The dual interactions-driven consensus model: Probing the non-cooperative behavior through order effects

Abstract

An improved consensus mechanism is designed with decision-makers (DMs) to pursue the possibility of working with different partial agreements on the evaluations of alternatives. Existing works focus on static cooperation and non-cooperation behaviors in the consensus-reaching process but ignore the individual dynamic cooperation degree caused by trusted information. Since the cooperation and non-cooperation behaviors of DMs affect each other, a quantum model is developed to manipulate such a phenomenon, with an emphasis on the order effects that occur when DMs contact the interference information with different orderings. Four cases are analyzed to determine how the individual state vectors about cooperation evolve and are structured, and a series of optimization models are built to obtain the updated cooperation degrees of DMs. To reflect the effect of interaction between evaluation criteria and mutual inference between DMs on the consensus, a multi-objective consensus model with a two-fold consensus level based on the Choquet integral is constructed to obtain the optimized modified information. The built models can be applied in decision problems with interactions between criteria and between DMs when individual cooperative attitudes in group negotiations change dynamically due to information interference. In the constructed consensus model, the willingness of individual modification and the satisfaction of each DM with the group consensus are fully integrated. Therefore, it is seen as an expansion of the conventional consensus model in the quantum decision-making environment. Furthermore, result and comparison analysis are performed to illustrate the advantages of the proposed model.