Robust planning for electric vehicle charging stations under congestion

The last decades have witnessed the rise of electric vehicle (EV) sales, accompanied by a growing demand for readily accessible public EV charging facilities. Unlike refueling a fossil fuel vehicle, charging an EV requires significantly more time, which may lead to congestion if the public charging infrastructure is not well-designed. In this paper, we study the strategic planning of public EV charging stations, aiming to place chargers with a limited investment budget to maximize the coverage of uncertain charging demand. To ensure service quality under possible congestion, we introduce two types of chance constraints to mitigate long waiting times and reduce demand loss in situations with limited waiting space. Given the challenges in accurately estimating charging demand and charging time, we apply a robust approach to model this problem with uncertain charging demand arrival and service rates. The robust model is then reformulated into an equivalent mixed integer linear program of moderate size, which is tractable by commercial solvers. A case study based on data from Nanjing demonstrates the effectiveness of the proposed robust approach and provides insights into real-world applications. Extensions with a general charging process and decentralized driver selection of charging stations are also discussed and verified through extensive numerical experiments, which indicates the stable performance of the proposed approach under general settings.