Safety risk analysis of restricted-speed train accidents in the United States

Zhipeng Zhang, Xiang Liu

Research output: Contribution to journalArticlepeer-review

10 Scopus citations

Abstract

An understanding of accident risk based on historical safety data can support the development and prioritization of effective accident prevention strategies. While previous studies have focused extensively on the safety risks associated with high-frequency-high-consequence accidents, comparatively little work has been undertaken to evaluate railroad risk and safety under restricted-speed operations. In the United States, restricted speed is defined as a speed that permits stopping within one-half the range of vision, but not exceeding 20 miles per hour. Human-reliant restricted-speed operation has been common on U.S. freight railroads for over a century. Recently, a series of severe accidents due to violations of restricted speed rules has triggered renewed interest in understanding and improving restricted-speed operational safety. To this end, this paper develops a statistical analysis of restricted-speed train accidents occurring between 2000 and 2016, based on data from the U.S. Federal Railroad Administration. Our study quantitatively analyzes the distribution of restricted-speed accident frequency, severity, risk, and other pertinent characteristics. The research finds that while the overall train accident rate has declined substantially, there is no significant improvement regarding the restricted-speed train accident rate over the studied period. In order to characterize the risk profile of this type of accident, two alternative risk measures, namely mean value and Conditional Value at Risk (CVaR), are developed to estimate an annual restricted-speed train accident risk. In particular, the CVaR represents the expected consequence in worst-case scenarios, which more effectively characterizes the low-probability-high-consequence restricted-speed accidents under certain circumstances. Furthermore, based on a micro-level study employing developed Fault Tree Analysis, effective human-error mitigation actions (e.g. valid medical program, alerter system) and advanced train control (e.g. Positive Train Control) are discussed as two primary restricted-speed accident risk prevention strategies to improve the safety level of train operations at restricted speeds.

Original languageEnglish (US)
Pages (from-to)1158-1176
Number of pages19
JournalJournal of Risk Research
Volume23
Issue number9
DOIs
StatePublished - 2020

All Science Journal Classification (ASJC) codes

  • General Engineering
  • Safety, Risk, Reliability and Quality
  • General Social Sciences
  • Strategy and Management

Keywords

  • Train accident
  • railroad
  • risk analysis
  • safety

Fingerprint

Dive into the research topics of 'Safety risk analysis of restricted-speed train accidents in the United States'. Together they form a unique fingerprint.

Cite this