Dynamic trust network-driven consensus modeling with endogenous adjustment and exogenous modification under a quasi-Z-number environment

Abstract

The Z-number is widely used in decision-making because of its unique dual ability to capture human knowledge and facilitate judgments on realistic issues. To improve the applicability of Z-numbers, this paper puts forward the concept of quasi-Z-number and studies the social network and consensus models with quasi-Z-information to address multi-attribute group decision-making problems. First, a dynamic trust network among decision makers (DMs) is constructed. The endogenous adjustment mechanism is presented due to the interaction among DMs. Second, a feedback mechanism for non-consensus opinions is established. In the feedback, an exogenous modification mechanism is constructed through the minimal modification consensus model to help non-consensus DMs improve their consensus level. Furthermore, a minimum tolerance increment consensus model is proposed to cope with the situation where the consensus threshold is not reached within a specified number of iterations. Third, the practicability of the proposed decision-making system is illustrated by the case of a green energy investment in which all technicalities are shown. Its comprehensive performance is exhibited in the sensitivity analysis. The validity of the information form and methodology is demonstrated in the comparative analysis.