Developing a framework for resilience assessment of electric vehicle charging station networks

Abstract

The growing adoption of electric vehicles (EVs) demands a resilient charging infrastructure for supporting sustained usage. This study introduces a framework for generating EV charging station (EVCS) networks using weighted Voronoi diagrams. EV Charging Attractiveness Potential (ECAP) metric, reflecting varying levels of EVCS attractiveness, is developed considering visits to commercial points of interest, daily trips per person, direct current fast charger equivalents, and EV penetration rates. The resilience of EVCS networks across twelve U.S. states is assessed by analyzing relative global efficiency under random and targeted node removal strategies, including single and multiple node removals. Results indicate higher resilience against random versus targeted attacks and single versus simultaneous node removals. Using a 0.2 relative global efficiency threshold, Florida, New York, and Washington are most vulnerable, reaching the threshold with fewer than 10% of stations removed, while Colorado, North Carolina, Oregon, and Texas collapse with 10%–15% removed, and Indiana, Maine, Minnesota, and Wisconsin remain resilient until more than 15% are removed. Notably, the most critical stations do not necessarily align with areas of high EV charging demand, indicating that resilience is influenced by more than demand alone. Furthermore, the distribution of critical stations varies across different community types (i.e., city, suburban, town, and rural), highlighting the importance of local socio-demographic and geographic factors in shaping network resilience. These findings provide valuable insights for policymakers and stakeholders, underscoring the need for strategic planning and deployment of EVCS infrastructure to enhance resilience and ensure reliable access to charging facilities across diverse regions.