Work

Modeling the Interdependence of Vehicle-Level Injury Severities of Bus–Taxi Crashes: A Random-Parameters Bivariate Probit Approach

Year: 2026

Type: article

Abstract: Prior studies have typically analyzed the injury severity of bus or taxi passengers at the crash level or single-vehicle level, neglecting vehicle-level interdependence between them. To address the gap, this research sets out to analyze the factors contributing to the vehicle-level injury severities of transit bus–taxi crashes, with consideration of their interdependence and heterogeneities. The random-parameters bivariate probit model, which can capture both unobserved heterogeneity and within-crash correlation between bus and taxi injury outcomes, was advocated for the joint analysis. In the model, the factors related to the two vehicles and their drivers, together with other factors (e.g., roadway, environment, and crash configuration), were used as the explanatory variables. A total of 3404 two-vehicle bus–taxi crash records in Hong Kong, China, from 2009 to 2019 were used for model estimation. The results indicate that taxi driver age, taxi age, crash location, and collision manner resulted in heterogeneous effects on bus injury severity, and the time of day yielded a heterogeneous effect on taxi injury severity. In addition, bus driver error and street light resulted in fixed yet moderate (less than 6%) effects on bus injury severity, while taxi driver gender, speed limit, rainfall, and collision manner resulted in fixed effects on taxi injury severity, where female drivers and front collisions significantly increased the likelihood of fatality and severe injury with their marginal effects more than 20%. Based on the findings, tailored strategies pertaining to safety education, law enforcement, vehicle safety devices, and traffic management and control were proposed to mitigate crash outcomes involving public buses and taxis.

Source: Applied Sciences

DOI: 10.3390/app16062783

Authors: Jing Huang, Zheliang He, Jun Li, Qiang Zeng, Xiaofei Wang

Institutions: South China University of Technology, Guangzhou Municipal Engineering Design and Research Institute

Language: English

Topic: Traffic and Road Safety

Subfield: Safety, Risk, Reliability and Quality

Field: Engineering

Domain: Physical Sciences

SDGs: Sustainable cities and communities, Good health and well-being, Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $2,896