on the concept that the chosen alternative should have the shortest distance from the positive ideal solution (PIS) and the farthest from the negative- ideal solution (NIS) for solving a multiple criteria decision-making problem. According to the simulation comparison of Zanakis, Solomon, Wishart, and Dublish (1998), this technique has the few- est rank reversals among the eight methods of MCDM (Shih, 2008). Chen (2000) extended TOPSIS to fuzzy environments; this ex- tended version used fuzzy linguistic value (represented by fuzzy number) as a substitute for the directly given crisp value in grade assessment. This modified TOPSIS is a practical method and fits hu- man thinking under actual environment (Wang, Cheng, & huang, 2009). The difference between TOPSIS and fuzzy TOPSIS chiefly lies in rating approaches. The merit of fuzzy TOPSIS is using fuzzy num- bers instead of precise numbers (Chen & Tsao, 2008). Fuzzy TOPSIS is flexible and efficient method that is easily under- stood by practitioners and researchers. Fuzzy TOPSIS method is extended in this paper for selecting a proper risk assessment model. 4. Proposed method A systematic approach to extend the TOPSIS is proposed to solve the risk assessment model selection problem under a fuzzy envi- ronment in this section. Assume that there is a committee of k decision makers K ¼ð1; 2; . . . ; kÞ who are responsible for assessing the project risks. Once the set of possible models are selected, the committee determines the best of these models. Model selection first requires Table 1 Linguistic variables for the ratings. Linguistic variables Fuzzy triangular Very low/very poor (0, 0, 1) Low/poor (0, 1, 3) Medium low/medium poor (1, 3, 5) Medium/fair (3, 5, 7) Medium high/medium good (5, 7, 9) High/good (7, 9, 10) Very high/very good (9, 10, 10) 9108 A. KarimiAzari et al. / Expert Systems with Applications 38 (2011) 9105–9111 identification of decision attributes (criteria). Various techniques exist in order to reach a consensus among the experts (Bryson, Mobolurin, & Joseph, 1997). Nominal Group Technique (NGT), Delphi (Van De Ven & Delbecq, 1974), Focus Groups and Brainstorming (Stewart & Shamdasani, 1990) are formal and more useful group management techniques. When comparing the NGT with other group processes the NGT has a number of advantages over other group processes (Potter, Gordon, & Hamer, 2004) thus the NGT technique is suggested to obtain decision criteria/factors. Selected criteria can be classified into two types: benefit factors (C1) and cost ones (C2). After this, members in the risk assessment group are required to provide their judgment on the basis of their knowledge and exper- tise for each model. Decision makers make decisions on the basis of their knowledge of the facts and personal experience. Their judgments and prefer- ences are often vague, inexact, imprecise and uncertain by nature which makes it difficult to estimate their preference with an exact numerical value since crisp data are inadequate to model real-life situations. Decision makers describe their preference with words or sentences in a natural or artificial language. In these circum- stances values are not numbers but linguistic terms. A linguistic variable is a variable whose values (namely linguistic values) have the form of phrases or sentences in a natural language (Von Altrock, 1996). Linguistic values can be represented by fuzzy num- bers. A fuzzy number is a convex fuzzy set, characterized by a given interval of real numbers, each with a grade of membership be- tween 0 and 1 (Wang & Elhag, 2006). In the following, some basic definitions of fuzzy sets theory will be reviewed briefly from Cheng and Lin (2002), Kaufmann and Gupta (1985), and Raj and Kumar (1999). A real fuzzy number A is described as a fuzzy subset of the real line R with member function fA that represents uncertainty. A membership function is defined from universe of discourse to [0, 1] (see Fig. 1). A triangular fuzzy number can be defined as a triplet (a, b, c); the