Application of Bayesian Statistics to Identify Highway Sections with Atypically High Rates of Median-Crossing Crashes

Abstract

This paper describes a Bayesian statistical technique for using crash records to estimate the frequency and rate of median-crossing crashes (MCCs) on each set of highway sections, in cases where MCCs are not explicitly identified in computerized crash records. This technique requires an analyst to review only a subset of hardcopy accident reports to produce a training sample, which is then used to identify computerized data associated (possibly imperfectly) with whether a crash was an MCC. This association can then be exploited to use larger sets of computerized records to increase statistical power over that provided by the training sample alone. This technique is applied to data from Minnesota's freeways and rural expressways. Estimates that allow highway sections to be ranked according to the estimated frequency or density of MCCs, or to the estimated MCC rate, are computed, and then the estimated frequency rankings are reported.