Energy Intensity Mapping of Battery-Electric vs. Diesel Heavy Haulage in Surface Mining: The Interplay of Payload Dynamics and Ambient Temperature

Decarbonizing heavy-duty transport in the mining sector requires a deep understanding of the interplay between specific energy consumption, payload dynamics, and ambient thermal stressors. This study presents an integrated, physics-informed machine learning framework to compare the energy intensity of battery-electric (EV) and diesel internal combustion engine (ICE) tippers on a real quarry route in Poland. We develop a bidirectional, route-aware model using physical force balance and high-resolution elevation data to estimate net energy consumption and regenerative braking potential over a complete closed-loop cycle. Furthermore, an Artificial Intelligence analysis utilizing a Random Forest regressor is implemented to simulate and quantify the non-linear impacts of ambient temperature, haul road rolling resistance, and payload mass on the specific energy intensity ( E s p e c ). Results indicate that while EV energy demand surges in sub-zero climates due to parasitic battery thermal management loads, electric powertrains exhibit a profound thermodynamic advantage during loaded downhill segments, acting as net energy generators via recuperation. The proposed multi-factor approach provides a robust predictive tool for optimizing fleet deployment, infrastructure positioning, and decarbonization pathways in transitionary mining environments.