Electric vehicle charging station accessibility and land use clustering: A case study of the Chicago region

Land use mixing, balanced land uses, and transportation accessibility have previously been indicated as significant impactors of travel behavior, yet this relationship has not been examined in the EVSE accessibility literature. Using an application of the unsupervised machine learning (ML) clustering algorithm Density-based spatial clustering of applications with noise (DBSCAN), this research identifies 34 areas of spatially clustered level-1, level-2, and DC Fast EVSE charging infrastructure in the Chicago Metropolitan Area. Results indicate that charging access is imbalanced across suburban and urban communities and much of this disparity can be tied to EVSE clustering and its associated land use regimes in the metropolitan area. The majority of EVSE clusters are comprised primarily of level-2 charging and exist in isolated commercial developments to the affluent north and west of the city. Only 26% of clusters are associated with mixed land uses that occur in higher-income dense neighborhoods such as Evanston. Level-3 charging forms a smaller proportion of clustered charging across the region but is primarily unclustered. From a travel behavior perspective, this research highlights a need for a wider abundance of public fast charging options for lower socioeconomic communities and not merely utilitarian charging allocations that perpetuate accessibility for the wealthy.