Estimating Incident Duration Considering the Unobserved Heterogeneity of Risk Factors for Trucks Transporting HAZMAT on Expressways. Issue 2 (February 2019)

• Record Type: Journal Article
• Authors: Hong, Jungyeol; Tamakloe, Reuben; Park, Dongjoo; Choi, Yoonhyuk
• Is Part Of: Transportation research record. Volume 2673:Issue 2(2019)
• Page Start: 232
• Page End: 242
• Publication Date: 2019-02
• Subjects: Transportation -- Periodicals; Roads; Transport -- Périodiques; Routes -- Périodiques; Routes -- Conception et construction -- Périodiques; Roads; Transportation; 388.05
• DOI: 10.1177/0361198119827925
• Languages: English
• ISSNs: 0361-1981

Abstract

Traffic accidents involving vehicles transporting hazardous materials (HAZMAT) on expressways not only delay traffic flow but can also cause large-scale casualties and socio-economic losses. Therefore, rapid response to and prevention of these accidents is important to minimize such loss. To ensure more efficient accident response, this study applied a random parameter hazard-based Weibull modeling approach to measure the relationship between crash characteristics and accident duration for trucks transporting HAZMAT. The study focuses on finding the key factors that have an impact on the accident duration of these vehicles as well as a statistical method to estimate the accident duration. The analysis is based on raw crash data from 2007 to 2017, obtained from the Korea Expressway Corporation, of crashes that involved HAZMAT trucks. The study found that crashes occurring during peak times of the day; crashes occurring on segments at the mainline, ramp, and roadways with a guardrail; and the number of vehicles involved in a crash, result in random parameters. In addition, the weather, season, crash severity, truck size, crash location, type of accident report, roadside features (e.g., guardrails), and status after a crash, can be used to explain the accident duration. The random parameters hazard-based model is found to have a better fit than a fixed model since it is able to capture the unobserved heterogeneity in the hazard function.