Citation

Bandyopadhyaya R, Mitra S. Procedia Soc. Behav. Sci. 2013; 104: 1011-1019.

Copyright

(Copyright © 2013, Elsevier Publishing)

DOI

10.1016/j.sbspro.2013.11.196

PMID

unavailable

Abstract

Crash injury severities are recorded in ordinal scales with fatal crashes ranking highest in the scale and property damage minor crashes ranking lowest in the scale. Various researchers have attempted to model injury severity outcomes by taking injury severity levels as either simple categorical variables or ordered categorical variables. The severity level of crashes vary with the collision partners, crash time, roadside activity characteristics and road inventory characteristics. Collision partner is the key attribute in determining the severity outcome of crashes, but severity of crashes vary even if the collision partners remain same. The influence of other factors on severity outcome may be fixed or random. This study makes an attempt to model the crash severity outcome for vulnerable road user involving crashes, crashes involving passenger cars and crashes involving only heavy vehicles using multinomial and ordered models. The model is developed using data from National Highway-6 (NH-6,West Bengal, India), considering crash time, roadside activity characteristics and road inventory details as contributing factors. The crash severity outcomes are categorized into three categories viz. minor injury crashes, major injury crashes and fatal crashes. The study also attempts to identify factors influencing higher severity crashes.