Built environment and public electric vehicle charging: an investigation using POI data and computer vision

Abstract

Public EV charging stations (EVCSs) are essential for EV adoption. This study analyzes Seoul’s public EV charging patterns in relation to the urban built environment. Built-environment data were collected from land-use maps, POI data, and panorama images, with computer vision extracting scene features. A spatiotemporal analysis revealed peak afternoon usage, with additional late-evening peaks near mega-retail stores on weekdays. Public EVCSs were used more on weekdays, with central business districts experiencing the highest demand, sometimes nearing overuse. Cluster analysis identified unique built-environment patterns, with high-usage stations having more parking areas. Computer vision detected highways, parking lots, and crosswalks as common surroundings. Outlier analysis highlighted fast chargers in business districts. Findings suggest POI data and computer vision complement each other in assessing built environments, offering insights to optimize public EVCS usage.

Team

Junfeng Jiao , Seung Jun Choi

Acknowledgments

This research was supported by National Science Foundation [NSF-2125858; 2133302; 1952193], the UT Good System Grand Challenge [Good Systems], and the USDOT Cooperative Mobility for Competitive Megaregions University Transportation Center at The University of Texas at Austin [USDOT CM2].

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

The cover image is sourced from Pexels, is free of copyright issues, and can be used for educational purposes. https://help.pexels.com/hc/en-us/articles/360042295174