Predictive Risk Modeling of Haulage Fleet Automation in Quarrying: From Historical Baselines to Autonomous Electric Emerging Hazards

The deployment of autonomous haulage systems and battery electric vehicles represents a promising pathway to decarbonizing the extractive industry while mitigating severe occupational hazards. This study investigates a 40-year historical dataset (1984–2024) from the U.S. Mine Safety and Health Administration across the stone and sand and gravel sectors to establish baseline injury trajectories and evaluate the safety impacts of fleet electrification and automation. Predictive and probabilistic modeling shows that removing onboard drivers achieves a net fatal risk reduction between 58% and 92%, even after incorporating emerging high-power battery charging hazards. However, haulage automation reconfigures shift dynamics without inherently eliminating ground-level hazards; residual non-fatal frequency rates depend critically on daily task reallocation across manual handling and auxiliary ground logistics. Furthermore, semi-quantitative assessment confirms a critical risk migration toward high-voltage maintenance (R = 12) and confined underground charging (R = 12), where thermal runaway off-gassing elevates explosion risks. Maximizing the safety benefits of autonomous electric haulage requires shifting safety governance from driver training to advanced high-voltage protocols, sensor maintenance and dedicated ventilation standards.