Group decision technique for multiple criteria evaluation problems: the preferential difference and rank approach through data envelopment analysis

摘要

In a multiple criteria evaluation problem, ranking or selecting the alternatives is usually carried out using analytic hierarchy process (AHP) due to its simplicity and popularity. When applied in a group environment, computation of AHP becomes more difficult and complicated. Although the geometric mean method (GMM) is known to be most common used with AHP for aggregation of judgments of the experts in the group, it may ignore variations among the experts' opinions. This paper proposed a group decision technique based on the preferential difference and rank concept. The proposed method used the preferential difference to accommodate strong opinions of the experts in the group while using the DEA/AR exclusion model to give more priority to the alternatives that were put in high ranks more often. The data set used by previous literature, simulated data sets and the case study of Thailand sugar industry for identification of the SWOT sub-factors with their priorities were used to test the effectiveness of the proposed method. In this paper, the group decision techniques by Huang et al. (2009), Angiz et al. (2012) and GMM were used as baseline methods to compare against the proposed method.