CHARGE-MAP: An integrated framework to study the multicriteria EV charging infrastructure expansion problem

The widespread adoption of electric vehicles (EVs) in recent years has necessitated the development of effective charging infrastructures. However, charging infrastructure expansion is a multifaceted problem that requires careful consideration of the existing infrastructure, spatiotemporal distribution of charging demands, power-grid capacity, and budget constraints. To approach this complex problem, we present charge-map, a data-driven simulation-optimization framework, focused on ensuring meaningful charging experience for individual EV owners. charge-map integrates three modules: an agent-based simulation module that estimates spatiotemporal distribution of charging demands by modeling EV adopter mobility and charging behavior; an optimization module that determines optimal new charging station/charger locations and capacities, while minimizing expected detour distances and wait-times with a limited number of new stations; and a power module that determines how to connect the stations to the power grid while maintaining its stability. Using the state of Virginia (consisting of 95 counties and 38 independent cities) as a case study, our results show that charge-map can meet the demand of [Formula: see text]198,600 predicted EVs with 1,305 new public charging stations and 2,164 new chargers. It reduces average detour distances for charging by 66% and wait-times at stations by 72% compared to the existing infrastructure. Furthermore, transformer capacity requirement analysis reveals that only 1.8% of residential transformers require upgrades, while over 80% of commercial charging locations can be supported with modest transformer infrastructure (25 to 50 kVA). This indicates that targeted investments can facilitate cost-effective EV integration. Consequently, charge-map provides policymakers and urban planners with crucial data-driven insights for effective EV charging infrastructure expansion.