Wet and Wild: Developing and Evaluating an Automated Wet-Pavement Motorist Warning System

Abstract

A fully automated motorist warning system developed for wet-pavement conditions was evaluated. The demonstration took place on one expressway interchange ramp where 69 percent of total recorded crashes had been classified as “run-off” crashes during wet-pavement conditions. However, fewer than one-half of all wet-pavement crashes occurred during rain. The potential solution was to develop an automated, dynamic motorist warning system to attract attention to the advisory speed limit signs and thus encourage motorists to reduce vehicle speed. A pavement sensor embedded in the roadway activated two flashing beacons located above the signs whenever moisture was detected. Infrared radar recorded vehicle speed at the site. Speeds and volumes were grouped into a matrix according to weather conditions and time periods, which were based on sunlight visibility and peak traffic hours. The reduction of the 85th-percentile speeds served as the first measure of effectiveness, and only “like” conditions were compared for system evaluation. In total, more than 27,000 wet-pavement vehicle speeds were compared before and after system activation. The average reduction in travel speed was 16 km/h (10 mph) during heavy rain and 8 km/h (5 mph) during light rain; the standard deviation for vehicle speed also was reduced after system activation. No run-off crashes were reported at the site after the first week of the evaluation period.