Development of Fuzzy Pattern Recognition Model for Underground Metal Mining Method Selection

Abstract

Selection of underground metal mining method is a crucial task for the mining industry to excavate the ore deposit with proper safety and economy. The objective of the proposed study is to demonstrate the application of a fuzzy pattern recognition model for the decision-making of the most favourable underground metal mining method for a typical ore deposit. The model considers eight factors (shape, depth, dip, rock mass rating [RMR] of ore zone, RMR of footwall, RMR of hanging wall, thickness of the ore body, grade distribution), which influence the mining method, as input variables. The weights of these factors were determined using the analytic hierarchy process (AHP). The study used the pair-wise comparison method to determine the relative membership degrees of qualitative and quantitative criteria as well as weights of the criteria set. The model validation was done with the deposit characteristics of Uranium Corporation of India Limited (UCIL), Tummalapalle mine selected. The weighted distances for easiest to adopt are found to be 0.1436, 0.0230, 0.0497, 0.2085, 0.0952, 0.1228, and 0.1274, respectively, for block caving, sublevel stoping, sublevel caving, room and pillar, shrinkage stoping, cut and fill stoping, and squares set stoping. The results indicate that the room and pillar mining method is having the maximum weighted distance value for the given ore deposit characteristics and thus assigned the first rank. It was observed that the mining method selected using fuzzy pattern recognition model and the actual mining method adopted to extract the ore deposit are the same.