Exploring the temporal correlations of factors affecting traffic safety on mountain freeways: Through new crash frequency modelling methods
PLOS ONE · 08.04.2025 · 20 (4) · e0319831
DOI: 10.1371/journal.pone.0319831 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 08.04.2025. Verfasst von Liang Zhang und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Traffic and Transportation Engineering, Changsha University of Science and Technology, Changsha, China. Hauptthemen laut MEDLINE: Accidents, Traffic, Safety, Models, Theoretical. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Hunan Province, Research Fund of Hunan Provincial Department of Education.
Abstract (Originalsprache)
The potential factors contributing to safety risks on mountainous freeways exhibit significant seasonal clustering and temporal correlations. However, these temporal characteristics have not been accurately captured by existing crash modeling methods, which severely compromise model fit and may lead to erroneous conclusions. This study makes three major contributions. Firstly, a multidimensional crash dataset involving design features, traffic conditions, pavement performance, and weather conditions was established based on eight quarterly datasets of mountain freeways in China. Secondly, two new crash modeling methods considering temporal correlations were proposed. The first model embedded an autoregressive structure and a time linear trend function within a Poisson model, while the second model incorporated an autoregressive structure and time-varying regression coefficients within a Poisson model. The superiority of the new models over seven existing time-correlated models was validated in terms of goodness-of-fit and prediction accuracy, and the significant associations between crash frequencies across different quarters were also confirmed. Moreover, this study quantitatively analyzed the causes of crash frequency on mountainous freeways in China, revealing several significant conclusions. For instance, special road sections such as interchanges, tunnels, and service areas exhibit higher crash risks. Increased traffic volumes, especially with a higher proportion of trucks, are associated with elevated crash risks. Enhancing pavement smoothness and skid resistance was found to effectively mitigate crashes. Moderate rainfall increases crash risks, whereas heavy rainfall alters travel plans and paradoxically reduces crash frequency. To the best of our knowledge, this study introduced the first temporal correlation modeling method specifically addressing the unique temporal characteristics of safety-influencing factors on China’s mountainous freeways, offering valuable insights for the development of effective safety countermeasures.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- School of Traffic and Transportation Engineering, Changsha University of Science and Technology, Changsha, China.
- Civil Engineering, Hunan City College, Yiyang, China.
ORCID-Profile: Zhongxiang Huang
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Hunan Province
- Research Fund of Hunan Provincial Department of Education
Zitieren
APA
Zhang, L., Huang, Z., Kuang, A., Yu, J., Zhu, L., & Yang, S. (2025). Exploring the temporal correlations of factors affecting traffic safety on mountain freeways: Through new crash frequency modelling methods. PLOS ONE, 20(4), e0319831. https://doi.org/10.1371/journal.pone.0319831
Vancouver
Zhang L, Huang Z, Kuang A, Yu J, Zhu L, Yang S. Exploring the temporal correlations of factors affecting traffic safety on mountain freeways: Through new crash frequency modelling methods. PLOS ONE. 2025;20(4):e0319831. doi:10.1371/journal.pone.0319831
BibTeX
@article{Zhang2025,
author = {Zhang, Liang and Huang, Zhongxiang and Kuang, Aiwu and Yu, Jie and Zhu, Lei and Yang, Songtao},
title = {Exploring the temporal correlations of factors affecting traffic safety on mountain freeways: Through new crash frequency modelling methods},
journal = {PLOS ONE},
year = {2025},
volume = {20},
number = {4},
pages = {e0319831},
doi = {10.1371/journal.pone.0319831}
}
EUVIMED