The number of miles traveled each year by the U.S. large-truck population is a topic of interest for many reasons, one of which is safety. Although the number of accidents involving large trucks may be easily calculated from accident data, it is often more informative to know their risk of accident involvement per mile of travel. This requires accurate travel data. Compared in this paper are three sources of truck travel data: the Truck Inventory and Use Survey conducted by the Bureau of the Census; the National Truck Trip Information Survey conducted by the University of Michigan Transportation Research Institute; and annual estimates published in Statistics by the Federal Highway Administration. Each data source yields different estimates of annual travel by large trucks, which is to be expected considering the difficulty of collecting travel data. The overall conclusion, however, is that the Truck Inventory and Use Survey and the National Truck Trip Information Survey estimates are reasonably close to each other, whereas Statistics estimates are significantly higher. The implication of this finding is that the procedures used by the states and the Federal Highway Administration to generate Statistics data lead to artificially and systematically high estimates of travel by large trucks.

This article discusses problems related to truck safety including truck exposure data and truck crash data, and summarizes findings of recent research in Michigan. It also suggests directions for future work. It is noted that the differences in safety (as measured by involvement in traffic crashes) between singles and doubles are fairly small and, when only multivehicle crashes are considered, doubles are probably safer. The Michigan study found that, in general, doubles in Michigan are probably safer overall. The study has also shown that truck safety decreases as one moves down the hierarchy of the highway system - the more restrictive geometry of nonlimited-access highways and local roads leads to higher crash rates.

The National Truck Trip Information Survey (NTTIS) is part of a research program at the University of Michigan Transportation Research Institute (UMTRI) to study the safety of large trucks (trucks with gross vehicle weight ratings greater than 10,000 lb) on the highway. The objective of the NTTIS is to provide descriptive information on the national population of large trucks and their use. This information will be combined with data from a companion survey of the fatal accident experience of all large trucks in the United States, called Trucks Involved in Fatal Accidents (TIFA), to estimate involvement rates (fatal accident involvements per hundred million vehicle miles) for a broad range of truck configurations and use. Presented in this paper is a brief discussion of the overall methodology of the research program as background. The sampling frame, sample design, and survey methods are described, and preliminary estimates of the national population of large trucks are presented. The survey design and preliminary results are compared to the 1982 Truck Inventory and Use Survey conducted by the Bureau of the Census.