Multi-period Charging Infrastructure Planning under Uncertainty: Challenges and Opportunities

Long-term charging infrastructure planning is imperative to sustain the rapid adoption of electric vehicles (EVs) in line with climate goals. While the literature on spatial planning of charging infrastructure is well documented, the temporal dimension has received limited attention. This paper comprehensively reviews the literature on multi-period charging infrastructure planning under uncertainty. It examines the complex interplay between EV mobility and the energy sector. Four gaps are identified after examining 44 pertinent studies published from January 1990 to March 2024. Firstly, current models are predominantly deterministic and myopic, lacking a forward-looking approach to accommodate future uncertainties. Secondly, most studies rely on EVs’ aggregated mobility and charging patterns, leading to inaccurate charging demand forecasts and suboptimal plans. Addressing this requires integrating vehicle-level agent-based models that accurately depict EVs’ charging patterns, and their interactions with charging stations and the grid. Thirdly, the impact of improved charging infrastructure on EV adoption is generally ignored. Joint consideration of charging demand forecasting with infrastructure planning is essential to incorporate such infrastructure-demand feedback loops. Lastly, current planning frameworks show limited integration of grid expansion, operations, and renewable energy sources To address these gaps, we propose a dynamic programming-based framework and solution approach to this planning problem.