Abstract
The optimization of haul networks within large-scale mining operations is a critical factor in reducing queue times and fuel consumption, two primary concerns in the industry that significantly impact productivity and cost. This case study describes a research project conducted across four progressive phases aimed at addressing these challenges through a combination of key performance indicators (KPIs), predictive analytics and assisted decision-making.
The initial phase focused on the deployment of data visualization tools to improve the understanding impacts and scheduling of shovel maintenance, with the goal of reducing unscheduled equipment failures. By monitoring KPIs such as Mean Time Between Failures (MTBF) and Mean Time To Repair (MTTR), and planned maintenance compliance the study sought to enhance the reliability of critical mining equipment.
The second phase involved an analysis of historical equipment downtime data. By applying Monte Carlo Simulation, Time Series Analysis, and Survival Analysis, the research aimed to provide predictive insights into future equipment failures, allowing for more informed and strategic decision-making regarding the impacts of equipment health and maintenance schedules.
In the proposed future work, Queue Prediction methodologies will link Fleet Management System (FMS) data with environmental variables, production constraints, and a variety of other variables to better understand and forecast the factors contributing to increased queue times. Here, Graph Neural Networks (GNN) offer a promising avenue for simulating and predicting the future state of mining haul networks. And finally, Short Interval Control will develop a system to enhance shovel productivity by providing dynamic goal setting, managing real-time bottlenecks, offering supervisor guidance, and facilitating response tracking. The objective is to create a responsive system that minimizes disruptions and aligns with short-term operational objectives.
Collectively, these phases represent an integrated approach to improving decision-making processes, enhancing productivity, and ensuring the highest standards of operational efficiency and equipment readiness in mining operations.