Award Date

5-2010

Degree Type

Dissertation

Degree Name

Doctor of Philosophy in Engineering

Department

Civil and Environmental Engineering

First Committee Member

Mohamed S. Kaseko, Chair

Second Committee Member

Moses Karakouzian

Third Committee Member

Nader Ghafoori

Fourth Committee Member

Alexander Paz

Graduate Faculty Representative

Ashok Singh

Number of Pages

188

Abstract

Crashes are the result of complex interactions between several factors representing driver, roadway, vehicle, and environmental characteristics. Understanding to what degree each factor contributes to the severity of a crash is not a simple task. The outcomes of crashes in the US have been an average of 42,000 deaths and 3 million injuries per year. To better understand the role of significant contributors to crashes, three sets of models using multinomial logit and one using ordered probit were calibrated. To calibrate these models, forty two independent predictor variables including driver/occupants characteristics, crash environment at the crash location, crash characteristics, and vehicle characteristics were used. In total, twenty variables were found to be statistically significant.

The crash data used for this study was from the state of North Carolina. The obtained data included all the crash records for the year 2003. Vehicle dimensions were incorporated into the final database.

The contributions of this study were twofold: First, the evaluation of the impact of passenger-vehicle dimensions on the injury severity. The following is a condensed summary of the findings:

An increase in the vehicle front overhang was more likely to decrease the risk of suffering an evident injury for two groups: drivers age 66 and older, and the male drivers. In addition, an increase in the vehicle rear overhang was more likely to reduce the risk associated to fatal injury for three groups: female drivers, drivers age <= 25, and drivers age 66 and older. Further, an increase in the vehicle width was more likely to increase the risk of sustaining injury for drivers age 66 and older.

Second, although the findings of this research were consistent with other researches, some differences identified as discussed below.

An increase in vehicle weight increased the risk of sustaining a fatal injury for two groups: drivers age<= 25 and female drivers. Furthermore, an increase in number of occupants did not pose an extra risk of fatal injury for two groups: drivers age 46-65 and female drivers. Moreover, dark roads with no lighting posed an extra risk of sustaining a fatal injury for drivers age <=25, but posed the lowest risk of injuries for female drivers. Further, head on crashes imposed a higher risk of sustaining a fatal injury for two groups: drivers age<=25 and female drivers. Finally, roads with no divided medians posed a higher risk of injury for two groups: drivers age <= 25 and male drivers.

Keywords

CCrash; Injury severity; Multinomial logit model; Ordered probit model; Probabilities; Traffic accident victims; Traffic accidents; Vehicle Dimensions; Vehicles – Sizes; Wounds and injuries

Disciplines

Civil Engineering | Transportation

Language

English


Share

COinS