A Model for Risk Evaluation in Construction Projects Based on Fuzzy MADM

Transcription

1 A Model for Risk Evaluation in Construction Projects Based on Fuzzy MADM S. Ebrahimnejad 1, S. M. Mousavi, S. M. H. Mojtahedi 1 Department of Industrial Engineering, Islamic Azad University - Karaj Branch, Iran Department of Industrial Engineering, Graduate School, Islamic Azad University - South Tehran Branch, Member of Young Researchers Club, Tehran, Iran (Ibrahimnejad@kiau.ac.ir) Abstract - The number, size and complexity of new projects in Iran have created an extra burden on construction participants and resulted in lots of risks. In this paper, at first we identify the significant risks in construction industry project; then we introduce some effective criteria and attributes is used for risk evaluating in construction industry. Afterwards the problem is defined in fuzzy MADM field. Therefore, fuzzy TOPSIS and fuzzy LINMAP methods are presented to evaluate the high risks in the project. This study compares the modeling mechanisms of the two methods and their performances in modeling a set of project risk data. Finally, a case study is used to illustrate the procedure of the proposed model at the end of paper. Keywords - Construction projects, risk evaluation, fuzzy MADM I. INTRODUCTION The increasing growth of the Iran construction Industry calls for massive development of infrastructures and assets. While this brings opportunities to project stakeholders, employing effective risk management techniques coped risks associated with variable construction activities is of importance to implement the projects aligning with project objectives including time, cost, quality, safety and environmental sustainability. Therefore, it is important to identify and assess the significant risks in the Iran construction industry in order to help local companies and international companies who do or plan to work in Iran to pay attention to these significant risks [1]. Construction risk assessment can be well modeled with using fuzzy set theory. In fact, there have been limited attempts to exploit fuzzy logic within the construction risk management domain. Kangari [] presented an integrated knowledge-based system for construction risk management using fuzzy sets. The system, called Expert-Risk, performs risk analysis in two situations: before construction, and during construction. Chun and Ahn [3] proposed the use of fuzzy set theory to quantify the imprecision and judgmental uncertainties of accident progression event trees. Peak et al. [4] proposed the use of fuzzy sets for the assessment of bidding prices for construction projects. Ross and Donald [5] describe a method for assessing risk based on fuzzy logic and similarity measures. Carr and Tah [6] presented a model that involves the relationships between risk factors, risks, and their impacts based on cause and effect diagrams. Also Dikman et al. [7] proposed a fuzzy risk assessment for international construction projects. This methodology utilizes the influence diagramming method and estimate a cost overrun risk rating. Zeng et al. [8] presented a risk assessment model based on fuzzy reasoning and modified Analytical Hierarchy Process (AHP) to handle the uncertainties arising in the construction process. The evaluation on previous researches and studies shows that higher risk evaluating modeling problem has not been considered as equal as project risk assessment problem and few studies had been performed in the Iran construction industry. The proposed model in this paper allows risks to be ranked for management priority via fuzzy multi criteria decision making. Most organizations have limited resources to manage all risks equally on project. To overcome this problem, the organization can assess and prioritize the significant risks of project, so that an appropriate level of effort can be applied to the management of those projects. In particular, resources will be directed to manage project with the higher risk ranking. Moreover, we present new effective criteria assigned for risk evaluating in construction projects. The paper is organized as follows: In Section II, we briefly introduce some basic concepts on fuzzy sets to pave the way for the fuzzy TOPSIS and fuzzy LINMAP methods modeling. In Section III, we propose project risk evaluating model in construction industry and new criteria based on developing risk concepts. Section IV investigates a case study using the proposed project risk evaluating to illustrate their potential applications in construction industries. The discussion of results is provided in Section IV.E. Finally, Conclusions are offered in Section V. II. FUZZY MULTI ATTRIBUTE DECISION MAKING PROBLEM The aim of the MADM is to obtain the optimum alternative that has the highest degree of satisfaction for all of the relevant attributes. MADM problem can be dealt with using several existing methods such as the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) developed by Hwang and Yoon [9], the Linear Programming Technique for Multidimensional Analysis of Preference (LINMAP) developed by Srinivasan and Shocker [10]. The TOPSIS and LINMAP methods are two well-known MADM methods and they are used in various industrial fields. In the TOPSIS method, the decision matrix D and the weight vector W are given as crisp values a priori; a /08/$ IEEE 305

