Productivity monitoring and improvement through digitalisation of load-haul-dump (LHD) machinery at Mimosa underground operations

Abstract

This study evaluates the impact of digital monitoring on the performance of underground Load-Haul-Dump (LHD) machines in a room-and-pillar mining operation at Mimosa Mine. The research aimed to (i) assess the limitations of the existing manual monitoring system, (ii) quantify the benefits of digitalisation via a Fleet Management System (FMS), and (iii) identify data driven opportunities for productivity improvement. A quantitative approach was applied, integrating real-time telemetry from sensor-enabled LHDs with parallel manual records. Key variables: cycle time, payload, tramming distance, and loading rate were captured and analysed using Python scripting and Stata 15. Results showed that the manual system consistently overstated tonnage and operating time, leading to an underestimation of productivity. These discrepancies were both statistically significant and operationally material. Productivity based on FMS data ranged between 36–37 tonnes per hour (tph), compared to 35.2 tph from manual data and a historical baseline of 34 tph. Although the effect size was small, even a 3% uplift translates into a potential annual net gain exceeding US$1.7 million. Correlation analysis revealed that cycle time had a strong negative relationship with productivity (r = –0.77), tramming distance a moderate negative effect, while payload exhibited only a weak correlation. The loading rate was identified as a more actionable lever for operator performance management. The FMS enabled high-resolution monitoring of variables such as cycle time and tramming distance, uncovering inefficiencies invisible to manual systems. These insights support targeted interventions to reduce tramming distances and cycle times, alongside the introduction of a digital performance management framework for LHD operators.