Influence measure-based large-scale group decision making with linear uncertain preference relations

Abstract

This research focuses on the problem of large-scale group decision-making (LSGDM) based on influence measure under linear uncertain preferences. The value of this research is that it improves the performance of the current clustering algorithms, increases the efficiency of consensus reaching for major decision-making events, thus reducing the cost of feedback adjustments, and at the same time reflecting the risk attitude of the decision-makers (DMs) during situations of uncertainty. First, an approach for measuring the influence of the DMs is presented. Based on the influence measure, a clustering method is proposed to classify the experts into subgroups. Then, a method for determining the weights of the subgroups is developed by combining the influence and credibility of the subgroup leader as well as the intra-subgroup consensus level. Subsequently, a feedback adjustment method is provided to reduce the disagreement among the DMs by considering the unit adjustment cost and the corresponding adjustment willingness of the DMs. Further, based on the trust risk and credibility of the DMs, a dual management mechanism of non-cooperative behavior is established to hasten consensus. Finally, a case study is presented to demonstrate the practicality of the suggested approach, while simulation experiments and comparative analysis are conducted to verify the model.