2 positive ideal solution (PIS) and a negative ideal solution (NIS) are generated from D directly; the best compromise alternative is then defined as the one that has the shortest distance to the PIS and the farthest from the NIS. However, in the LINMAP method, the weight vector W and the PIS are unknown a priori. The LINMAP method is based on pair wise comparisons of alternatives given by the decision maker and generates the best compromise alternative as the solution that has the shortest distance to the PIS. In the LINMAP and TOPSIS methods, all the decision data are known precisely or given as crisp values. However, under many conditions, crisp data are inadequate or insufficient to model real-life decision problems [11,1,13]. We briefly review the rationale of fuzzy; as follows: Definition.1. A fuzzy set a ~ in a universe of discourse x is characterized by a membership function μa ~ ( x) which associates with each element x in X, a real number in the interval [ 0,1]. The function value μ a ~ ( x) is termed the grade of membership of x in a ~ [11]. Definition.. Let a ~ = ( a1, a, a3 ) and b ~ = ( b 1, b, b3 ) be two triangular fuzzy numbers, then the vertex method is defined to calculate the distance between them, Refer to (1) [13]: ~ 1 d( a ~, b ) = [( a1 b1 ) + 4( a b ) + ( a3 b3 ) ]. (1) 6 Property.1. Assuming that both = a 1, a, a3 and b ~ = ( b 1, b, b3 ) are real numbers, then the distance d a ~ ~, b is identical to the Euclidean distance. measurement ( ) a ~ ( ) Property.. Let a ~, b ~, and c ~ be three triangular fuzzy numbers. The fuzzy number b ~ is closer to fuzzy number a ~ than the other fuzzy number c ~ if, and only if, d a ~ ~, b d a ~, c ~ [11]. ( ) ( ) A. The Normalization Method The vector normalization is used to calculated r ~ ij. aij aij aij r ~ 1 3 ij =,, i = 1,,..., m : j = 1,,..., e j e j e j n. () where = m e j a ij i= 1. B. Fuzzy TOPSIS and fuzzy LINMAP methods The readers can follow the procedure of fuzzy TOPSIS and fuzzy LINMAP in [11,13,14]. III. PROJECT RISK EVALUATEING NEW MODEL Risk analysis happens throughout the project life cycle. As new risks become evident and identified, the project manager should route the risks through the risk analysis process. The end result of project risk analysis is risk ing. A. New Effective Criteria Probability and impact criteria are not sufficient for covering all aspects of project risks solely, on the other hand, MCDM gives an opportunity to take advantage of exact and appropriate criteria in order to increase the precision of final risks evaluating. Therefore, new criteria are presented in this paper which are based on developing risk concepts for more precise risk analysis: 1- Probability criterion: Risk probability assessment investigates the likelihood that each specific risk will occur. - Impact criterion: Risk impact assessment investigates the potential effect on a project objective such as time, cost, scope, or quality, including both negative effects for threats and positive effects for opportunities. 3- Quickness of Reaction toward risk criterion: The duration takes until organization responses to occurred event. 4- Event measure quantity criterion: Expected required resources for settling the event in an appropriate time and with standard equipment. 5- Event ability criterion: This criterion is defined as follows: Event ability = Threats value / opportunities value. Threats value is risk time expected value or risk cost expected value after risk occurrence and opportunities value is risk time expected value or risk benefit expected value in case of not happening risk. Considering all different concepts of criteria which are defined above, no specific relation can be shaped in decision maker's mind based on relationships between criteria, therefore, there is independency between criteria. B. Fuzzy Theory Importance in Risk Evaluating The project risk experts or decision makers (DMs) can provide a precise numerical value, a range of numerical values, a linguistic term or a fuzzy number. So, fuzzy linguistic terms are much easier to be accepted and adopted by the DMs to provide precise numerical judgments about the criteria of each risk event. Therefore a linguistic term or a fuzzy number can be used in the proposed model. C. Fuzzy Membership Function Among the commonly used fuzzy numbers, triangular fuzzy numbers are likely to be the most adoptive ones due to their simplicity in modeling and easy of interpretation. We feel that a triangular fuzzy number can adequately 306

