Operational Factor Analysis for Predictive Maintenance in Mining Fleets

Abstract

Technical failures in truck operations within the mining industry can lead to significant profit losses and disruptions across interconnected processes such as hauling, processing, storage, and material transportation. These disruptions can lead to reduced efficiency and increased costs. Hence, a comprehensive understanding of the critical operational factors that contribute to technical failures and disruptions is essential for reducing downtime, improving reliability, and optimizing maintenance schedules. Several studies have highlighted the importance of understanding machine failure patterns and operational disruptions in industrial contexts, particularly in high-cost sectors like mining. Most existing research focuses on individual component reliability or general maintenance strategies, but few works examine the interplay between operational factors and maintenance scheduling, in a holistic manner. For example, studies on condition-based maintenance (CBM) and predictive maintenance (PdM) emphasize the role of real-time monitoring of single-component wear but often fail to address the role of broader operational factors—such as fleet utilization and environmental conditions—in predicting and preventing breakdowns. This study bridges this gap by analyzing performance metrics, event logs and failure data to pinpoint high-risk components based on frequency of failure and severity. Following this, a correlation analysis is performed on real-world operation and utilization data to identify which factors contribute to specific failures. The findings are presented through a correlation matrix heatmap that visualizes these relationships and provides actionable insights. The insights from this study can assist mining companies in refining both long-term and short-term maintenance schedules, minimizing unplanned downtimes and improving operational efficiency.