Spatial heterogeneity analysis on distribution of intra-city public electric vehicle charging points based on multi-scale geographically weighted regression

Abstract

The rapid rollout of electric vehicle (EV) charging infrastructure is critical for enhancing EV penetration and building an efficient e-mobility system. However, research concerning the impact of the built environment on the deployment of public EV charging points (EVCPs) and its spatial variations remains insufficient. To address this gap, an innovative perspective to assess the performance of public EVCP spatial distribution is firstly developed considering service accessibility and capacity. Using multi-source data, including the POI data, second-hand house data and EVCP static data by the end of 2021, multiscale geographically weighted regression (MGWR), GWR and OLS are conducted in the Beijing case study to assess the influence of the built environment and socio-economic characteristics on the distribution of intra-city public EVCPs. Two sets of comparison are conducted, one for EVCPs with different charging powers (AC and DC) and another for those with varying charging capacities, categorized as distributed and centralized EVCPs. MGWR performs better than OLS and GWR statistically with the largest adjusted R2 and lowest AICc and RSS. The results demonstrate an imbalanced and spatially diverse deployment of EVCPs, with DC and distributed EVCPs displaying a clear concentration trend within the densely populated core area. Besides, 9 variables have emerged as statistically significant factors which vary across analysis groups in significance, coefficients, and bandwidths, showing complex interaction mechanism with EVCPs of various locations and attributes. The conclusions provide insights for policymaking aimed at planning and deploying public EVCPs in megacities.