Adverse weather can significantly change the condition of the roadway within a short period of time, often with little or no warning to motorists or response personnel charged with protecting public safety. Traditional crash data is both aggregate and subjective in nature, limiting it to (1) accurately identify those weather conditions under which safety levels are minimized and (2) effectively guide crash-preventative response to adverse weather conditions (i.e., the dispatch of sand trucks or the dissemination of slippery condition warnings). The advent and expanded use of Road Weather Information Systems (RWIS), capable of providing detailed numeric data related to roadway surface, air and dew temperature and wind speed in addition to categorical information such as the presence of precipitation, road surface condition, and wind direction, shows potential for improving the identification of weather-related factors contributing to low levels of safety and for improving guidance provided to response personnel during or preceding times of adverse weather. While demonstrating potential, this investigation revealed several significant issues associated with the use of RWIS data for improving adverse weather-related crash prediction and response: (1) the categorical nature of some RWIS-reported data elements limits its usefulness in guiding response actions, (2) RWIS data is limited in historical timeline and ease of accessibility, and (3) RWIS data are highly localized spatially (i.e., reporting the pavement surface status only at the location of the in-road sensor) which results in substantial discrepancies between officer-reported and RWIS-reported crash data. (A) For the covering abstract of the conference see ITRD no 207828. The reprints are also available at the web - http://www.vti.se/pdf/reports/K18APart1.pdf; http://www.vti.se/pdf/reports/K18APart2.pdf and http://www.vti.se/pdf/reports/K18APart3.pdf.
Abstract