A Fuzzy Multi-Criteria Decision-Making Framework for Optimal Mining Method Selection in Complex Geological Conditions

Mining Method Selection (MMS) is a complex, multi-criteria decision-making problem influenced by geological, technical, economic, environmental, and safety factors. Traditional methods, such as Nicholas, University of British Columbia (UBC), and reformed UBC approaches, rely on expert judgment and fixed decision matrices, often failing to account for uncertainty, interdependencies, and conflicting criteria. To address these limitations, this study proposes the Flexible Mining Method Selection (FMMS) framework, which integrates Fuzzy Entropy (FE), Fuzzy Analytic Network Process (FANP), and Modified Fuzzy TOPSIS. FMMS dynamically adjusts criteria weights based on entropy-driven uncertainty measures, enabling adaptive, data-driven decisions while capturing interdependencies among selection criteria. Applied to the Gol Gohar iron ore deposit in Iran, FMMS identified Top Slicing as the most reliable alternative, achieving the highest crisp score (0.1066) with the lowest variance (0.000052) and standard deviation (0.0063), compared to the lowest-ranked score of 0.0932 for Stopping and Pillar. This contrasts with conventional Nicholas, UBC, and reformed UBC evaluations, which favored Cut and Fill but produced inconsistent rankings across weighting adjustments. Sensitivity analysis over a weighting parameter range of p = 0–10 confirmed stable rankings under varying weight scenarios, with Block Caving, Sublevel Stoping, and Sublevel Caving showing particularly robust behavior once internal criterion relationships were incorporated. These results demonstrate that FMMS effectively balances objective factors (geology, cost, safety, environmental impact) with subjective expert evaluations, offering a flexible, scalable, and systematic approach that enhances efficiency, accuracy, and sustainability in mining method selection. Furthermore, it lays a foundation for future AI-based decision-support systems in mining.