Forecasting short-term road surface temperatures considering forecasting horizon and geographical attributes – an ANN-based approach

Abstract

The ability to forecast road surface temperature (RST) in advance is an important asset for winter maintenance (WRM) operations. It effectively allows for the reduction of WRM cost through more efficient use of their maintenance resources. However, considering extensive road networks that must be monitored and the extent at which RST varies over time, WRM agencies have constantly been looking for better ways to generate accurate road weather forecasts as they strive to optimize their WRM services and maintain safe travels. To tackle this, this study developed RST forecasting models using an Artificial Neural Network (ANN). RST measurements collected by six selected stationary road weather information systems (RWIS) stations in Alberta, Canada were utilized to validate the feasibility and applicability of the proposed method developed herein. The developed models were found to generate highly accurate results with mean absolute error (MAE) values of 0.64, 1.20, 1.59, 2.16, 2.56, and 3.03 °C for 1-h, 2-h, 3-h, 4-h, 5-h, and 6-h ahead forecasts, respectively. The novelty of this study lies in investigating the probable effect of some external factors on model performance, where it was revealed that forecasting horizon and geographical attributes influenced forecasting accuracies. Upon investigating the hypothesis that locational attributes would affect forecasting accuracies, the results confirmed that accuracy improved with increasing latitude, and decreasing elevations – worthwhile findings that can potentially lead to developing more refined models for generating highly accurate location-specific RST forecasts.