A Network-Based Fuzzy MARCOS-based Legal Framework for Dispute-Resolution Selection
Abstract
The selection of a dispute-resolution mechanism represents a complex multi-dimensional optimization problem characterized by uncertainty, conflicting objectives, and strong interdependency among decision variables. This work proposes an intelligent network-based fuzzy optimization framework using the MARCOS algorithm to solve uncertain legal decision problems in the commercial dispute process. The proposed computational framework integrates fuzzy information modeling, criterion interaction network analysis, centrality-driven feature weighting, and fuzzy MARCOS-based alternative optimization to achieve robust decision classification under incomplete and subjective information. Moreover, the developed algorithmic architecture transforms linguistic expert evaluations into fuzzy numerical representations and constructs a weighted criterion interaction network to identify the structural importance of decision attributes. Eight evaluation features are optimized for five dispute-resolution alternatives: negotiation, mediation, arbitration, litigation, and conciliation. Expert knowledge from 16 decision-makers is incorporated to generate the fuzzy decision environment. The computational results demonstrate that outcome predictability (C6) and enforceability (C3) are the most influential optimization features with criterion weights of 0.128 and 0.128, respectively, followed by relationship preservation (0.126). IN addition, the proposed fuzzy MARCOS optimization algorithm identifies arbitration as the optimal solution with a utility value of 0.7851, followed by mediation (0.7357) and litigation (0.7102). Robustness experiments using ±20% weight perturbation confirm complete ranking stability, while expert-level validation achieves the highest mean utility for arbitration (0.722) and a Rank-1 frequency of 25%. In general, the proposed framework provides an explainable intelligent optimization approach for complex decision environments involving uncertain human knowledge.