membership function of the fuzzy number A is defined as: fA ¼ 0; x 6 a; x�a b�a ; a 6 x 6 b; c�x c�b ; b 6 x 6 c; 0; x 6 c: 8>>>< >>>: ð1Þ This representation is useful for arithmetic operation on fuzzy numbers. With this notation, the arithmetic operations on fuzzy numbers are defined as follows: ða1; b1; c1ÞðþÞða2; b2; c2Þ¼ ða1 þ a2; b1 þ b2; c1 þ c3Þ; ð2Þ ða1; b1; c1Þð�Þða2; b2; c2Þ¼ ða1 � c2; b1 � b2; c1 � a2Þ; ð3Þ ða1; b1; c1Þð�Þða2; b2; c2Þ¼ ða1 � a2; b1 � b2; c1 � c2Þ; ð4Þ ða1; b1; c1Þð�Þða2; b2; c2Þ¼ ða1 � c2; b1 � b2; c1 � a2Þ; ð5Þ ða1; b1; c1Þ �1 ¼ 1 c1 ; 1 b1 ; 1 a1 � � ; ð6Þ k �ða1; b1; c1Þ¼ ðka1; kb1 þ kc1Þ: ð7Þ Fig. 1. A triangular fuzzy number. According to the vertex method stated by Chen (2000), the dis- tance between fuzzy numbers (a1, b1, c1) and (a2, b2, c2) is calculated as: dðA1; A2Þ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 3 ða1 � a2Þ 2 þðb1 � b2Þ 2 þðc1 � c2Þ 2 h ir : ð8Þ Obviously, in the risk assessment model selection problem the models and the factor set Cj = (j = 1, 2, . . . , j) are finite, so it is very convenient to denote the rating of models on factors by fij. Then a decision problem can be concisely expressed as the fol- lowing decision matrix: Dk ¼ ½f kij �m�n; f kij ¼ðaij; bij; cijÞ is a linguistic variable, indicating the rating of each ith method with respect to each jth factor respect to kth DM. These linguistic variables can be described by triangular fuzzy numbers as shown in Table 1. In the decision making process, different attributes have differ- ent importance. Suppose wi = (i = 1, 2, . . . , m) is the relative weight of factor Ci, where wi P 0 and Pi¼n i¼1 wi ¼ 1. Denote a weight vector by w = (w1, w2, . . . , w6) T. Establishing the relative importance of fac- tors can be obtained by either directly assigning or indirectly using pair-wise comparisons (Cook, 1992). In many real-life cases, a deci- sion maker cannot generally specify exact attribute weights but can provide value ranges (Xu & Chen, 2007) thus it is suggested in this paper that linguistic variables are used for assigning the pri- ority weights of factors. The linguistic variable schemes in the rat- ing set and weighting set, shown in Tables 1 and 2, respectively, are used in this study to evaluate the ratings of strategies with respect to different factors and the importance of the factors. If there is consensus among DMs with respect to rating and importance of factors suppose k = 1 in proposed procedure. The procedure of the TOPSIS method consists of the following steps: 4.1. Establish a normalized matrix The decision matrix must first be normalized so that the ele- ments will be unit-free. The structure of the normalized matrix for the kth decision maker can be expressed as follows: Table 2 Linguistic variables for the importance weight of each criterion. Linguistic variables Fuzzy triangular Very low (0, 0, 0.1) Low (0, 0.1, 0.3) Medium low (0.1, 0.3, 0.5) Medium (0.3, 0.5, 0.7) Medium high (0.5, 0.7, 0.9) High (0.7, 0.9, 1) Very high (0.9, 1, 1) A. KarimiAzari et al. / Expert Systems with Applications 38 (2011) 9105–9111 9109 Rk ¼ ½rkij�m�n; k ¼ 1; 2; . . . ; K; i ¼ 1; 2; . . . ; m; j ¼ 1; 2; . . . ; n; ð9Þ where the rkij is the normalized value of f k ij ¼ðaij; bij; cijÞ which be cal- culated by the following relations: � If jth criterion is a benefit one: rkij ¼ aij c�j ; bij c�j ; cij c�j ! ; ð10Þ where c�j ¼ max cij. � And if jth criterion is a cost one: rkij ¼ a�j cij ; a�j bij ; a�j aij � � ; ð11Þ Table 3 Importance weight of criteria from three decision-makers. DM1 DM2 DM3 C1 MH M M C2 VH H H C3 L VL VL C4 H MH VH where a�j ¼ min aij . 4.2. Construct the weighted normalized decision matrix The columns of the normalized decision matrix for kth decision maker by the associated priority weights of factors are multiplied to construct the weighted normalized decision matrix as follows: V k ¼ ½v kij�m�n; k ¼ 1; 2; . . . ; K; i ¼ 1; 2; . . . ; m; j ¼ 1; 2; . . . ; n; ð12Þ where v kij ¼ r k ijð�Þw k j and v k ij ¼ðv ij1; v ij2; v ij3Þ is a triangular fuzzy number. 