Abstract

The rapid electrification of the transportation sector is a critical component of global sustainability goals; however, “range anxiety” and the lack of accessible charging infrastructure remain significant barriers to widespread Electric Vehicle (EV) adoption. Strategic placement of charging stations is essential to maximize utilization and ensure economic viability, yet traditional planning methods often lack granular, data-driven insights.

This project presents a comprehensive machine learning framework for identifying optimal EV charging station locations. By leveraging diverse geospatial datasets—including traffic volume, population density, Points of Interest (POI) distribution, and road network connectivity—we engineered a robust feature set to model localized charging demand.

We employed an XGBoost Regressor to predict a continuous demand score for high-resolution grid cells, moving beyond simple classification to quantify the intensity of potential usage. To ensure practical applicability, the raw model predictions were integrated into a Multi-Criteria Decision Making (MCDM) system. This secondary layer weighted the predicted demand against suitability factors such as land availability and urban centrality.

The study successfully identified and ranked the top 300 high-priority locations, primarily clustered in commercial and mixed-use zones with high traffic throughput. The resulting framework provides urban planners and stakeholders with a scalable, evidence-based tool for optimizing infrastructure investment and accelerating the transition to electric mobility.

Chen, Tianqi, and Carlos Guestrin. 2016. “XGBoost: A Scalable Tree Boosting System.” In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–94. ACM.

Guo, Su, and Huiru Zhao. 2016. “Optimal Siting and Sizing of Electric Vehicle Charging Stations: A Review.” International Journal of Energy Research 40 (10): 102–20.

Knuth, Donald E. 1984. “Literate Programming.” The Computer Journal 27 (2): 97–111.

McKenna, Russell, Mogens Graf Plessen, and Frances Sprei. 2020. “High-Resolution Geospatial Assessment of Residential Electric Vehicle Charging Demand.” Nature Energy 5 (10): 101–10.

Morro-Mello, I., and A. Padilha-Feltrin. 2021. “Machine Learning Approaches for Electric Vehicle Charging Infrastructure Planning: A Review.” Journal of Energy Storage 36: 102353.

Name, Author, and Co-Author Name. 2023. “HERO: A Hybrid Energy Resource Optimization Dataset for EV Charging Stations.” Journal of Renewable and Sustainable Energy 10 (4): 100–115.

OpenStreetMap contributors. 2025. “OpenStreetMap Data.” https://www.openstreetmap.org.

Pedregosa, F., G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, et al. 2011. “Scikit-Learn: Machine Learning in Python.” Journal of Machine Learning Research 12: 2825–30.

Wu, Yuhua, Tao Zhang, and Rui Zhang. 2019. “Site Selection of EV Charging Stations Using Fuzzy MCDM Based on GIS.” Sustainable Cities and Society 50: 101686.