Answers are grounded in this paper and cite their sources.
Although a wealth of research has examined the built environment (BE)-travel association, evidence on shared electric-bike (e-bike)–bus transfers remains limited. To address this gap, this study applies an integrated framework combining automated machine learning and GeoShapley to analyze the nonlinear associations between the BE and potential shared e-bike–bus transfer demand, along with identifying spatial heterogeneity in these associations. The research results indicate that geographic locations, land use mix, and bus route density make the largest contributions to model predictions. The associations between BE variables and potential transfer demand are nonlinear; the predictive contributions of some variables tend to level off at higher values. Additionally, GeoShapley reveals that the predictive contributions of bus route density, land use mix, distance to central business district, and sports and leisure facilities vary across geographic locations. This study provides empirical evidence and planning implications for optimizing urban multimodal transportation systems.
No takes yet. Share an insight, caveat, or question.
Yin et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: