01371353

Component

An analysis of vehicle crash data determined that a significant number of severe crashes occur on the curves of two-lane roads. Most of these crashes have at least some level of injury due to the nature of the collisions. Hence, crash severity modeling on horizontal curves was applied towards reducing the severity of these accidents. Crash severity is dependent upon several factors including driver behavior, traffic conditions, travel speed, road geometry, vehicle characteristics and pavement condition. The paper discusses a study that aims to identify the relationship between pavement condition indices and crash-level injury severity. The study highlights the benefit of linking different databases for safety maintenance of roadways. The analysis dataset used is a coalition of a crash database and Pavement Management Information System (PMIS) network level condition assessments, both maintained by the Texas Department of Transportation (TxDOT). An ordered probit injury severity model is developed using variables available in these databases. The study establishes the relationship between pavement condition indices and injury severity of crashes occurring on two-lane horizontal curves. The sensitivity of the pavement indices with respect to the probability of a crash becoming fatal was found to be low. A poor correlation is also apparent between pavement skid index and injury severity.

This paper was sponsored by TRB committee AFD10 Pavement Management Systems

01362476

12-1285

English

12p

2012

Transportation Research Board 91st Annual Meeting

Digital/other

References; Tables

Crash characteristics; Crash severity; Highway curves; Injury severity; Sensitivity analysis; Two lane highways

Pavement Management Information System

Pavement conditions

Highways; Pavements; Safety and Human Factors; I80: Accident Studies

Transportation Research Board Annual Meeting 2012 Paper #12-1285

TRIS, TRB

Feb 8 2012 5:01PM