Multi-Objective Simulation-Based Optimization of Truck-Shovel Systems in Open-Pit Mining Using a Hybrid Algorithm

Truck-shovel systems are among the most crucial components in open-pit mining operations, directly influencing operational efficiency, equipment utilization, and transportation costs. Given the multi-objective nature of such systems-balancing production, cost, and resource usage traditional optimization methods often fall short in capturing their dynamic and uncertain behavior. This paper presents a hybrid simulation-based optimization approach that integrates Non-Dominated Sorting Genetic Algorithm II (NSGA-II), Differential Evolution (DE), and Variable Neighborhood Search (VNS). The proposed method is designed to improve the balance between global exploration and local exploitation in solving complex multi-objective problems. A discrete-event simulation model of the truck-shovel system at the Sarcheshmeh copper mine was developed using Arena software to evaluate performance under real-world constraints and stochastic conditions. Experimental results show that the hybrid algorithm outperforms individual metaheuristics in terms of both cost and throughput. Specifically, it achieved a ۱% reduction in monthly transportation costs and a ۱۰% increase in average production throughput. These findings highlight the effectiveness of the proposed hybrid method as a practical decision-support tool for mining operations, enabling more efficient equipment planning and resource allocation under uncertainty.

کلیدواژه ها:

Simulation-Based Optimization ، Truck-Shovel System ، Hybrid Metaheuristic Algorithm