Safety Effectiveness of Pedestrian Crossing Treatments

Over the last decade, the transportation agencies in Oregon have systematically enhanced many pedestrian crossings at mid-block locations with Rectangular Rapid Flashing Beacons (RRFBs), Flashing Yellow Beacons (Flash), and high visibility crosswalk markings (Hi-Vis). Enhancements often included the installation of refuge medians. This study explored the safety performance of these enhanced crossings, categorized by enhancement type. Data were collected on 191 crossings that included installation year, geometric features, surrounding land use, traffic volumes, and the number of crashes. Because pedestrian volume at the locations was unavailable, a pedestrian activity level variable was developed. Target crashes for analysis were identified as pedestrian and rear-end. The analysis of the before-after crash patterns showed a reduction in the pedestrian crash severity after the installation of the crosswalk enhancements. Risk ratios, calculated by the unadjusted crash frequency relative to the years of operation in each analysis category, were calculated. For pedestrian crashes, risk ratios increased with the number of lanes, posted speed, and estimated pedestrian activity level. Similar trends were observed for rear-end crashes. Due to sample size limitations, safety effectiveness was only estimated for the 19 RRFBs locations. Lack of pedestrian volumes limited the development of a safety performance function (SPF) for the pedestrian crash types. However, a rear-end crash SPF was estimated. Standard methods to estimate a crash modification factor (CMF) were attempted. The recommended CMF for pedestrian crashes is 0.64 +/- 0.26 using a simple before-after analysis and 0.93 +/- 0.22 for rear-end crashes using an empirical Bayes analysis. The final report that the presentation is derived from is available at: https://doi.org/10.15760/trec.168