CO2 Emission Function Development for Diesel Usage in Gold Mining Haulage Operations

This study develops a CO2 emission function to estimate greenhouse gas emissions from fuel consumption in mine haulage operations, particularly focusing on diesel trucks utilized in small‐scale mining (SSM) haulage with a limited technological level. The methodology, validated in a Brazilian gold mine, integrates real‐time monitoring with accurate emissions estimation, facilitating achievable reduction targets and promoting sustainable mining practices. The analysis indicated an estimation of 510 t CO2e based on a year of operational data, in contrast to the 523 t CO2e formally reported by the company, achieving an accuracy of 97% per the mine's Annual Sustainability Report. Through multivariate analysis, key variables impacting CO2 emissions were identified, and predictive modeling techniques, including partial least squares regression (PLSR) and random forest (RF), were utilized to assess the accuracy and robustness of the function. The PLSR model demonstrated appropriate performance under low‐ and high‐variability conditions, achieving a root mean square error (RMSE) of 0.02 and 0.07, a mean absolute percentage error (MAPE) of 0.24% and 0.47%, and calibration accuracy of 94.5% and 92.2%, respectively, in emissions prediction. The sensitivity analysis conducted on both models revealed that operator efficiency, fleet management, and material management are critical factors in achieving emission reductions of 15%–20%, 25%, and 30%, respectively. This method highlights its capacity for broad implementation, enabling SSM operations to play a crucial role in global efforts to combat climate change.