Influential factors associated with consecutive crash severity: A two-level logistic modeling approach
Artikel i vetenskaplig tidskrift, 2020

A consecutive crash series is composed by a primary crash and one or more subsequent secondary crashes that occur immediately within a certain distance. The crash mechanism of a consecutive crash series is distinctive, as it is different from common primary and secondary crashes mainly caused by queuing effects and chain-reaction crashes that involve multiple collisions in one crash. It commonly affects a large area of road space and possibly causes congestions and significant delays in evacuation and clearance. This study identified the influential factors determining the severity of primary and secondary crashes in a consecutive crash series. Basic, random-effects, random-parameters, and two-level binary logistic regression models were established based on crash data collected on the freeway network of Guizhou Province, China in 2018, of which 349 were identified as consecutive crashes. According to the model performance metrics, the two-level logistic model outperformed the other three models. On the crash level, double-vehicle primary crash had a negative association with the severity of secondary consecutive crashes, and the involvement of trucks in the secondary consecutive crash had a positive contribution to its crash severity. On a road segment level, speed limit, traffic volume, tunnel, and extreme weather conditions such as rainy and cloudy days had positive effects on consecutive crash severity, while the number of lanes was negatively associated with consecutive crash severity. Policy suggestions are made to alleviate the severity of consecutive crashes by reminding the drivers with real-time potential hazards of severe consecutive crashes and providing educative programs to specific groups of drivers.

Multi-level model

Traffic hazard

Unobserved heterogeneity

Consecutive crash

Crash severity

Författare

Fanyu Meng

Southern University of Science and Technology

Pengpeng Xu

The University of Hong Kong

Cancan Song

Tongji University

Kun Gao

Chalmers, Arkitektur och samhällsbyggnadsteknik, Geologi och geoteknik

Zichu Zhou

Tongji University

Lili Yang

Southern University of Science and Technology

International Journal of Environmental Research and Public Health

1661-7827 (ISSN) 1660-4601 (eISSN)

Vol. 17 15 1-16 5623

Ämneskategorier

Transportteknik och logistik

Infrastrukturteknik

Farkostteknik

DOI

10.3390/ijerph17155623

Mer information

Senast uppdaterat

2020-09-01