The transportation sector is a major contributor to greenhouse gas emissions in the United States, motivating the transition toward electric vehicles (EVs). However, maintaining continuous EV operation, particularly during battery disruptions, remains a critical challenge. This study proposes a novel system based on autonomous mobile charging pods (pods) to provide on-demand EV charging as a flexible complement to fixed charging infrastructure. A mixed-integer linear programming (MILP) model is developed to optimize the routing and operation of pods with the objective of minimizing vehicle miles traveled (VMT). To address large-scale networks, a hybrid simulation–optimization framework combining Simulation of Urban Mobility (SUMO) and Adaptive Large Neighborhood Search (ALNS) is employed, while Gurobi validates the model on smaller instances. A comprehensive scenario analysis is conducted, considering variations in customer demand, time window lengths, depot numbers, and charging rates. Results show that increasing the number of depots and adopting Level 3 fast charging substantially improves system efficiency, reducing VMT and the number of required pods. Extending customer time windows further enhances routing flexibility, while platooning opportunities identified through SUMO simulations provide additional travel savings, reducing energy consumption and yielding an average fleet-wide energy saving of roughly 2.5% across all scenarios. Overall, the proposed framework provides insights for policymakers and practitioners to evaluate fleet sizing, routing strategies, and operational efficiency in mobile EV charging systems. • An integrated MILP–SUMO–ALNS mobile EV charging framework. • The ALNS heuristic offers near-optimal solutions and reduces computation time. • Increasing the number of depots reduces both required pods and vehicle miles traveled. • Faster charging further enhances operational efficiency and reduces route requirements. • Additional travel and energy savings are achieved with platooning.
Moghaddam et al. (Mon,) studied this question.