Unified Granular-number-based AHP-VIKOR multi-criteria decision framework

Abstract

In most of the real-life problems, data and information in a decision-making process are usually imprecise because of the experts’ subjective judgments. The main problem arises when experts, who have different degrees of knowledge about multiple conflicting criteria and alternatives, deal with uncertain data in diverse forms. There are two aims of this paper, the first of which is to define granulation of linguistic information in heterogeneous (fuzzy, rough, interval, or crisp) contexts for group decision-making problems, and thereby its transformation in a homogenous perspective using unified granular number. Second, based on the above, we construct a flexible multi-criteria decision-making (MCDM) framework integrating the Analytic Hierarchy Process and the VIKOR compromise-ranking method in a granular domain, so that we can evaluate the weights of different criteria and hence prioritize alternatives. This methodology can applied in prioritizing risk responses to manage green supply chain risks from the perspective of a plastic manufacturing company. The robustness of our model is monitored by conducting a sensitivity analysis on alternative ranking. In addition, to establish the stability and validity of the ranking result, a comparative analysis of ranking by the proposed method is done with other existing MCDM methods in uncertain domains, namely, through the use of crisp, grey, and rough sets.