The chance of accident involvements of the Interstate Commerce Commissionauthorized, large, single-unit trucks and tractor-trailers was investigated using the 1977 Bureau of Motor Carrier Safety accident and the Highway Cost Allocation Study exposure data. The model used was discrete-multivariate and capable of simultaneously analyzing both the accidents and the exposure. The variables that were found to be important predictors of accident involvements include trailer style, vehicle configuration, number of axles of power unit, trip length, road class, road surface condition, loading status, day/night, driver experience, and driver age. Particularly high accident involvement rates, of 200 involvements per 10 8 mi or higher, were shown by all van singles in local service, 3-axle-tractor singles in local service, 2-axle straight trucks in local service, and flatbed doubles in over-the-road service. Low accident involvement rates of less than 50 involvements per 10 8 were shown by all 3-axle straight trucks and 2-axle straighl Lrucks in over-the-road service. van singles and tanker singles in over-the-road service showed moderate involvement rates (less than 100 involvements per 10" mi), while 2-axle-tractor van doubles and 2-axle-tractor tanker doubles showed higher rates (120 to 200 involvements per 10 8 mi).

This paper contains the results of an investigation of the association between the chance of accident involvements and several influencing factors for the nation's large single-unit trucks and tractortrailers. The methodology developed for this study is aimed at obtaining stable estimates of the probabilities of accident involvements. The findings are useful as input for policy planning (including usage regulations for certain types of trucks and their route assignments, driver education, and training programs). They also provide input for vehicle tests and simulations, accident countermeasures, as well as identifying gaps in data collection efforts concerned with truck accidents and usage.