An interactive consensus reaching model with updated weights of clusters in large-scale group decision making. (January 2022)

• Record Type: Journal Article
• Authors: Liao, Huchang; Wu, Zheng; Tang, Ming; Wan, Zhengjun
• Is Part Of: Engineering applications of artificial intelligence. Volume 107(2022)
• Publication Date: 2022-01
• Subjects: Consensus reaching process -- Large-scale group decision making -- Interactive decision making -- Minimum cost consensus -- Modification degree; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2021.104532
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: Consensus reaching process generates a group decision approved by all experts despite their possible divergent preferences. To circumvent a calculated and nominal consensus without exchanges of views, an interactive consensus reaching strategy is necessary even though it may bring costs, especially for large-scale group decision making problems. To allow experts to make cost-effective preference modifications to reach consensus in large-scale group decision making, this study proposes a dynamic interactive consensus reaching model. Firstly, a minimum-cost-consensus model that focuses on clusters in a large-scale context is introduced, in which the unit preference adjustment cost can be determined objectively. Then, we apply the minimum-cost-consensus solution to develop a feedback mechanism to activate discussions between experts and support preference modifications. The modification degree is defined towards each expert cluster to measure the cost performance of a cluster pertaining to the modifications. On this basis, we update the weights of clusters so as to improve the cost performance. An illustrative example about the grading management of high-alert medication is presented. By comparison, the interactive consensus model only costs 10.1 percent more than an automatic consensus model but gets group consensus with the exchanges of expert views.