Exploring spatial heterogeneity of e-scooter’s relationship with ridesourcing using explainable machine learning

Abstract

The expansion of e-scooter sharing system has introduced several novel interactions within the existing transportation system. However, few studies have explored how spatial contexts influence these interactions. To fill this gap, this study explored the spatial heterogeneity in e-scooter’s relationship with ridesourcing using data from Chicago, IL. We developed a Light Gradient Boosting Machine (LightGBM) to estimate e-scooter sharing usage using ridesourcing trips along with associated built environment and socio-demographic variables. The model was interpreted using SHapley Additive exPlanations (SHAP). Results indicated that the threshold effects, where the positive relationship between e-scooter sharing and ridesourcing significantly weakened beyond a certain value, were more pronounced in areas with lower population density, fewer jobs, and fewer young, highly educated population. This is primarily attributed to the limited competitiveness of e-scooter sharing in these areas. These findings can assist cities in harmonizing e-scooter sharing and ridesourcing thus promoting sustainable transportation systems.