Modeling the Probability of Hazardous Materials Release in Crashes at Highway–Rail Grade Crossings

Amirfarrokh Iranitalab, Yashu Kang, Aemal Khattak

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Crashes at Highway–Rail Grade Crossings (HRGCs) that involve a truck or a train carrying hazardous materials (hazmat) expose people and the environment to potentially severe consequences of hazmat release. This research involved statistical modeling of the probability of hazmat release from trucks and/or trains in crashes at HRGCs to identify factors associated with hazmat release. The Federal Railroad Administration (FRA) HRGC crash dataset (2007–2016) yielded two subsets of crashes: 1) those involving hazmat-carrying trucks, and 2) those involving hazmat-carrying trains. Results from a logistic regression model using data subset 1 (crashes involving hazmat-carrying trucks) with hazmat release/no release as the response variable showed that standard flashing signal lights, railroad crossbucks, and railroad classes II and III (relative to railroad class I) were associated with lower hazmat release probability from hazmat-carrying trucks. Hazmat release probability from trucks was higher with freight train involvement. Results from a logistic regression model using data subset 2 (crashes involving hazmat-carrying trains) revealed that hazmat release probability from trains was lower with warmer temperature. However, the probability of release from trains was greater with railroad class II (relative to railroad class I), type of highway user (different types of trucks and motorcycle relative to automobiles), and weather conditions (fog, sleet or snow, relative to clear). A comparison of the results from this study with HRGC crash severity studies highlighted the importance and usefulness of this study.

Original languageEnglish (US)
Pages (from-to)28-37
Number of pages10
JournalTransportation Research Record
Volume2672
Issue number10
DOIs
StatePublished - Dec 1 2018

Fingerprint

Hazardous materials
Railroads
Trucks
Logistics
Crossings (pipe and cable)
Railroad crossings
Motorcycles
Fog
Snow
Automobiles

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Mechanical Engineering

Cite this

Modeling the Probability of Hazardous Materials Release in Crashes at Highway–Rail Grade Crossings. / Iranitalab, Amirfarrokh; Kang, Yashu; Khattak, Aemal.

In: Transportation Research Record, Vol. 2672, No. 10, 01.12.2018, p. 28-37.

Research output: Contribution to journalArticle

@article{78ba6a16db5c49df8c13ec3abb07b62e,
title = "Modeling the Probability of Hazardous Materials Release in Crashes at Highway–Rail Grade Crossings",
abstract = "Crashes at Highway–Rail Grade Crossings (HRGCs) that involve a truck or a train carrying hazardous materials (hazmat) expose people and the environment to potentially severe consequences of hazmat release. This research involved statistical modeling of the probability of hazmat release from trucks and/or trains in crashes at HRGCs to identify factors associated with hazmat release. The Federal Railroad Administration (FRA) HRGC crash dataset (2007–2016) yielded two subsets of crashes: 1) those involving hazmat-carrying trucks, and 2) those involving hazmat-carrying trains. Results from a logistic regression model using data subset 1 (crashes involving hazmat-carrying trucks) with hazmat release/no release as the response variable showed that standard flashing signal lights, railroad crossbucks, and railroad classes II and III (relative to railroad class I) were associated with lower hazmat release probability from hazmat-carrying trucks. Hazmat release probability from trucks was higher with freight train involvement. Results from a logistic regression model using data subset 2 (crashes involving hazmat-carrying trains) revealed that hazmat release probability from trains was lower with warmer temperature. However, the probability of release from trains was greater with railroad class II (relative to railroad class I), type of highway user (different types of trucks and motorcycle relative to automobiles), and weather conditions (fog, sleet or snow, relative to clear). A comparison of the results from this study with HRGC crash severity studies highlighted the importance and usefulness of this study.",
author = "Amirfarrokh Iranitalab and Yashu Kang and Aemal Khattak",
year = "2018",
month = "12",
day = "1",
doi = "10.1177/0361198118780885",
language = "English (US)",
volume = "2672",
pages = "28--37",
journal = "Transportation Research Record",
issn = "0361-1981",
publisher = "US National Research Council",
number = "10",

}

TY - JOUR

T1 - Modeling the Probability of Hazardous Materials Release in Crashes at Highway–Rail Grade Crossings

AU - Iranitalab, Amirfarrokh

AU - Kang, Yashu

AU - Khattak, Aemal

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Crashes at Highway–Rail Grade Crossings (HRGCs) that involve a truck or a train carrying hazardous materials (hazmat) expose people and the environment to potentially severe consequences of hazmat release. This research involved statistical modeling of the probability of hazmat release from trucks and/or trains in crashes at HRGCs to identify factors associated with hazmat release. The Federal Railroad Administration (FRA) HRGC crash dataset (2007–2016) yielded two subsets of crashes: 1) those involving hazmat-carrying trucks, and 2) those involving hazmat-carrying trains. Results from a logistic regression model using data subset 1 (crashes involving hazmat-carrying trucks) with hazmat release/no release as the response variable showed that standard flashing signal lights, railroad crossbucks, and railroad classes II and III (relative to railroad class I) were associated with lower hazmat release probability from hazmat-carrying trucks. Hazmat release probability from trucks was higher with freight train involvement. Results from a logistic regression model using data subset 2 (crashes involving hazmat-carrying trains) revealed that hazmat release probability from trains was lower with warmer temperature. However, the probability of release from trains was greater with railroad class II (relative to railroad class I), type of highway user (different types of trucks and motorcycle relative to automobiles), and weather conditions (fog, sleet or snow, relative to clear). A comparison of the results from this study with HRGC crash severity studies highlighted the importance and usefulness of this study.

AB - Crashes at Highway–Rail Grade Crossings (HRGCs) that involve a truck or a train carrying hazardous materials (hazmat) expose people and the environment to potentially severe consequences of hazmat release. This research involved statistical modeling of the probability of hazmat release from trucks and/or trains in crashes at HRGCs to identify factors associated with hazmat release. The Federal Railroad Administration (FRA) HRGC crash dataset (2007–2016) yielded two subsets of crashes: 1) those involving hazmat-carrying trucks, and 2) those involving hazmat-carrying trains. Results from a logistic regression model using data subset 1 (crashes involving hazmat-carrying trucks) with hazmat release/no release as the response variable showed that standard flashing signal lights, railroad crossbucks, and railroad classes II and III (relative to railroad class I) were associated with lower hazmat release probability from hazmat-carrying trucks. Hazmat release probability from trucks was higher with freight train involvement. Results from a logistic regression model using data subset 2 (crashes involving hazmat-carrying trains) revealed that hazmat release probability from trains was lower with warmer temperature. However, the probability of release from trains was greater with railroad class II (relative to railroad class I), type of highway user (different types of trucks and motorcycle relative to automobiles), and weather conditions (fog, sleet or snow, relative to clear). A comparison of the results from this study with HRGC crash severity studies highlighted the importance and usefulness of this study.

UR - http://www.scopus.com/inward/record.url?scp=85049034311&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049034311&partnerID=8YFLogxK

U2 - 10.1177/0361198118780885

DO - 10.1177/0361198118780885

M3 - Article

AN - SCOPUS:85049034311

VL - 2672

SP - 28

EP - 37

JO - Transportation Research Record

JF - Transportation Research Record

SN - 0361-1981

IS - 10

ER -