Spatiotemporal associations between the built environment and carbon emission reduction potential of shared on-demand transit
4 affiliations
- 1Center for Intelligent Transportation Systems and Unmanned Aerial Systems Applications Research, State Key Laboratory of Ocean Engineering, School of Naval Architecture, Ocean and Civil Engineering, Shanghai Jiao Tong University
- 2School of Vehicle and Mobility, Tsinghua University
- 3School of Transportation Engineering, Chang'an University
- 4Shanghai Urban-Rural Construction and Transportation Development Research Institute
Abstract
Shared on-demand transit has emerged as a promising mobility solution for reducing transport-related carbon emissions and supporting more sustainable urban environments. However, limited evidence exists on how built environment characteristics are associated with its carbon emission reduction potential across space and time. This study aimed to quantify the maximum carbon emission reduction achievable through ODT and identify the built environment factors associated with its spatial variation. An integer linear programming model combined with a multi-passenger travel optimization framework was developed to estimate achievable carbon emission reductions using travel demand data. An XGBoost model was then used to examine their relationships with built environment characteristics, while SHAP and partial dependence analyses were employed to interpret nonlinear, threshold, and spatially heterogeneous relationships. The results showed that ODT could reduce carbon emissions by up to 17.67% in the morning and 14.48% in the evening. Built environment characteristics exhibited pronounced spatiotemporal heterogeneity, with employment- and education-related land uses showing higher nonlinear thresholds in the evening than in the morning. Local interpretation further revealed substantial spatial variation in feature contributions, with areas surrounding educational facilities, employment centers, and suburban metro stations associated with greater carbon emission reduction potential. These findings provide spatial evidence for targeted ODT implementation and highlight the potential of built environment characteristics to inform low-carbon mobility strategies and sustainable urban transport planning.
Related
- Associations of school food environment characteristics (availability, pricing, and marketing) with adolescents’ dietary intake and anthropometric outcomes
- Associations between community built environment and leisure-time physical activity among older adults: a systematic review and meta-analysis
- Associations Among family environment, self-identity, cognitive emotion regulation strategy use, and physical activity in adolescents: A multi-theory integration perspective
- Shared reading and independent book engagement in early childhood: associations with the home literacy environment in a Turkish pediatric outpatient population
- Household transportation mode choice in the context of emission reduction goals
BibTeX
@article{wei2026spatiotemporal,
title = {{Spatiotemporal associations between the built environment and carbon emission reduction potential of shared on-demand transit}},
author = {Xiaofan Wei and Zhe Zhang and Chao Li and Meigen Xue and Hongdi He},
journal = {Frontiers in Public Health},
year = {2026},
month = oct,
doi = {10.3389/fpubh.2026.1927997},
url = {https://www.frontiersin.org/articles/10.3389/fpubh.2026.1927997/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Spatiotemporal associations between the built environment and carbon emission reduction potential of shared on-demand transit
AU - Xiaofan Wei
AU - Zhe Zhang
AU - Chao Li
AU - Meigen Xue
AU - Hongdi He
PY - 2026
DA - 2026/10/20
JO - Frontiers in Public Health
DO - 10.3389/fpubh.2026.1927997
UR - https://www.frontiersin.org/articles/10.3389/fpubh.2026.1927997/full
AB - Shared on-demand transit has emerged as a promising mobility solution for reducing transport-related carbon emissions and supporting more sustainable urban environments. However, limited evidence exists on how built environment characteristics are associated with its carbon emission reduction potential across space and time. This study aimed to quantify the maximum carbon emission reduction achievable through ODT and identify the built environment factors associated with its spatial variation. An integer linear programming model combined with a multi-passenger travel optimization framework was developed to estimate achievable carbon emission reductions using travel demand data. An XGBoost model was then used to examine their relationships with built environment characteristics, while SHAP and partial dependence analyses were employed to interpret nonlinear, threshold, and spatially heterogeneous relationships. The results showed that ODT could reduce carbon emissions by up to 17.67% in the morning and 14.48% in the evening. Built environment characteristics exhibited pronounced spatiotemporal heterogeneity, with employment- and education-related land uses showing higher nonlinear thresholds in the evening than in the morning. Local interpretation further revealed substantial spatial variation in feature contributions, with areas surrounding educational facilities, employment centers, and suburban metro stations associated with greater carbon emission reduction potential. These findings provide spatial evidence for targeted ODT implementation and highlight the potential of built environment characteristics to inform low-carbon mobility strategies and sustainable urban transport planning.
SN - 2296-2565
ER -