A novel hybrid framework combining physics-based modeling, statistical techniques, and machine learning has been proposed to optimize the operational reliability of Load Haul Dumper fleets in underground coal mining operations.
In underground mining activities, equipment availability depends directly on the frequency of machine failure and the mean time to repair. The degradation of diverse subsystems, maintenance schedules, maintenance quality, and operating intensity influence equipment failure, preventing static reliability metrics from accurately mapping mean time between failure and mean time to repair.
The research framework considers Load Haul Dumpers as multi-state repairable systems with time-varying failure properties.
The study analyzed empirical operational and maintenance records from Load Haul Dumpers over an evaluation period of 77 months. The underlying failure and repair processes were modeled using non-homogeneous Poisson process models alongside semi-Markov models.
Subsystem criticality across the equipment was evaluated using reliability importance measures. In addition, the probabilistic risk of equipment failure was determined through survival analysis techniques, specifically employing Kaplan-Meier estimation and Cox proportional hazard analysis.
Machine learning models integrated with explainable artificial intelligence capabilities were developed to predict key reliability parameters for the repairable machinery fleets.
Subject of Research: Reliability optimization of underground mining Load Haul Dumper equipment
Article Title: A hybrid physics–statistics–machine learning framework for reliability optimization of repairable Load Haul Dumper equipment
Article References: Siddiqui, M. A. H., Raj, P., Hasan, M. E., & Dewangan, S. (2026). A hybrid physics–statistics–machine learning framework for reliability optimization of repairable Load Haul Dumper equipment. Discover Applied Sciences. https://doi.org/10.1007/s42452-026-09621-x
Image Credits: AI Generated
DOI: 10.1007/s42452-026-09621-x
Keywords: Load Haul Dumper, underground coal mining, equipment reliability, mean time to repair, mean time between failure, multi-state repairable systems, time-varying failure, non-homogeneous Poisson process, semi-Markov models, reliability importance measures, survival analysis, Kaplan-Meier estimation, Cox proportional hazard analysis, explainable AI, machine learning, maintenance scheduling, subsystem degradation
News Source: Denise Maddox. (November 13, 2026). Hybrid Framework Proposed for Reliability Optimization of Load Haul Dumpers. Scienmag.



