Predictive Reliability Modelling and Maintenance Optimization of Belt Conveyor Systems in Underground Coal Mines

Belt conveyor systems are vital for efficiency and safety in underground coal mining but remain vulnerable to frequent failures stemming from wear, fatigue, and component misalignment. Using a this dataset comprising breakdown records, time before failure, frequency, repair duration, and operational hour this study applies predictive reliability modelling with Weibull distribution analysis to assess the performance of key components such as belting, motors, gearboxes, rollers, drums, couplings, and structural elements. Reliability parameters including the shape factor (β), scale factor (η), and Mean Time Between Failures (MTBF) were estimated, and preventive maintenance intervals were determined at target reliability thresholds (R* = 0.90, 0.80, 0.70). The analysis shows that predictive maintenance, when scheduled from data set, can reduce downtime by 22–28%, extend MTBF by 15–20%, and considerably improve both safety and system availability. The study demonstrates that incorporating datasets into condition-based monitoring and reliability modelling offers a scalable framework for optimizing maintenance strategies, minimizing costs, and enhancing the dependability of conveyor systems in underground coal mines.带式输送机系统对井下煤矿的生产效率与作业安全至关重要，但仍易因磨损、疲劳及部件错位频繁发生故障。本数据集包含故障记录、故障前时长、故障频次、维修时长与运行时长等数据，本研究基于该数据集采用威布尔分布（Weibull Distribution）分析进行预测性可靠性建模，以评估输送带、电机、齿轮箱、托辊、滚筒、联轴器及结构件等核心部件的运行性能。研究估算了形状因子（β）、尺度因子（η）与平均无故障时间（Mean Time Between Failures, MTBF）等可靠性参数，并基于目标可靠性阈值（R*=0.90、0.80、0.70）确定了预防性维修间隔。分析结果表明，基于本数据集制定的预测性维修方案可将停机时间缩短22%~28%，将平均无故障时间延长15%~20%，并显著提升作业安全性与系统可用率。本研究证实，将数据集纳入状态监测与可靠性建模流程，可为优化井下煤矿带式输送机系统的维修策略、降低运维成本并提升系统可靠性提供可扩展的框架。