Data-driven robust cost consensus model with individual adjustment willingness in group decision-making

Abstract

In the consensus reaching process, the moderator, as the representative of collective interests, needs to consume certain resources to help decision-makers (DMs) reach a consensus. However, in real life, when the adjustment opinions provided by the moderator are too different from the initial opinions of the DMs, they may not accept the adjustment. On the other hand, it is difficult for the moderator to determine the unit consensus cost of every DM. However, existing consensus methods ignore the above two problems. Therefore, a data-driven robust minimum cost consensus model with individual adjustment willingness (DDR-MCCM-IAW) is proposed. First, an IAW function based on the approximation degree between the adjustment opinion and the original opinion is provided to construct the MCCM-IAW. Subsequently, principal component analysis and kernel density estimation are combined to obtain a data-driven uncertainty set of unit consensus cost, to reduce the conservatism of classical robust cost consensus models. Meanwhile, the tractable DDR-MCCM-IAW is obtained by using linear dual theory. Finally, to test the effectiveness and rationality of the proposed method, some experiments and comparative analysis are presented.