Large-scale group decision-making with non-cooperative behaviors and heterogeneous preferences: An application in financial inclusion

Non-cooperative behavior is a common situation in large-scale group decision-making (LSGDM) problems. In addition, decision makers in LSGDM often use different preference formats to express their opinions, due to their educational backgrounds, knowledge, and experiences. Heterogeneous preference information and non-cooperative behaviors bring challenges to LSGDM. This study develops a consensus reaching model to address heterogeneous LSGDM with non-cooperative behaviors and discuss its application in financial inclusion. Specifically, the cosine similarity degree is introduced to build a distance measure for different preference structures. Clustering analysis is employed to divide large-scale groups and handle non-cooperative behaviors in LSGDM. A consensus degree and a weighting process are proposed to decrease the influence of non-cooperative behaviors and facilitate the consensus reaching process. The convergence of the proposed approach is proven by theoretical and simulation analyses. Experimental studies are carried out to compare the performances of the proposed approach with existing methods. Finally, a real-life example from the targeted poverty reduction project in China is presented to validate the proposed approach. The selection of beneficiaries in finance inclusion is difficult due to the lack of credit history, the large number of participants, and the conflicting views of participants. The results showed that the proposed consensus model can integrate opinions of participants using diverse preference formats and reach an agreement efficiently. (c) 2020TheAuthor(s). Published by Elsevier B.V.

Automated summary

Non-cooperative behavior and heterogeneous preference information are common in large-scale group decision-making problems, posing challenges. A consensus reaching model is developed to address these issues and applied in financial inclusion. The model utilizes cosine similarity degree and clustering analysis to handle diverse preferences and non-cooperative behaviors, achieving efficient consensus.