Integrating experts' weights generated dynamically into the consensus reaching process and its applications in managing non-cooperative behaviors

The consensus reaching process (CRP) is a dynamic and iterative process for improving the consensus level among experts in group decision making. A large number of non-cooperative behaviors exist in the CRP. For example, some experts will express their opinions dishonestly or refuse to change their opinions to further their own interests. In this study, we propose a novel consensus framework for managing non-cooperative behaviors. In the proposed framework, a self-management mechanism to generate experts' weights dynamically is presented and then integrated into the CRP. This self-management mechanism is based on multi-attribute mutual evaluation matrices (MMEMs). During the CRP, the experts can provide and update their MMEMs regarding the experts' performances (e.g., professional skill, cooperation, and fairness), and the experts' weights are dynamically derived from the MMEMs. Detailed simulation experiments and comparison analysis are presented to justify the validity of the proposed consensus framework in managing the non-cooperative behaviors.