The objective of the study described in this paper is to identify common site features that may contribute to high collision rates under wet pavement conditions. To minimize falsely identified high collision concentration locations (HCCL) in evaluating factors contributing to high collision rate, this study accessed the magnitude of false positives (i.e., identifying sites for safety improvements that should not have been selected) by comparing HCCLs identified by the existing conventional sliding moving window approach with the ones identified by the Continuous Risk Profile (CRP) approach and the safety investigators field evaluation notes. The result shows that CRP approach can reduce the false positive rate by 30%.
Significant shifts in collision distribution across traveling lanes were observed at some of the HCCLs under wet and dry pavement conditions. Speeding was the primary collision factor regardless of pavement condition, but it became a more dominant factor under wet pavement conditions at all observed locations.
Rapid spatial changes (i.e., vertical and horizontal curve over a short distance), narrower lane width, lack of median, and wider total freeway width were some of the notable geometric features observed at these sites. Other features responsible for diminishing drivers’ visibility are also contributing factors.