4.3. Calculate the separation measure from the ideal and the negative ideal solutions for each decision maker The positive ideal solution indicates the most preferable alter- native, and the negative ideal solution indicates the least prefera- ble alternative. So determine the fuzzy positive ideal solution (FPIS, A+) and fuzzy negative ideal solution (FNIS, A�) as follows (Chen, Lin, & Huang, 2006): Aþ ¼ ðvþ1 ; v þ 2 ; . . . ; v þ n Þ; A � ¼ ðv�1 ; v � 2 ; . . . ; v � n Þ; ð13Þ where vþj ¼ maxiðv ij3Þ and v � j ¼ miniðv ij1Þ; i ¼ 1; 2; . . . ; m; j ¼ 1; 2; . . . ; n. For kth, the distance of each alternative from A+ and A� can be currently calculated as: dkþi ¼ Xn j¼1 dðv kij; v þ j Þ; i ¼ 1; 2; . . . ; m; j ¼ 1; 2; . . . ; n ð14Þ and dk�i ¼ Xn j¼1 dðv kij; v � j Þ; i ¼ 1; 2; . . . ; m; j ¼ 1; 2; . . . ; n; ð15Þ where d(⁄, ⁄) represented the distance measurement between two fuzzy numbers. 4.4. Calculate the overall separation measure from the ideal and the negative ideal solutions To derive group preferences provided by multiple decision mak- ers and combine the group synthesis and prioritization stages into a single integrated stage, the geometric mean with the modified TOPSIS approach is employed. The overall separation measure calculated as: �dþi ¼ YK k¼1 dkþi !1 k ; i ¼ 1; 2; . . . ; m ð16Þ and �d�i ¼ YK k¼1 dk�i !1 k ; i ¼ 1; 2; . . . ; m: ð17Þ 4.5. Calculate the relative closeness to the ideal solution The relative closeness of the alternative Aj with respect to ideal solution A+ is defined as: Ci ¼ �d�i �dþi þ �d�i ; ð18Þ where Ci range belong to the closed interval [0, 1] and i ¼ 1; 2; . . . ; m. 4.6. Rank the alternatives A set of alternatives can now be preference ranked according to the descending order of Ci. The one with the maximum value of Ci is the best. 5. Numerical example To illustrate the group based fuzzy TOPSIS approach introduced above; risk assessment model selection problem faced by XYZ is presented. The case company XYZ is one of the Iranian construc- tion companies. Recently FNP Co. has taken a huge project in road construction. Risk management team is formed to manage risks in the project. Three experts with high qualification regarding project risks are se- lected to form a group. Risk assessment model selection is one of the fundamental tasks of the team. The proposed method com- monly taken decides among a three possible models. After the NGT technique is employed, the group identifies four criteria, for the risk assessment model selection, as follows: � Implementation cost. � External party’s approval. � Complexity. � Usability. The group, respectively, compares the four criteria and evalu- ates their degree of satisfaction with every model. The compared and evaluated grades are shown in Table 3 (see Tables 1 and 2 for the linguistic value and degree of importance). The ratings of the three consultants by the decision makers against the various criteria are shown in Tables 4. The linguistic evaluations are shown in Tables 3 and 4 are converted into trian- gular fuzzy numbers. After establishing a normalized matrix, the weighted normalized fuzzy decision matrix is calculated. To save space the other matrixes are omitted and only the separation from the ideal and the negative ideal solutions for each DM is shown in Table 5. Next, to derive group priorities, the group’s aggre- gated separation distances are generated by its geometric mean. Table 4 Ratings of the 3 consultants by the DMs under the various criteria. C1 C2 C3 C4 DM1 A1 M L ML ML A2 H VL VH L A3 MH VH M VH DM2 A1 ML ML M L A2 VH VL VH VL A3 M VH ML VH DM3 A1 ML M ML L A2 H L VH VL A3 M VH M VH Table 5 The distance measurement. DM1 DM2 DM3 d�1 d þ 1 d�1 d þ 1 d�1 d þ 1 A1 2.33 1.1 2.11 0.9 2.02 1.1 A2 2.78 0.4 2.58 0.1 2.58 0.3 A3 1.18 2.1 1.17 1.7 1 1.9 Table 6 The final closeness coefficient of each model. Model Overall A1 0.441 A2 0.321 A3 0.545 9110 A. KarimiAzari et al. / Expert Systems with Applications 38 (2011) 9105–9111 Table 6 represents the result. At last according