3 represent the seven level fuzzy linguistic variables and thus, is used for the analysis hereafter. Based on these assumptions, Table I shows the linguistic terms defined for the criteria of project risk event in this paper. TABLE I THE RELATIONS BETWEEN LINGUISTIC VARIABLES AND TRIANGULAR FUZZY NUMBERS Linguistic variables Very Low (VL) Low (L) Medium Low (ML) Medium (M) Medium High (MH) High (H) Very High (VH) Triangular fuzzy numbers (0, 0, 0.1) (0, 0.1, 0.3) (0.1, 0.3, 0.5) (0.3, 0.5, 0.7) (0.5, 0.7, 0.9) (0.7, 0.9, 1) (0.9, 1, 1) The proposed model contains fundamental segments which are illustrated in Fig.1. Project risk planning and identification are focused in first segment, output of this segment is Risk Breakdown Structure (RBS). In second segment, fuzzy evaluating for high risks are determined by taking advantage of fuzzy MADM. Details of the proposed case and application of model are discussed respectively in the following sections. Finally, their advantages and disadvantages these methods are compared in Section IV. E. IV. CASE STUDY (CONSTRUCTION INDUSTRY) A. Purpose The purposes of this case study are: 1- Identify sources of risk, uncertainty and the significant risks for construction industry projects. - Design a project risk evaluating model for the significant risks using fuzzy MADM. 3- Compare the modeling mechanisms of the two methods of Fuzzy MADM and their performances in modeling a set of risk data for the mentioned project. B. Project Definition South Pars gas field in one of the largest independent gas reservoirs in the world situated within the territorial waters between Iran and the state of Qatar in the Persian Gulf. It is one of the country s main energy resources. This gas field covers an area of 9700 square kilometers, of which 3700 square kilometers belongs to Iran. Fig. 1. A Project Risk Evaluating Model in Fuzzy Environment The Iranian Portion is estimated to contain some 14 Trillion Cubic Meter (TCM) of gas reserves and some 18 billion barrels of gas condensates [15]. The contract type of above mentioned project is MEPCC, which includes Management, Engineering, Procurement, Construction and Commissioning. C. Scope and Project Risks According to mentioned description in section IV.B., this project includes uncertainties and risks that create opportunities and threats based on predefined objects and stakeholder's aims. 307

4 WBS Level 0 WBS Level 1 The initial Risks 1- Project management disabilities - Lack of attention to law and regulations 3- Economical inflation Management - 4- Fluctuating currencies exchange rate 5- Increase in international crude oil price 6- Lack of attention to contract requirements 7- Communication matters between consortium members 8- Weak client 9- Delay in paying and receiving project's invoices Engineering Basic Design Detail Design 1- Inaccessibility to foreign design consultants 4- Failure in transmitting data from basic design to detail design - Design failures 3- Change in project specifications 5- Lack of expert human resources 6- Lack of design quality Procurement Construction Equipment and Bulk Material Long Lead Items Spare Parts Site Preparation 1- International relations 4- Incorrect long lead item time schedule 1- Soil and site bed problems - Ambiguity in project cash injection 5- Imperfect data transmission to vendors - Unsuitable weather conditions Camp Construction 4- HSE matters 5- Workers riots Site Establishment 7- Change in construction scope of work 8- Lack of experienced workers 3- Inappropriate vendor list 6- Lack of experience in inspection and forwarding 3- Heavy lifting matters 6- Lack of communication between central office and site office 9- Contagious diseases Plant Construction 10- Subcontractor interferences 1- Delay in equipment delivery to site 10- Delay in paying subcontractors invoices 11- Deficiency in QA/QC inspections and audits Commissioning Pre-commissioning Commissioning 1- Non-consideration of precommissioning requirements - Lack of pre-commissioning materials quality 3- Non-consideration to commissioning procedures Fig.. Risk Breakdown Structure Information is used in the MEPCC project risk identification included key