Ant Colony Optimization for the Dynamic Electric Vehicle Routing Problem

Traffic congestion significantly affects the efficiency of electric vehicles (EVs), especially during extended periods of low-speed conditions, which, in this context, will violate battery capacity of the vehicle. This study addresses the dynamic electric vehicle routing problem (DEVRP), focusing on minimizing the impact of traffic congestion. Using the proven adaptation capabilities and behaviors of ant colonies, we applied the ant colony optimization (ACO) approach to improve vehicle performance under dynamic traffic conditions. Specifically, our experimental findings, on a set of benchmark generated test cases, demonstrate the effectiveness of transferring knowledge from previously optimized environments rather than optimizing from ground up. The advantage of ACO in DEVRP highlights the importance of adaptive-learning, knowledge-based, and decision-making in optimizing EV routes, presenting a promising path for future research in intelligent transportation systems.