to the closeness of the 3 consultant the A3 is the best. 6. Conclusion Despite its importance to the success of project management, risk management is rarely approached with the same rigor as other project management processes such as project scope and schedul- ing. A process of risk management has involved risk identification, risk assessment and risk mitigation. The identification and assess- ment of project risk are the critical procedures for projecting suc- cess. Many construction project risk assessment techniques are currently used in the construction industry but insufficient atten- tion has been paid by researchers to a select suitable risk assess- ment model. To address this decision problem, in this paper a group based fuzzy TOPSIS approach is developed with an effective algorithm to improve the quality and effectiveness of decision making. TOPSIS provides good evaluations and it appears to be more appropriate than other MCDM methods. Construction project would require interaction between dissim- ilar, yet contractually integrated parties, owners, designers, con- tractors, sub-contractors, suppliers, manufacturers, and others. As a result, construction is described as a collaborative teamwork pro- cess where parties with different interests, functions, and objec- tives, share a common goal, which is successful completion of a project. In line with a group decision environment of the problem, the approach provides a simple and effective mechanism to make comparative and absolute judgments in a conventional manner. The proposed method can discriminate successfully and clearly among risk assessment methods. Further research can apply this method to other decision situa- tions in construction industry, like project selection, performance selection, vendor selection, etc. References Bajaj, J. (1997). Analysis of contractors approaches to risk identification in New South Wales, Australia. Construction Management and Economics, 15, 363–369. Baloi, D., & Price, A. D. F. (2003). Modelling global risk factors affecting construction cost performance. International Journal of Project Management, 21, 261–269. Bannerman, P. L. (2008). Risk and risk management in software projects: A reassessment. The Journal of Systems and Software, 81(12), 2118–2133. Bowles, J. B., & Pelaez, C. E. (1995). Fuzzy logic prioritization in a system failure mode, effects and criticality analysis. Reliability Engineering and System Safety, 50, 203–213. Bryson, N., Mobolurin, A., & Joseph, A. (1997). Generating consensus fuzzy cognitive maps. Intelligent Information Systems, 8(10), 231–235. Carr, V., & Tah, J. H. M. (2001). A fuzzy approach to construction project risk assessment and analysis: Construction project risk management system. Advanced in Engineering Software, 32, 847–857. Chen, C. T. (2000). Extensions of the TOPSIS for group decision-making under fuzzy environment. Fuzzy Sets and Systems, 114, 1–9. Chen, C.-T., Lin, C.-T., & Huang, S.-F. (2006). A fuzzy approach for supplier evaluation and selection in supply chain management. International Journal of Production Economics, 102, 289–301. Chen, T.-Y., & Tsao, C.-Y. (2008). The interval-valued fuzzy TOPSIS method and experimental analysis. Fuzzy Sets and Systems, 159, 1410–1428. Cheng, C. H., & Lin, Y. (2002). Evaluating the best main battle tank using fuzzy decision theory with linguistic criteria evaluation. European Journal of Operational Research, 142(1), 174–186. Cho, H. N., Choi, H. H., & Kim, Y. B. (2002). A risk assessment methodology for incorporating uncertainties using fuzzy concepts. Reliability Engineering & System Safety, 78, 173–183. Choi, H., Cho, H., & Seo, J. W. (2004). Risk assessment methodology for underground construction projects. Journal of Construction Engineering and Management, 130(2), 258–272. Cook, R. L. (1992). Expert systems in purchasing applications and development. International Journal of Purchasing and Management, 18, 20–27. Dikmen, I., Birgonul, M. t., Anac, C., Tah, J. H. M., & Aouad, G. (2008). Learning