Journal
TRANSPORTATION RESEARCH PART D-TRANSPORT AND ENVIRONMENT
Volume 110, Issue -, Pages -Publisher
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.trd.2022.103426
Keywords
Ridesourcing; Metro; Built environment; Nonlinearity; Machine learning; GBDT
Funding
- Research Foundation - Flanders (FWO) [1257022N]
- China Scholarship Council [202008320289]
- Fundamental Research Funds for the Central Universities [2242022R10009]
Ask authors/readers for more resources
This study investigates the complex relationship between the built environment, metro systems, and ridesourcing usage using the gradient boosting decision tree method. The findings reveal different patterns and effects based on different dates and metro origins/destinations, providing valuable insights for optimizing ridesourcing services and formulating appropriate land use planning.
Although ridesourcing has served as an emerging feeder mode to the metro system to solve the first/last-mile issue, research on metro-integrated ridesourcing usage is rather limited. This paper applies a gradient boosting decision tree (GBDT) method to investigate the nonlinear relationship between the built environment and metro-originated and metro-destinated usage, using ridesourcing trip record data. The results show that built environment factors (i.e., density, diversity, and destination accessibility) have significant nonlinear and threshold effects on the integrated usage, which differentiate between weekdays and weekends. Different patterns are also observed between metro-originated and metro-destinated usage. Employment density has a more significantly positive effect on weekday metro-originated usage than metro-destinated usage. The distance to metro stations does exist an effective range. These findings could help not only transportation network companies optimize ridesourcing services but also transportation planners formulate tailor-made land use interventions to facilitate intermodal mobility.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available