Citation

Ghomi H, Bagheri M, Fu L, Miranda-Moreno LF. Traffic Injury Prev. 2016; 17(8): 833-841.

Affiliation

Associate Professor, Department of Civil and Applied Mechanics , McGill University , luis.miranda-moreno@mcgill.ca.

Copyright

(Copyright © 2016, Informa - Taylor and Francis Group)

DOI

10.1080/15389588.2016.1151011

PMID

26980425

Abstract

OBJECTIVE: The main objective of this study is to identify the main factors associated with injury severity of the vulnerable road user (VRU) involved in the accidents at highway railroad grade crossings (HRGC) using data mining techniques.

METHODS: This paper applies Ordered Probit model, Association Rules and Classification-Regression Tree (CART) algorithms to the U.S. Federal Railroad Administration (FRA) HRGC accident database for the period of 2007-2013, to identify VRU injury severity factors at HRGCs.

RESULTS: The results show that train speed is the key factor influencing the injury severity. Further analysis illustrated that the presence of an illumination does not reduce the severity of accidents for high speed train. Also there is a greater propensity towards fatal accidents for elderly road users compared to the younger individuals. Interestingly, during the night, injury accident involving a female road user is more sever in comparison to a male.

CONCLUSIONS: The Ordered Probit Model was the primary technique, and CART and Association Rules act as the supporter and identifier of interactions between variables. All of the three algorithms results, consistently show that the most influential accident factors such as the train speed, VRU age and gender. The findings of this research could be applied for identifying high-risk hotspots and developing cost-effective countermeasures targeting VRUs at HRGC.

Language: en