A novel multi-criteria sorting method based on the linguistic polyhedral hesitant fuzzy consensus-reaching model

This paper proposes FUCOMSort, a novel consensus-based multi-criteria sorting method that utilizes the Full COnsistency Method (FUCOM) to categorize alternatives into predefined categories. By reducing pairwise comparisons, it significantly improves sorting efficiency and classification model consistency. Specifically, the LPHFSs Dombi Weighted Average (LPHFDWA) operator is introduced to effectively aggregate expert information. In addition, a redefined hybrid centrality measure, combining trust relationships with K-core decomposition, is proposed to evaluate their substantial impact on the classification outcomes. A consensus-reaching process is further constructed using trust networks and regret theory, addressing conflicts in expert evaluations. FUCOM is extended to the LPHFSs environment, incorporating both subjective and objective weights through linear equations. The proposed FUCOMSort method reduces pairwise comparisons to just n-1, thereby substantially improving efficiency in large-scale multi-criteria sorting problems. To validate the proposed method, we applied it to the classification of green patent values and conducted sensitivity analysis and comparative analysis. The analysis results demonstrate that the method possesses strong robustness and effectiveness.