A novel interval type-2 fuzzy decision model based on two new versions of relative preference relation-based MABAC and WASPAS methods (with an application in aircraft maintenance planning)

Y. Dorfeshan | Seyed Meysam Mousavi

---

URL :   http://research.shahed.ac.ir/WSR/WebPages/Report/PaperView.aspx?PaperID=116893
Date :  2019/07/01
Publish in :    Neural Computing and Applications

Link :  https://link.springer.com/article/10.1007/s00521-019-04184-y
Keywords :

Abstract :

---

In this paper, concerning advantages of relative preference relation (RPR) and interval type-2 fuzzy sets (IT2FSs), a new IT2F-RPR based on multi-attributive border approximation area comparison (MABAC) method is introduced for determining the critical path of production projects under group decision-making process. IT2FSs are more capable than classic fuzzy sets in coping with uncertainty and providing more degree of freedom to model the uncertain conditions. Also, the RPR is better than defuzzification because it does not lose the fuzzy message, and the method avoids pairwise comparisons. In fact, the MABAC method is extended by the RPR for reducing the time complexity. Furthermore, an extended MABAC method is developed under IT2FSs to better address the uncertainty. Weights of decision makers (DMs) are computed by introducing a new IT2F-RPR-based MABAC concept in the proposed new decision model. Also, experts’ opinions are aggregated by using weights of the DMs. Moreover, weights of criteria are determined based on a new extended weighted aggregated sum-product assessment (WASPAS) method by using the DMs or experts’ opinions on the importance of criteria and weights of DMs. As a matter of fact, weights of DMs are determined by a new version of extended MABAC method, and weights of criteria are specified using a new version of WASPAS method for the first time in the literature. Finally, a case study about aircraft maintenance planning is solved to address the calculation process and applicability of the proposed method. Computational results of the presented model are accurate and suitable for real-life situations.