Spatiotemporal planning of electric vehicle charging infrastructure: Demand estimation and grid-aware optimization under uncertainty

Deploying strategically planned charging infrastructure is essential to support the rapid electrification of transportation. This study presents a three-phase framework for planning and deploying electric vehicle (EV) chargers. First, charging demand is forecasted using real-time spatial and temporal data, including traffic and point-of-interest activity. Second, a two-stage stochastic maximum coverage model identifies optimal charging locations and capacities while accounting for demand uncertainty and installation costs. Third, power flow simulations assess the impact of added chargers on grid performance to ensure safe integration within hosting capacity limits. The framework is demonstrated using data from Austin, Texas, by integrating regional transportation activity with a real power grid model. Simulation results compare deployment strategies and assess trade-offs in grid load and service coverage. This integrated approach supports infrastructure planning that balances cost, accessibility, and grid reliability in growing urban EV networks.