Fatality rates for crashes involving heavy vehicles on highways: A random parameter tobit regression approach

Few studies have analyzed the impacts of freight movements (large truck) on crash fatality rates. This study explores a novel application of a method, namely the random parameters tobit regression model to large truck fatality rates by using the nationwide Fatality Analysis Reporting System. By utilizing random parameter tobit regression the authors examined crash rates (instead of frequencies) in per million truck-miles traveled and ton-miles of freight in the United States as continuous censored variables. The empirical and statistical results illustrate that the random-parameters tobit regression model provides a better understanding of the fatality rates per million truck-miles traveled and ton-miles of freight over the fixed parameter tobit model. Factors related to the crash mechanism, temporal and spatial characteristics, road and environmental attributes, vehicle configuration, drivers and passenger attributes were found to be statistically significant. Some exposure to injury severity related factors also were found to be significant with random parameters that vary across the observations.