Black Ice Prediction Model for Road Pavement Using Weather Forecast Data and GIS Database

Abstract

Black ice is a thin coating of ice on the road surface, which strongly reduces friction at the tire-road surface, resulting in dangerous driving when it happens. An appropriate diagnostic of black ice could prevent traffic accidents as well as provide timely notice to drivers. Therefore, this study aims at developing a black ice prediction model to diagnose the probability of black ice formation. Several combinations that can form road ice have been considered, including freezing rain, hoar frost, freezing of wet roads. In addition, black ice risky index (BRI) has been computed to reflect the probability of black ice formation. To acquire a fast prediction and high accuracy, the existing Geographical Information System (GIS) database and meteorological data have been utilized. GIS database includes road geometry and location of automatic weather stations, while the meteoritical data consists of air temperature, wind speed, humidity, cloud cover. The model has been developed based on the Python programming language. A 5-km road condition was observed from 1 December to 31 December 2021 to determine the model accuracy. Based on the results from the prediction model, black ice formation has been verified when the BRI is higher than 0.8. The model may be useful to develop black ice diagnostic program.

Keywords

• api
• asphalt pavement
• black ice
• gis
• meteorological data
• prediction model