project documents, such as the project WBS, project execution strategy, project charter, cost and schedule assumptions, scope definitions, engineering designs and studies, structured questionnaire, economic analysis and any other relevant documentation about the project and its purpose. Other information such as historical data, theoretical analysis, informed opinions of experts and the concerns of stakeholders also are used. In addition to mentioned risk identification methods, some other techniques based on group decision methods shall be focused remarkably in construction projects. These methods include; Brainstorming, Pin card, Gallery, Battle- Belmuden-Brainwriting (BBB), Collective Note Book (CNB) and Nominal Group Technique (NGT) among others [16]. The proposed risk Structure shows the risk groups, risk categories and risk events at the lowest level. Project risks are divided into five groups, Management, Engineering, Procurement, Construction and Commissioning. D. Project Risk Evaluating Application Project risk evaluating is considered based on presented model in section III. The first segment of proposed model is the identification of risks that are effective on activities explored during the life cycle of project (See Fig..). Then in following segment, significant risks and determined criteria are used as fuzzy MADM input data. Ten top risks which they could crucial impacts on project's objectives have been chosen by considering experts' opinions, and other risks with minor impacts and low probability of occurrence have not been considered in calculation process. Decision matrix is Iranian oil and gas industry experts' data which is derived by using fuzzy Delphi method, mentioned matrix is shown in Table. II. The decision makers provide their preferences between final risks in fuzzy LINMAP method as follow: S ={(8,1),(8,),(8,3),(8,4),(8,5),(8,6),(8,7), (8,10),(1,3),(1,4),(1,5),(1,7),(1,10),(7,5),(7,6), (7,10),(3,4),(3,5),(3,10),(9,),(9,4),(9,5),(9,6), (9,10),(4,5),(4,6),(4,10),(,5),(,6),(,10),(10,5),(6,5)} (3) Calculation results of two fuzzy MADM methods are shown in Table. III. E. Discussion of Results In this section, we compare the merits and demerits of 308

5 TABLE II FUZZY DECISION MATRIX Final Risks Description Probability Impact Quickness of Reaction toward risk Event measure quantity Event ability A 1 International relations VH VH H M VH / VH A Design failures M H H H H / MH A 3 Communication matters between consortium members H M L MH M / VH A 4 Delay in paying and receiving project's invoices MH H ML L H / H A 5 Change in construction scope of work M ML H ML ML/ M A 6 HSE matters ML M MH VH M / M A 7 Delay in equipment delivery to site H H MH ML H/ H A 8 Economical inflation VH MH M H MH / M A 9 Lack of attention to contract requirements ML VH VL MH VH / MH A 10 Ambiguity in project cash injection MH MH VH M MH/ H Weight H VH M MH ML TABLE III FUZZY TOPSIS AND FUZZY LINMAP RESULTS Fuzzy TOPSIS Fuzzy LINMAP High Risks D j D Final Risk Final Risk j CC` i CC` Ranking i High Risks S i Ranking S i A A A A A A A 0.58 A A A A A A A A A A A A A 0.58 A A A A A A A A A A A A A A A A A A A A fuzzy TOPSIS and fuzzy LINMAP methods and their differences. These methods can be utilized to model project risk evaluating. Each of them has its advantages and disadvantages and none of them can always perform better than the other one in any situations. Computational result shows that fuzzy TOPSIS distributes alternatives (risks) between PIS and NIS and measures the situation of each alternative toward PIS and NIS, whereas fuzzy LINMAP evaluates alternatives with only one ideal solution, therefore alternatives are distributed in limited range. Fuzzy LINMAP shows the distance of alternatives (risks) to ideal solution closer than what they are in comparison with fuzzy TOPSIS. In spite of this, fuzzy LINMAP'S advantages toward fuzzy TOPSIS are as following: 1- Criteria's weights are input data for fuzzy TOPSIS. If DM makes a mistake in estimating criteria's weight, fuzzy TOPSIS output will change impressively, whereas criteria's weights are generated from the fuzzy LINMAP output and they do not depend on DMs opinion. - Pair wise comparison in S set does not need to be transitive in fuzzy LINMAP method. 3- If alternatives be more than criteria, result of Linear Programming (LP) will be more precise. 4- Same ranking risks can be identified and grouped from fuzzy LINMAP output, it makes response planning easy to deal with risks in case of risks numerousness. For more description, after grouping risks, we can determine each risk level change by choosing a couple of criteria and a couple of risk response resources with allocating required resources. For instance, Δ R, ΔR 1 shall be assigned for changing the risk level from very high to medium high in fuzzy MADM matrix based on criteria (C1,C) with assuming that there are two resources (R1,R). V. CONCLUSION The main objective of this paper was to understand the significant risks in construction projects in Iran and to 309

