A DATA DRIVEN DECISION MAKING APPROACH FOR LONG-WALL MINING PRODUCTION ENHANCEMENT

Abstract

Machine failures have destructive effects on continuity of operation and lead to production losses in long-wall mines, making proper maintenance scheduling essential. This paper models the reliability of the whole production chain in an Iranian long-wall mine including the drum shearer, Armored Face Conveyor (AFC), hydraulic powered supports, Beam Stage Loader (BSL), and main conveyer belt. Analyzing the computational results and failure frequencies, we rank the critical components and develop a reliability-based preventive maintenance schedule for all equipment. In respect to the data classification, conveyor belt with failure abundance of 41.5 percent is the most critical, while powered supports with the failure abundance of 1.2 percent shows the best performance. Approximately, the reliability of the production process after four hours reaches nearly to zero. Implementing the schedule, computational results suggest an increase of approximately 67.7 percent in the average production per shift.