Citation

Chen H, Zhang H, Tong Z, Jing Y, Zhang L, Liu S, Zhang Y, Chen C, Liu Y. Travel Behav. Soc. 2024; 35: e100743.

Copyright

(Copyright © 2024, Elsevier Publishing)

DOI

10.1016/j.tbs.2024.100743

PMID

unavailable

Abstract

In the context of climate change, transportation resilience to flood threats has received significant attention. However, measures to improve practical resilience have not been addressed and elucidated. Various studies from the domains of disaster and transport have been devoted to commuting vulnerability analysis and transport network robustness assessment, whereas a systematic perspective is rarely used to analyze the damage mechanism of floods to the commuting system. To fill this research gap, we constructed an index system from the dimensions of hazard, transport network condition, and commuting pattern as driving factors of commuting risk during floods. The contribution to commuting loss was first assessed to guide intervention prioritization based on flood modeling and commute simulation. We adopted gradient boosting decision trees to examine the relationship between driving factors from three dimensions and commuting. The results of this study conducted in Wuhan indicate that the contribution rate of commuting pattern is the highest, followed by hazard, and finally road network conditions. The effective thresholds of the driving factors were also identified to guide urban governance. This study is a novel exploration of disaster impact mechanism that can provide new insights for the improvement of commuting resilience practices under climate change.

Language: en

Keywords

Commuting resilience; Influencing mechanism; Location-based service data; Machine learning; Urban flood