6 develop a model to analysis them. Most organizations have limited resources to manage all risks equally on project. To overcome this problem, the organization can assess and prioritize the significant risks of project. The new evaluating model based on fuzzy TOPSIS as rapid MADM and fuzzy LINMAP as exact MADM were presented. Moreover, we presented new effective criteria assigned for risk evaluating in projects. Probability and impact criteria were not sufficient for covering all aspects of project risks solely. Therefore, new criteria demonstrated which were based on developing risk concepts for more precise risk analysis. The other objective of the present research was to illustrate the application of proposed model using a construction project risk data. For this, an construction industry project was selected. At the end of paper their advantages and disadvantages these methods were compared. It is found that fuzzy LINMAP showed the distance of alternatives (risks) to ideal solution closer than what they are in comparison with fuzzy TOPSIS. In spite of this, importance fuzzy LINMAP'S advantages toward fuzzy TOPSIS were as following: Criteria's weights were input data for fuzzy TOPSIS. If DM makes a mistake in estimating criteria's weight, fuzzy TOPSIS output will change drastically, whereas criteria's weights were generated from the fuzzy LINMAP output and they did not depend on DMs opinion. It is also found that If alternatives be more than criteria, result of Linear Programming (LP) will be more precise. Moreover, same ranking risks can be identified and grouped from fuzzy LINMAP output, this causes allocating available resources precisely and appropriately to risks in response stage which is important stage after risk evaluation. ACKNOWLEDGMENT The authors would like to thanks all Iranian oil and gas experts that took apart in our research. [6] V. Carr, J. H. M. Tah, A fuzzy approach to construction project risk assessment and analysis: construction project risk management system, Advances in Engineering Software, vol. 3, pp , 001. [7] I. Dikmen, M. T. Birgonul, S. Han, Using fuzzy risk assessment to rate cost overrun risk in international construction projects, International Journal of Project Management, vol. 5, pp , 007. [8] J. Zeng, M. An, N. J. Smith, Application of a fuzzy based decision making methodology to construction project risk assessment, International Journal of Project Management, vol. 5, pp , 007. [9] C. L. Hwang, K. Yoon, Multiple Attributes Decision Making Methods and Applications, Springer-Verlag, Berlin, [10] V. Srinivasan, A. D. Shocker, Linear programming techniques for multidimensional analysis of preference, Psychometrika, vol. 38, pp , [11] T. Yang, C. Hung, Multiple attribute decision making methods for plant layout design problem, Robotics and Computer-Integrated Manufacturing, vol. 3, no. 1, pp , 007. [1] D. Li, J. B. Yang, Fuzzy linear programming technique for multi attribute group decision making in fuzzy environments, Information Sciences, vol. 158, pp , 004. [13] H. Xia, D. Li, J. Zhou, J. Wang, Fuzzy LINMAP method for multi attribute decision making under fuzzy environments, Journal of Computer and System Sciences, Vol. 7, pp , 006. [14] C. L. Hwang, K. Yoon, Fuzzy Multiple Attribute Decision - Making, Springer Verlag, Berlin Heidberg, 199. [15] [16] A. Makoui, S. M. H. Mojtahedi, S. M. Moosavi, Introducing new and practical risk identification methods in infrastructure projects, in Proc. first international risk cong., Tehran, Iran, 007. REFERENCES [1] S. Ebrahimnejad, S. M. Moosavi, A. Ghorbanikia, Risk identification and assessment in Iran construction supply chain, in Proc. first international risk cong., Tehran, Iran, 007. [] R. Kangari, Construction risk management, Civil Engineering and System, vol. 5, pp , [3] M. Chun, K. Ahn Assessment of the potential application of fuzzy set theory to accident progression event trees with phenomenological uncertainties, Reliability Engineering and System Safety, vol. 37, no. 3, pp. 37 5, 199. [4] J. H. Peak, Y. W. Lee, J. H. Ock, Pricing construction risk - fuzzy set application, ASCE J Construction Engineering Management, vol. 119, no.4, pp , [5] T. Ross, S. Donald, A fuzzy multi-objective approach to risk management, in Proc. second Computing in civil engineeering congress, vol., New York, pp ,