The electric vehicle routing problem of a new mobile charging service

Kanghui Ren,Maosheng Li,Xuekai Cen,Helai Huang

DOI: https://doi.org/10.1093/tse/tdae012

2024-03-13

Transportation Safety and Environment

Abstract:Abstract A novel mobile charging service that utilizes vehicle-to-vehicle (V2V) charging technology has recently been proposed as a supplement to fixed charging infrastructure (CI), enabling electric vehicles (EVs) to exchange electricity. This study formulates a vehicle routing problem (VRP) of vehicle-to-vehicle (V2V) charging, optimizing the routing of DVs to service RVs while taking into account their willingness to join the V2V charging platform. A mixed integer linear programming (MILP) model is established to optimize the VRP-V2V (i.e. the VRP of V2V charging), which is known to be NP-hard. To solve large-scale instances for real-world applications, we propose an adaptive large neighborhood search (ALNS) algorithm, which, when combined with the structure of the VRP-V2V problem, utilizes four local search procedures to enhance solution quality following destroy and repair operators. Results indicate that the proposed ALNS algorithm outperforms the optimization solver CPLEX in small-scale instances, and can solve large-scale instances that are infeasible using CPLEX solver. In a numerical analysis of Changsha's large-scale network, we demonstrate that the V2V platform can save an average of 33.1% on the charging cost of recharging vehicles, hence raising customer satisfaction with charging services and reducing range anxiety. The platform's profitability is also increased by using V2V charging in areas lacking fixed charging infrastructure.

transportation science & technology

What problem does this paper attempt to address?

This paper discusses a novel electric vehicle (EV) charging problem, namely the vehicle-to-vehicle (V2V) charging vehicle routing problem. In situations where the current fixed charging infrastructure cannot meet the charging demand of EVs, V2V charging technology allows EVs to charge each other. The study establishes a mixed integer linear programming (MILP) model to optimize V2V charging vehicle routing problem (VRP-V2V), considering the willingness of vehicle owners to join the V2V charging platform. To solve large-scale instances, they propose an Adaptive Large Neighborhood Search (ALNS) algorithm, which outperforms the commercial solver CPLEX in small-scale instances and can handle large-scale problems that CPLEX cannot solve.

Through numerical analysis, this study shows that the V2V platform can reduce EV charging costs, with an average saving of 33.1%, improve charging service satisfaction, and reduce range anxiety. In addition, V2V charging can increase the profitability of the platform in areas lacking fixed charging infrastructure. The paper also reviews related literature, including mobile charging services, V2V charging technology, and solutions to vehicle routing problems, and points out that existing research mainly focuses on technological feasibility, while the deployment and operational details of V2V charging have not been sufficiently studied.

Overall, the paper aims to address how to effectively utilize V2V charging technology by designing reasonable vehicle scheduling strategies to improve charging efficiency and service quality, while considering the economic and time factors of users choosing the V2V platform.