Abstract

Introduction: In mining, worker fatigue is a leading cause of both lost time incidents and decreased per-hour productivity. Scientific and technological innovation in the field of fatigue risk management has evolved in recent years, with approaches in the field transitioning from purely “reactive” methods of detecting fatigue to “predictive” ones. Predictive fatigue technology projects the likelihood of an operator’s fatigue prior to the start of a shift, providing an individualized risk forecast for each hour ahead. With the increased adoption of these technologies throughout the mining sector, new best practices are emerging for how to leverage these data to achieve meaningful safety and productivity improvements. Two key practices have emerged: 1) The “dynamic assignment” of rest breaks and critical tasks by supervisors, and 2) the implementation of highly targeted sleep improvement programs. We quantify the benefits of these applications on both productivity and safety metrics, leveraging data from telematics and video systems from two major North American mine sites.

Methodology: Two mine sites using Readi, the predictive fatigue technology solution from Vancouver-based Fatigue Science, agreed to participate in this study. Data collection took place in 3-4 month periods in 2019-2021. At each site, the majority of operators used wrist-worn wearables to capture sleep data. While sleep data remained private, Readi analyzed the data with a validated biomathematical fatigue model, known as SAFTE™, to generate a personalized fatigue prediction for each operator, for each hour he or she was on duty. Fatigue data were made available to supervisors and operators on a daily basis to enable the “dynamic assignment” of breaks and tasks, and to implement highly targeted fatigue reduction strategies. Fatigue data were then analyzed alongside corresponding telematics data for Dig Rate and Spot Time, as well as to in-cab video data from a reactive Hexagon OAS fatigue detection system. Relationships were examined between operator fatigue and their performance along each of these 3 metrics. Finally, a “Fatigue Risk Profile” was constructed for each mine site, revealing the percent of total on-duty time spent at various levels of fatigue. Using Fatigue Risk Profiles in different time periods, we were able to project a quantified impact of each of the two strategies.

Results and Conclusions: Our analysis was able to project significant productivity and safety benefits for a typical mine of 1,000 operators, similar to the mines that participated in the analysis. On an annualized basis, the analysis indicates a ~USD $5,530,000 increase in output per mine site from the strategies. Furthermore, data indicated a 13% reduction in overall lost time incidents (LTIs) and a 26% reduction in LTIs attributable to fatigue. We were able to estimate a $379,000 annual cost reduction associated with these reductions. In total, results indicate an annualized benefit of USD $5,910,000 for a typical mine with the profile used in the analysis. From this work, we conclude a material benefit of the ongoing use of predictive fatigue technology, particularly when paired with practices that engage supervisors and operators in daily workflows.