from risks: A tool for post-project risk assessment. Automation in Construction, 18, 42–50. Dikmen, I., Birgonul, M. T., & Han, S. (2007). Using fuzzy risk assessment to rate cost overrun risk in international construction projects. International Journal of Project Management, 25, 494–505. Elkington, P., & Smallman, C. (2002). Managing project risks: A case study from the utilities sector. International Journal of Project Management, 20, 49–57. El-Sayegh, S. M. (2008). Risk assessment and allocation in the UAE construction industry. International Journal of Project Management, 26, 431–438. Fan, M., Lin, N., & Sheu, C. (2008). Choosing a project risk-handling strategy: An analytical model. International Journal of Project Management, 112, 700–713. Hillson, D. (2002). Extending the risk process to manage opportunities. International Journal of Project Management, 20, 235–240. Hillson, D. (2004). Effective opportunity management for projects – exploiting positive risk. New York: Marcel Dekker. Hobbs, B. F., & Meier, P. M. (1994). Multi criteria methods for resource planning: An experimental comparison. IEEE Transactions on Power Systems, 9(4), 1811–1817. Huang, D., Chen, T., & Wang, M.-J. (2001). A fuzzy set approach for event tree analysis. Fuzzy Sets and Systems, 118, 153–165. Hwang, C. L., & Yoon, N. (1981). Multiple attributes decision making methods and application. Berlin: Springer-Verlag. Kaufmann, A., & Gupta, M. M. (1985). Introduction to fuzzy arithmetic: Theory and applications. New York: Van Nostrand Reinhold. Kaufmann, A., & Gupta, M. M. (1988). Fuzzy mathematical models in engineering and management science. Amsterdam: North-Holland. Klemetti, A. (2006). Risk management in construction project networks. Report 2006/ 2. Finland: Laboratory of Industrial Management, Helsinki University of Technology. Lam, K. C., Wang, D., Lee, p. T. K., & Tsang, Y. T. (2007). Modelling risk allocation decision in construction contracts. International Journal of Project Management, 25, 485–493. Lichtenstein, Sh. (1996). Factors in the selection of a risk assessment method. Information Management and Computer Security, 4, 20–25. Lyons, T., & Skitmore, M. (2004). Project management in the Queensland engineering construction industry: A survey. International Journal of Project Management, 22, 51–61. Mills, A. (2001). A systematic approach to risk management for construction. Structural Survey, 19(5), 245–252. Mohammad, A. M., & Al-Bahar, J. F. (1991). Project risk assessment using the analytic hierarchy process. IEEE Transactions on Engineering Management, 38(1), 46–52. Mustafa, M. A., & Al-Bahar, J. F. (1991). Project risk assessment using the analytic hierarchy process. IEEE Transactions on Engineering Management, 38(1), 46–52. Naaranoja, M., Haapalainen, P., & Lonka, H. (2007). Strategic management tools in projects case construction project. International Journal of Project Management, 25, 659–665. Ngai, E. W. T., & Wat, F. K. T. (2005). Fuzzy decision support system for risk analysis in e-commerce development. Decision Support Systems, 40, 235–255. Olcer, A. I., & Odabasi, A. Y. (2005). A new fuzzy multiple attributive group decision making methodology and its application to population/maneuvering system selection problem. European Journal of Operational Research, 166, 93–114. A. KarimiAzari et al. / Expert Systems with Applications 38 (2011) 9105–9111 9111 Olsson, R. (2007). In search of opportunity management: Is the risk management process enough? International Journal of Project Management, 25, 745–752. Perminova, O., Gustafsson, M., & Wikstrom, K. (2008). Defining uncertainty in projects – A new perspective. International Journal of Project Management, 26, 73–79. Potter, M., Gordon, S., & Hamer, P. (2004). The nominal group technique: A useful consensus methodology in physiotherapy research. NZ Journal of Physiotherapy, 32(3), 126–130. Project Management Institute (2000). A guide to the project management book of knowledge (PMBOK). Newtown Square, PA: Project Management