A Multi-Criteria and AI-Assisted Optimization Framework for EV Charging Station Optimization in Mixed Urban–Rural Contexts

This study develops a multi-criteria and AI-assisted optimization framework that integrates the Analytic Hierarchy Process (AHP), K-means clustering, and Genetic Algorithm (GA) optimization within a Geographic Information System (GIS) environment to optimize electric vehicle (EV) charging station deployment across Abu Dhabi’s urban–rural gradient. The model generates a community-level Spatial Suitability Index (mean = 0.47) based on residential, commercial, and accessibility factors, which inform clustering into five deployment typologies reflecting distinct socio-spatial characteristics. GA-based spatial optimization under two policy pathways, Progressive and Thriving, balances accessibility, grid proximity, and utilization efficiency. Results show that the Thriving scenario achieves approximately 15–20% higher network coverage and equity compared to the Progressive case, demonstrating the value of adaptive, data-driven optimization for mixed urban–rural contexts. The integrated AHP–Clustering–GA approach provides a transferable and scalable blueprint for equitable, low-carbon mobility infrastructure planning in rapidly developing regions.