A Sequential Three-Way Decision-Based Group Consensus Method With Regret Theory Under Interval Multi-Scale Decision Information Systems

The advancement of computational intelligence has led to the increased application of intelligent methods across various fields, including management, psychology, and machine learning. Despite this progress, a significant challenge in intelligent decision-making research is how to comprehensively account for the influence of irrational behaviors on decision outcomes and effectively coordinate the operations of multiple machines from diverse perspectives. Multi-attribute group consensus research holds valuable insights for addressing macroscopic-level problems, and the consensus reaching process (CRP) is a pivotal component of group decision-making (GDM). In the CRP, decision-makers may need to adjust their individual opinions to enhance group consensus. Meanwhile, multi-scale decision information systems (MSDISs) provide a holistic evaluation of complex issues from various angles. In light of these considerations, this paper introduces a novel approach named CRP-STWD-RT-IMSDIS. This method combines sequential three-way decisions (STWD) with regret theory (RT) within the framework of interval multi-scale decision information systems (IMSDISs). The paper outlines a unique scheme for constructing IMSDISs, determining optimal scale combinations through matrix theory, and utilizing multiple optimal scale combinations as evaluation opinions contributed by different experts to establish multi-expert group consensus. Additionally, the paper introduces a consensus threshold function and outlines a recognition rule based on STWD. Furthermore, it defines personalized dynamic adjustment parameters using RT within the feedback mechanism, effectively addressing the influence of psychological behavior on decision-making. To validate the effectiveness and superiority of the CRP-STWD-RT-IMSDIS method, the paper conducts an experimental analysis using real-world data sets.