An Overview of Energy Replenishment Strategies for the Electric Vehicle Routing Problem: Models and Solution Algorithms

 The electric vehicle routing problem (EVRP) is constrained by the limited driving range and time-consuming energy replenishment. These characteristics shift the focus of the EVRP from simple path optimization to an integrated optimization of routing and energy replenishment. Consequently, the energy replenishment strategy becomes a critical determinant of the feasibility and economic viability of EVRP solutions. This paper presents a systematic literature review structured around a core classification of replenishment strategies. The strategies are categorized into two primary modes: charging and battery swapping. This framework addresses common gaps in existing research, such as imprecise strategy definitions and fragmented analyses. For the charging strategy, we establish a three-dimensional classification framework, which comprises the charging function, charging policy, and charging station type. Within this context, wireless charging is considered as a special method of energy replenishment. The battery swapping strategy relies on battery swapping stations (BSSs): the EVRP with BSSs (EVRP--BSSs) and the BSS location-routing problem with electric vehicle (BSS--EV--LRP). Our review identifies several limitations in the current body of research. These include an imbalance between modeling accuracy and computational efficiency, insufficient coverage of diverse operational scenarios, and a superficial integration of emerging technologies. Furthermore, many studies lack a multi-stakeholder perspective that considers collaborative solutions. Future research should prioritize addressing these gaps. Key directions include developing effective methods for solving nonlinear charging functions and expanding research into more specialized scenarios. Additionally, there is a need to improve collaborative algorithms for battery swapping and to develop shared BSS models that serve multiple enterprises. The strategy-driven framework proposed here offers a clear reference for modeling and scenario adaptation in future EVRP studies.