Benders-and-Price approach for electric vehicle charging station location problem under probabilistic travel range

This paper investigates the optimal locations for refueling stations for electric vehicles. Electric vehicles have been successfully introduced into the market. However, their use seems to be limited to urban transport since recharging facilities are readily available only near home and work. Planning recharging infrastructure for electric vehicles is highly relevant because this will enable longer trips, including inter-state travel which requires multiple battery charges. Among various models to determine optimal locations of recharging stations, a flow refueling location model (FRLM) is considered in this study. It determines locations for recharging stations to maximize the flow that can travel between origin and destination pairs by refueling at built facilities. FRLM is extended by introducing a probabilistic consideration of the travel range which might vary depending on various factors including road conditions. We develop a mixed integer nonlinear programming formulation and propose a Benders-and-Price algorithm by combining the Benders decomposition and column generation to solve the proposed formulation. The proposed algorithm is validated using extensive computational experiments on two transport networks, including a real-life Texas highway network.