Opinion Quality Dynamic Management and Consensus Model with Quality Threshold for Group Decision Making

In group decision making (GDM), experts from a variety of fields collaborate to select the best alternative. Due to external influences or a lack of sufficient knowledge, experts may sometimes offer low-quality opinions on alternatives. In existing GDM problems, the opinion quality and the consensus with a quality threshold have never been explored simultaneously. To fill this gap, this paper proposes a novel GDM framework integrating opinion quality dynamic management and an improved minimum cost consensus model (MCCM) with a quality threshold in GDM. Firstly, opinion quality dynamic evaluations and management mechanisms are designed to improve the opinions of experts to some extent. Afterwards, the weights of the experts are determined by combining their social reputation and opinion quality. Furthermore, the impact of opinion quality is considered in the consensus, and an improved MCCM with a quality threshold is proposed to promote the consensus. A case study on selecting AI enterprises for an investment is provided to verify the applicability of the proposed opinion-quality-based GDM. Ultimately, the quantitative results show that the proposed model achieves a consensus cost of 411, which is 67.5% lower than the benchmark method M2. The proposed GDM framework only requires two iterations and satisfies the predefined opinion quality threshold and consensus level. The optimal alternative remains stable under various parameter settings, verifying the robustness and superiority of the proposed model.