Citation

Mirzahossein H, Adibi H, Khishdari A, Afandizadeh S, Severino A. Iran. J. Sci. Technol. Trans. Civil. Eng. 2023; 47(4): 2497-2506.

Copyright

(Copyright © 2023, Holtzbrinck Springer Nature Publishing Group)

DOI

10.1007/s40996-023-01072-3

PMID

unavailable

Abstract

Recognizing the factors that have the greatest significant effect on the probability of a crash occurring is an essential part of road safety research. Many traffic crashes occur on rural roads with two-lane two-way. The purpose of this study was to explore the effect that the geometric design characteristics of rural two-lane roads had on the frequency of resulting crashes in developing country of Iran with the case study of Karaj-Chalus Road. The width of the pavement, the width of the shoulders, the curve, access to the secondary roads, the overtaking lane, the condition of the lighting, and the roadside features are all explanatory variables. Poisson and negative binomial regressions were utilized to model the crash frequencies. According to the findings, Poisson regression performed better than the other one. This was demonstrated by the fact that the Akaike's information criterion value of goodness-of-fit for Poisson and NB was, respectively, 211.783 and 272.494. In addition, the results of the regression showed that lane width and shoulder width were the most influential factors. Both of these components had p values that were less than 0.05 in the confidence range for 95 percent, indicating that they were meaningful. Additionally, their elasticity estimates were −0.65 and −0.53, respectively. In the end, the case-studied segments were ranked in order of priority according to the road safety index.

Language: en

Keywords

Crash frequency; Negative binomial (NB) regression; Poisson regression; Regression; Safety; Two-lane rural roads