| This research evaluates the factors that contribute to fatality in large truck crashes. These contributing factors are related to truck drivers, vehicle defects, environmental conditions, and roadway geometry. This research investigates the statistically significant variables that have impact on the likelihood of fatalities in crashes involving large trucks. Logistic regression models are developed and estimated using SAS statistical software. The models and estimators are evaluated based on their (1) estimated regression parameters, (2) estimated total number of truck fatalities, (3) estimated relative frequencies of variables with truck fatalities involvement over time, and (4) overall goodness-of-fit. Data from the Federal Motor Carrier Safety Administration (FMCSA) are used to analyze the potential causes of fatalities in large truck crashes. The evaluation suggests that driver behavior, defective vehicles, environmental conditions, and roadway geometry are important factors in predicting the likelihood of fatalities in crashes involving large trucks. The results indicate that the likelihood of fatalities is decreased by reducing driver distractions and other impairments to judgments, improved vehicle inspections, and roadway geometry. The results of this research indicate that alcohol, inoperative brakes, roadway class, ran lights, conversation, and flow restriction are statistically significant contributing factors to fatalities in large truck crashes. Results of this analysis show that the combined effects of alcohol, inoperative brakes, and conversation contribute to fatalities. Combinations of these effects increase the probability of a fatality by 96% on two-lane highways. However, this study also shows that alcohol is the most significant variable in increasing the likelihood of a fatality in large truck crashes. The results indicate that in cases involving distracted drivers, the probability of fatality increases by 77% when alcohol is involved in crashes on two-lane highways. |
