Spatiotemporal clustering analysis of shared electric vehicles based on trajectory data for sustainable urban governance

Abstract

Shared electric vehicles (SEVs) have emerged as a promising mode of transportation with the growth of EVs and the sharing economy. However, current challenges, such as suboptimal station planning and inflexible operation schedules, are limiting the user experience and the growth of the industry. Therefore, our study adopts the affinity propagation (AP) clustering algorithm to divide the stations into two categories: the residential station and the business office station. We then apply the Spatial-Temporal Density-Based Spatial Clustering of Applications with Noise (ST-DBSCAN) algorithm to analyze the spatiotemporal distribution characteristics of SEVs from three aspects: hotspot area, travel volume and travel speed, based on trajectory data from the two types of stations. Results suggest that the distribution hotspot and travel volume of SEV at residential and business office stations is similar during the morning peak, concentrated in office, residential and transportation hub areas, whereas it is obviously different during the evening peak. Unlike the commuting demand of users in residential stations, users in business office stations prefer to use SEVs to travel to leisure and entertainment areas. Furthermore, we found that the driving speed of SEVs is primarily determined by actual road conditions and not significantly affected by station type. These findings provide valuable insights for SEV operators to optimize station layouts and spatial planning of vehicles during peak periods.