Trust evolution based minimum adjustment consensus framework with dynamic limited compromise behavior for probabilistic linguistic large scale group decision-making

To eliminate discrepancies among decision makers' (DMs') evaluations, consensus reaching process (CRP) is important in large scale group decision-making (GDM). In CRP, trust relationship can promote the decision group to reach consensus and limited compromise behavior can preserve the original information. However, these two key factors are constant in most of existing studies, which cannot reflect the actual situations. Therefore, this paper explores trust evolution and dynamic limited compromise behavior in CRP for probabilistic linguistic large scale GDM. Firstly, a novel clustering method is proposed to cluster DMs into several subgroups. According to trust degrees among DMs in different subgroups and that in same subgroups, trust degrees among subgroups and confidence degrees of subgroups are defined, respectively. Then, trust evolution model and dynamic limited compromise behavior model are constructed. The trust degree in the next round of adjustment is related to the current trust degree and change of similarity degree, while limited compromise behavior is related to trust degree and confidence degree. Thus, the trust evolution based minimum adjustment consensus framework with dynamic limited compromise behavior is proposed. Finally, the proposed method is applied to an actual example and comparison analyses demonstrate the practicability and superiority of this method.