<![CDATA[Fuel consumption in open-pit mining operations is a significant operational cost, making fuel efficiency an important research topic. This project seeks to investigate the use of adaptive machine learning (ML) methodologies to improve real-time fuel efficiency in mining trucks. A Systematic Literature Review (SLR) was conducted following the PRISMA protocol to examine 47 peer-reviewed articles published from 2015 to 2025. Thematic synthesis and bibliometric analysis identified five dominant categories of machine learning, with deep learning and fuzzy logic being the most common. Many studies have examined adaptive energy regulation for varying terrain and loads; however, only 20% have included driver behavior, highlighting a significant research gap. Reinforcement learning and hybrid systems show significant potential for scheduling and control in dynamic environments; however, they face challenges in real-time applications due to factors such as edge computing and limited data integration. This review describes advances in fuel optimization research through the integration of artificial intelligence, control theory, and mining logistics, and proposes future goals including the development of simplified models for vehicle applications, empirical testing in industrial fleets, and the utilization of behavior and telemetry data to enhance contextual awareness in systems]]>
Copyrights © 2026
