Improving Payload Efficiency in Open-Pit Mining: An Integrated Model using Six Sigma and Artificial Intelligence

Increasing the payload efficiency in open-pit mining is paramount for productivity, costs, and process variability. In this paper, an integrated methodological approach that combines Six Sigma, the PDCA cycle, and artificial intelligence is proposed to optimise the loading process in a copper mining operation. The goal is to reduce variability and increase control on the loading operation by knowing the variables that relate to the operator training, the adherence to preventive maintenance, and the monitoring of the operation. The proposed methodology was tested on an industrial case involving three truck models and used process capability indicators based on predictive analytics to assess their performance. Results indicate measurable gains in payload efficiency that range from 2.22% to 5.25% for three of the four truck models, despite the presence of a high-performance level already. Availability increased from 88% to 94%, and null payloads dropped, which improved the data reliability when making decisions. Overall, the findings demonstrate that the integration of established process improvement methodologies with artificial intelsligence provides for more rigorous and repeatable control of the loading operation than the two on their own. Although mined at a single site, the proposed framework lends itself to scalability and can also be implemented in other mines. This research contributes to the advancement of data-driven process optimization in mining by providing a robust and transferable model for enhancing payload efficiency and operational performance.