Institute. Project Management Institute (2004). A guide to the project management book of knowledge (PMBOK) (3rd ed.). Newtown Square, PA: Project Management Institute. Raj, P. A., & Kumar, D. N. (1999). Ranking alternatives with fuzzy weights using maximizing set and minimizing set. Fuzzy Sets and Systems, 105, 365–375. Sadiq, R., & Husain, T. (2005). A fuzzy-based methodology for an aggregate environmental risk assessment: A case study of drilling waste. Environmental Modelling & Software, 20, 33–46. Saremi, M., Mousavi, S. F., & Sanayei, A. (2009). TQM consultant selection in SMEs with TOPSIS under fuzzy environment. Expert Systems with Applications, 2742–2749. Shih, H.-S. (2008). Incremental analysis for MCDM with an application to group TOPSIS. European Journal of Operational Research, 186(2), 720–734. Shih, H. S., Syur, H. J., & Lee, E. S. (2007). An extension of TOPSIS for group decision making. Mathematical and Computer Modeling, 45, 801–813. Steiner, G. A. (1969). Top management planning. New York: Macmillan. Stewart, D. W., & Shamdasani, P. N. (1990). Focus groups: Theory and practice. London: Sage. Tah, J. H. M., & Carr, V. (2001). Towards a framework for project risk knowledge management in the construction supply chain. Advanced in Engineering Software, 32, 835–846. Teneyuca, D. (2001). Organizational leader’s use of risk management for information technology. Information Security Technical Report, 6(3), 54–59. Terano, T., Asai, K., & Sugeno, M. (1992). Fuzzy systems theory and its applications. San Diego, CA: Academic Press. Thevendran, V., & Mawdesley, M. J. (2004). Perception of human risk factors in construction projects: An exploratory study. International Journal of Project Management, 22, 131–137. Van De Ven, A. H., & Delbecq, A. L. (1974). The effectiveness of nominal, delphi, and interacting group decision making process. Academy of Management Journal, 17(4), 605–621. Von Altrock, C. (1996). Fuzzy logic and neurofuzzy applications in business and finance. New Jersey: Prentice-Hall. Wang, J.-W., Cheng, Ch.-H., & huang, K.-C. (2009). Fuzzy hierarchical TOPSIS for supplier selection. Applied Soft Computing, 9(1), 377–386. Wang, Y. M., & Elhag, T. M. S. (2006). Fuzzy TOPSIS method based on alpha level sets with an application to bridge risk assessment. Expert Systems with Applications, 31, 309–319. White, D. (1995). Application of systems thinking to risk management: A review of the literature. Management Decision, 3(10), 35–45. Williams, T. (1995). A classified bibliography of recent research relating to project risk management. European Journal of Operational Research, 85, 18–38. Xu, Z., & Chen, J. (2007). An interactive method for fuzzy multiple attribute group decision making. Information Sciences, 177, 248–263. Zanakis, S. H., Solomon, A., Wishart, N., & Dublish, S. (1998). Multi-attribute decision making: A simulation comparison of selection methods. European Journal of Operational Research, 107, 507–529. Zayed, T., Amer, M., & Pan, J. (2008). Assessing risk and uncertainty inherent in Chinese highway projects using AHP. International Journal of Project Management, 26, 408–419. Zeng, J., An, M., & Smith, N. J. (2007). Application of a fuzzy based decision making methodology to construction project risk assessment. International Journal of Project Management, 25, 589–600. Zou, P. X. W., Zhang, G., & Wang, J. (2007). Understanding the key risks in construction projects in China. International Journal of Project Management, 25, 601–614. Risk assessment model selection in construction industry Introduction Literature review Risk Risk management Risk assessment model Proposed method Establish a normalized matrix Construct the weighted normalized decision matrix Calculate the separation measure from the ideal and the negative ideal solutions for each decision maker Calculate the overall separation measure from the ideal and the negative ideal solutions Calculate the relative closeness to the ideal solution Rank the alternatives Numerical example Conclusion References