Optimal location planning for battery exchange and charging stations for electric scooters in leisure areas considering stochastic user dwell times

Stochastic factors significantly affect the optimal planning for locating electric scooter charging stations and scheduling battery exchanges in real operations. A plan that does not account for stochastic factors might turn out to be overly optimistic when implemented in real-world operations, where such stochastic disturbances are common. In this study, we apply mathematical programming and network flow techniques to develop a model for determining the optimal locations of battery exchange and charging stations for electric scooters, while accounting for stochastic user dwell times in leisure areas. The proposed stochastic model, designed to address a location-scheduling problem, is formulated as a mixed-integer network flow problem with side constraints. We develop a simulation-based evaluation method and conduct computational tests to assess the model’s performance. Computational tests using real data from a city in Taiwan show that the stochastic model outperforms the deterministic model and has significant potential for practical applications.