Application of multi-criteria decision analysis for optimal siting of electric vehicle charging stations in Oklahoma

Abstract

The rapid adoption of electric vehicles (EVs) necessitates a well-planned and resilient charging infrastructure to support widespread accessibility. This study presents a systematic, data-driven framework for optimal electric vehicle charging station (EVCS) site selection by integrating multi-criteria decision analysis (AHP & Fuzzy-AHP), spatial optimization using Voronoi diagrams, and sensitivity analysis. Key decision factors include accessibility, vehicle ownership density, proximity to points of interest and resilience to environmental hazards such as floods and wildfires. AHP and Fuzzy-AHP suitability maps are compared to identify areas of agreement and uncertainty, leading to an integrated suitability map that enhances decision-making confidence. A Voronoi-based spatial optimization approach is then employed to iteratively refine EVCS placement, ensuring fair coverage and minimizing redundancy. Results indicate that urban EVCS should be spaced every 6-10 km apart to accommodate high trip frequency, while rural areas require stations every 15-20 km, with additional placements at highway junctions to prevent accessibility gaps. The study also highlights the importance of a location specific charging mix, recommending Level 2 chargers for residential and workplace settings and DC Fast Chargers for highways and high-traffic areas. Beyond spatial optimization, this study emphasizes the importance of an adaptive, phased deployment strategy, where EVCS expansion is guided by real-world demand rather than static planning assumptions. The proposed framework is scalable and transferable, providing a replicable methodology for EVCS planning across diverse geographic regions. By bridging the gap between theoretical models and practical deployment, this study contributes to the development of a resilient and demand-responsive EV charging network.