System Dynamics Modeling for Estimating the Locations of Road Icing Using GIS

Road icing can cause large traffic accidents on highways because, unlike snowy roads, its location is difficult to identify and it can occur rapidly, even during rainy weather. In this study, the amount and location of road icing were modeled and simulated over time based on the system dynamics theory. The simulation is expressed on the geographic information system (GIS) and facilitates advance detection of the location and amount of road icing that occurs unexpectedly unlike previous studies. Modeling was designed to process spatial and meteorological data after combining them. The spatial data used for modeling were Hillshade, Water System, Bridge, and Road (Highway). Air temperature, cloudiness, vapor pressure, wind speed, and precipitation were used as meteorological data. The amount of road icing was estimated by scientifically designing the parameters related to its occurrence between spatial and meteorological data. Based on this, the amount of road icing by location was simulated per 1m2 using the GIS. The simulation results showed that the amount of road icing that began to increase from AM 08:00 reached its peak (an average of 213.62 g/m2) at noon and then slowly decreased. Additionally, when simulated with GIS, the sum amount of road icing between AM 12:00 and PM 13:00 was a maximum of 1707.292 (g/14 h) and a minimum of 360.082 (g/14 h) for each location. Hypothesis testing was conducted on whether road icing significantly occurs at actual points vulnerable to traffic accidents. Based on the results, the average significance level was calculated to be less than 0.05. Therefore, the alternative hypothesis that the model can estimate road icing in vulnerable areas was adopted. The verified simulation can be useful data to government agencies (e.g., road traffic authority) in their programs to prevent traffic accidents caused by road icing.