Selecting exposure measures for predicting crash rates on two-lane rural highways

A critical part of any risk assessment is identifying how to represent exposure to the risk involved. Recent research shows that the relationship between crash count and traffic volume is non-linear; consequently, a simple crash rate computed as the ratio of crash count to volume is not proper for comparing the safety of sites with different traffic volumes. To solve this problem, we describe a new approach for relating traffic volume and crash incidence on rural two-lane highway segments. Specifically, we disaggregate crashes into four types: (1) single-vehicle, (2) multi-vehicle same direction, (3) multivehicle opposite direction, and (4) multi-vehicle intersecting direction, and define candidate exposure measures for each (as a function of site traffic volumes) that we hypothesize will be linear with respect to each crash type. ^ This report first describes investigation using crash and physical characteristics data for highway segments from four states participating in the Highway Safety Information System (HSIS). We have used a hierarchical Bayesian framework to fit zero-inflated-Poisson regression models for predicting counts for each of the above crash types as a function of the daily volume, segment length, speed limit and pavement width. We found that the relationship between crashes and the daily volume is non-linear and varies by crash type, and is significantly different from the relationship between crashes and segment length for all crash types. These findings cast doubt on the usage of VTM and AADT as usual exposure measures. ^ Part III of the report describes an investigation using hourly volume counts for small samples of highway segments from two states. The results reveal how the relationship between crashes and hourly flow rate varies by time of day, thus improving the accuracy of crash count predictions. The results show that even accounting for time of day, the disaggregate exposure measure—hourly flow rate is indeed non-linear for each of the four crash types. These findings help us to further understand the relationship between number of crashes and flow rate, segment length and other risk factors, and facilitate more meaningful comparisons of the safety record of seemingly similar highway locations.