Article Preview
Top1. Introduction
Project procurement method (PPM) describes how the project participants are organized to interact, and how the owner’s goals and objectives are transformed into the finished facilities (Moon et al., 2011; ASCE, 1988; Chen et al., 2011). The PPM affects the objectives of a construction project, which are the schedule, cost, and quality (Chan et al., 2001; Khalil, 2002; Blayse and Manley, 2004; Shane et al., 2013; Mollaoglukorkmaz et al., 2013). There are several PPMs in the construction industry. The most common approaches are design-bid-build (DBB), construction management at risk (CM-at risk), design-build (DB), engineering-procurement-construction (EPC) and integrated project delivery (IPD) (Chen et al., 2010; Shi et al., 2014; Qiang et al., 2015; Li et al., 2015). The PPM features its characteristics and meets different situations and owner’s requirements (Alhazmi and Mccaffer, 2000). It was approved that the appropriate PPM can effectively get excellent project performance (Hong et al., 2008; Ojiako et al., 2008; Oyetunji and Anderson, 2006). Therefore, selecting a suitable PPM for a construction project is one of the vital decision-making issues for the owner in the planning stage.
The PPM selection problem is also called the project delivery system (PDS) in the engineering field. Many researchers have done a lot of work on the selection of PDS (Li et al., 2015; Liu et al., 2015; Konchar and Sanvido, 1998; Yngling and ShuHuiKerh, 2004; Ling and Liu, 2004). The aim of selecting the PDS is to achieve construction project performance better (Konchar and Sanvido, 1998; Yngling and ShuHuiKerh, 2004; Ling and Liu, 2004). As a powerful decision tool, analytical hierarchical process (AHP) was employed for PDS selection (Khalil, 2002; Alhazmi and Mccaffer, 2000; Mahdi and Alreshaid, 2005; Mafakheri et al., 2007). However, AHP has been criticized for its incapability to deal with uncertainty and its lack of sound statistical theory (Belton and Stewart, 2002) adequately. Moreover, multi-attribute utility was also applied to deal with the PDS selection decision making (Chen et al., 2011; Chan et al., 2001; Oyetunji and Anderson, 2006; Love et al., 1998). Case-based reasoning (CBR) is the process of solving new problems based on the solutions of similar past cases, which is suitable for selecting PDS for construction projects (Luu et al., 2003; Ng et al., 2005; Luu et al., 2006; Kumaraswamy and Dissanayaka, 2001). Li et al. (2015) proposed a decision-making model for the selection of PDS based on information entropy and unascertained set. From the perspective of value-added, Wang et al. (2013) have made a comparison to select of PDSs between DB and DBB. Tran and Molenaar (2015) have considered the risk factors and presented a risk-based modeling methodology to the selection of a project delivery method for the highway project. Dai et al. (2016) used a hybrid cross-impact technique for PDS decision-making for the highway project. Nevertheless, some shortcomings should be overcome, such as imprecise of evaluation criteria in nature (Ng et al., 2002). Therefore, the fuzzy set theory is also gradually applied to PDS selection (Ng et al., 2002; Khanzadi et al., 2016; Wang et al., 2014).
The selection of an appropriate PDS for a construction project is a typical multi-attribute decision-making problem under uncertainty(Ibbs et al., 2011), and the evaluation criteria have intense fuzziness (Ng et al., 2002). Many researchers have done much work on decision-making under uncertainty with the fuzzy set (Boran, 2011; Boran et al., 2011; Ashraf et al., 2014; Gupta et al., 2016; Büyüközkan and Güleryüz, 2016; Butt and Akram, 2016a,b; Nguyen, 2016; Habib et al., 2016; Zafar and Akram, 2017; Sarwar and Akram, 2017). Interval-valued intuitionistic fuzzy set (IVIFS) can effectively elucidate the fuzziness and uncertainty of material things (Nguyen, 2016; Atanassov, 1989; Xu, 2007a; Wei et al., 2011; Chen and Huang, 2017). The Spearman rank correlation coefficient (SRCC) is considered as one of the best nonparametric measures of relationship (Dikbas, 2018).SRCC assesses the linear relationships between the ranks of monotonically related variables. Even if the relationship between the variables is not linear. In fact, SRCC had tried to prove that ranks of measurements instead of raw measurements have significant advantages in correlation calculations (Dikbas, 2018).