Analysis of freight train collision risk in the United States

Tejashree Turla, Xiang Liu, Zhipeng Zhang

Research output: Contribution to journalArticlepeer-review

10 Scopus citations

Abstract

Rail transportation is pivotal for the national economy. Despite being rare, a train accident can potentially result in severe consequences, such as infrastructure damage costs, casualties, and environmental impacts. An understanding of accident frequency, severity, and risk is important for rail safety management. In the United States, extensive prior research has focused on risk analyses of train derailments and highway–rail grade crossing accidents. Relatively less work has been conducted regarding train collision risk. The US Federal Railroad Administration identifies various accident causes, among which the authors of this study have analyzed the major collision causes. For each major accident cause, the authors have analyzed its resultant collision frequency, severity (in terms of damage cost or casualties), and correspondingly the risk, which is the combination of the frequency and severity. The analysis was based on train collision data in the United States from 2001 to 2015. This analysis focuses on freight trains in the United States, due to their immense traffic exposure. On the temporal scale, collision rate (the number of collisions normalized by traffic exposure) has an approximately 5% annual reduction. In terms of collision cause, failures to obey signals, overspeeds, and violations of mainline operating rules accounted for more collisions than other causes. Two alternative risk measures, namely the expected consequence and conditional value at risk, were used to evaluate the freight train collision risk on main tracks, accounting for both the average and worst-case scenarios. This collision risk analysis methodology may provide the US Department of Transportation and railroad industry with information and decision support for identifying, evaluating, and implementing cost-effective risk mitigation strategies.

Original languageEnglish (US)
Pages (from-to)817-830
Number of pages14
JournalProceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit
Volume233
Issue number8
DOIs
StatePublished - Sep 1 2019

All Science Journal Classification (ASJC) codes

  • Mechanical Engineering

Keywords

  • Railroad
  • accident cause
  • freight
  • risk
  • train collisions

Fingerprint

Dive into the research topics of 'Analysis of freight train collision risk in the United States'. Together they form a unique fingerprint.

Cite this