Electric vehicles shared charging and parking: A reservation-based dynamic rolling allocation framework considering unpunctuality

Abstract

Installing charging stations in public parking lots is a widely adopted strategy to meet the growing demand for electric vehicle (EV) charging. However, in high-demand areas, repurposing conventional parking spaces as charging spaces can increase parking cruising time. Additionally, idle periods in charging spaces may reduce overall space utilization efficiency. This paper proposes a reservation-based dynamic rolling allocation framework that incorporates a novel charging space-sharing mechanism, aiming to enhance space utilization and maximize revenue in hybrid parking lots. To account for user unpunctuality, this paper models arrival and leave time uncertainty using a probability distribution and incorporate it into a robust optimization formulation, which is then transformed into a mixed-integer programming (MIP) model. This approach enables the estimation of a theoretical time buffer required to minimize scheduling conflicts between consecutive users, given a predefined maximum probability of parking or charging conflict. A case study based on simulated data derived from large-scale real-world records validates the effectiveness of the proposed model. Results show that the dynamic rolling allocation and sharing mechanisms significantly improve total revenue and reduce user conflicts compared to baseline approaches. Additionally, comprehensive parameter sensitivity analyses offer practical insights for managing hybrid parking lots and optimizing operational performance.