Road crashes are the major cause of unnatural deaths. Resultant, safety of road users has become one of the prime objectives of highway engineers apart from reliable and economic road network. This study is motivated to identify the safety requirements of different roads within the boundaries of Haridwar and Dehradun districts of Uttarakhand state, India. For analysis, six different roads were divided into 86 homogeneous segments of variable length. Essential traffic, geometric, and pavement surface variables, were selected and evaluated using Poisson regression (PR) and negative binomial regression (NB) model. Separate accident prediction models (APM’s) were developed for total, rear-end and head-on crashes. The APMs delineated important risk factors in terms of positive and negative association. The probability of a crash was identified using risk ratios. Further, the models were compared in terms of goodness of fit. Akaike information criterion (AIC), Schwarz's Bayesian (SBC) value showed that the NB model performs better than the PR model as a former model allows additional over-dispersion parameters in the model.
Analysis of Pavement and Geometric Factors of Selected Highways for Reduction in Road Accidents
Lecture Notes in Civil Engineering
Conference of Transportation Research Group of India ; 2021 December 14, 2021 - December 17, 2021
Proceedings of the Sixth International Conference of Transportation Research Group of India ; Kapitel : 16 ; 265-283
2022-09-29
19 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
NTRS | 1969
|Engineering Index Backfile | 1962
|Accidents on Suburban Highways - Tennessee's Experience
British Library Online Contents | 1995
|Impacts of Road-Trains on the Geometric Design of Highways
Online Contents | 2015
|