摘要

This paper aims to propose a new approach to complete the consensus reaching process for multiple attribute group decision making (MAGDM) where the experts' preferences can be expressed by hesitant fuzzy linguistic term sets (HFLTSs). The possibility distribution based approach is used to deal with HFLTSs since it is a natural way to express the possible information contained in an HFLTS. Two kinds of consensus measures are defined: one is based on distances to the collective preference and the other is based on distances between experts. This paper focuses on the first consensus measure and correspondingly, an interactive consensus reaching model is developed. Finally, a numerical example is provided to show the computational steps of the proposed model. The example indicates that only two rounds of interactions are needed in the given case. By comparison with the existing model which was based on the second consensus measure, it is found that the first consensus measure is less restrictive than the second one. In general, the existing model has to conduct more rounds than the proposed one if the same consensus threshold is specified. The two models also differ in the identification rules.