California Partners for Advanced Transportation Technology

This research involves the development of a tool that can be used to assess the changes in traffic safety tendencies that result from changes in traffic flow. The tool uses data from single inductive loop detectors, converting 30-second observations of volume and occupancy for multiple freeway lanes into traffic flow regimes. Each regime has a specific pattern of crash types, which were determined through nonlinear multivariate analyses of over 1,000 crashes on freeways in Southern California. These analyses revealed ways in which differences in variances in speeds and volumes across lanes, as well as central tendencies of speeds and volumes, combine in complex ways to explain crash taxonomy. This research may provide the foundation to forecast the crash rates, in terms of vehicle miles of travel, for vehicles that are exposed to different traffic flow conditions.