An Asynchronous Large-Scale Group Decision-Making Method With Punishment of Unstable Opinions and Its Application in Traffic Noise-Control Technologies Selection
Huchang Liao, Xiaowan Jin, Zeshui Xu, Enrique Herrera‐Viedma
Abstract
Traffic noise acts as a major disturbance to residents living in the vicinity of roads, railways, and airports, leading to various health problems to people. Evaluating different noise-control technologies by many experts according to multiple attributes can be seen as a multiattribute large-scale group decision-making (LSGDM) problem. The existing LSGDM models often neglected the time costed by consensus reaching progress, and methods to determine the number of clusters and the influence of unstable opinions of experts are rarely discussed. As experts spend different amounts of time in hesitation and revising their opinions, the adjusted opinions are always fed back to the moderator asynchronously. Considering the hesitancy time of each expert, this study designs a multiattribute LSGDM model with a minimum-time-based consensus mechanism to help select noise-control technologies for practical cases. Additionally, an iterative self-organizing data analysis clustering algorithm is adopted to cluster experts, and a punishment mechanism for experts with unstable opinions is proposed. A numerical example regarding the selection of traffic noise-control technologies for an expressway shows the effectiveness of the proposed model. Comparative analyses verify the advantages of our proposed method in saving the time of opinion adjustment process.