This study aimed to statistically analyze and understand the factors contributing to road icing accidents on Korean roads, particularly within sections designated as ice-prone. The objective was to assess the adequacy of the existing criteria for designating these sections and to provide insights for improving safety measures on icy roads. Road icing accident data were collected from Korea’s (traffic accident analysis system (TAAS) database, covering incidents from 2018 to 2022 on both expressways and general roads. The data were compiled to create a comprehensive database that incorporated geographic and weather-related factors influencing road icing accidents. Using geographic information systems (GIS), a spatial analysis was performed on the designated ice-prone and non-designated sections. The inverse distance weighting (IDW) method was applied to interpolate the meteorological data, and the temperature lapse rate was used to adjust for altitude-related variations in temperature, providing estimates for conditions such as the average temperature, relative humidity, and precipitation days at each accident location. The analysis revealed that only 2.9% of designated ice-prone sections reported icing-related accidents, whereas 97.6% of icing accidents occurred in non-designated sections. This discrepancy highlights potential gaps in the current criteria for designating ice-prone areas. Additionally, road characteristics such as traffic volume, lane count, speed limit, and curve radius, along with meteorological factors such as humidity, precipitation frequency, and solar radiation exposure, were observed to be significantly correlated with icing accidents. Regions with high relative humidity and low sunlight exposure exhibited a heightened risk of road icing accidents, suggesting the need for these parameters to designate ice-prone sections. This study underscores the need to refine Korea's criteria for designating ice-prone sections by incorporating a wider range of risk factors, including detailed terrain data, shaded areas, and road geometry factors, such as slopes. Integrating these additional elements into the designation process could enhance the precision and effectiveness of preventive measures and ultimately reduce the risk of icing-related accidents in Korea’s road networks. These findings provide a foundation for future policy and management decisions to enhance road safety during the winter.

도로 결빙이란 도로 표면에 형성된 얼음층으로 도로 결빙으로 인한 교통사고의 치사율은 결빙이 원인이 아닌 교통사고의 치사율과 비교하여 1.5배 높은 수치인 2.3으로 나타났다. 현재 국토교통부에서는 결빙사고 취약구간을 선정하고 관리하기 위하여 결빙 취약구간 평가기준표를 제시하였다. 그러나 도로 결빙은 노면 온도와 수분 공급에 따라 형성되며 기온, 구름량, 풍속, 풍향, 상대습도, 강수량 등 의 기상인자들이 복합적으로 작용하여 발생하며, 기존의 평가 기준은 이와 같은 인자들을 충분히 반영하지 못하여 결빙 형성을 예측 하고 평가하는 능력이 부족하다고 판단된다. 따라서 본 연구는 결빙 교통사고 데이터의 통계적인 분석을 통하여 결빙이 형성되는 기 상 조건을 구체화하고 결빙사고 및 결빙 형성을 예측하기 위한 기상학적 기준을 마련하는 것을 목적으로 진행되었다. 2018년 1월 1 일～2024년3월 15일 동안 발생한 결빙 사고와 사고 발생 당시 및 이전 6시간동안의 기상 데이터를 분석 데이터로 사용하였다. 이때, 역거리 가중법, 기온감률 등 공간보간기법을 적용하였다. 이후, 박스도표, 히스토그램, 경험적 누적분포함수 등의 통계분석을 적용하여 결빙사고의 기상 분포 특성을 확인하였다. 최종적으로 결빙사고의 몬테카를로 시뮬레이션을 활용하여 기온 및 습도에 따른 결빙사고 의 발생 확률을 계산하였다. 이와 같은 연구 결과는 결빙 형성을 예측하는 기온 및 습도의 기준점으로 제시할 수 있으며 더 나아가, 추후 결빙사고 예방 및 예보의 기준으로 활용이 가능할 것으로 기대된다.

PURPOSES : The purpose of this study is to statistically analyze the meteorological factors that contribute to the formation of road surface icing based on actual cases of icing accidents and provide directions for improving icing evaluation criteria. METHODS : In this study, we collected cases of domestic road icing accidents by searching news articles with the keyword ‘icing collision accidents’. Subsequently, we determined the latitude, longitude, and altitude of accident locations using satellite map service. We applied the Inverse Distance Weighting (IDW) method and temperature lapse rate to estimate meteorological data at each location. Finally, statistical analysis was conducted for temperature, humidity, and precipitation occurrence using probability density functions. RESULTS : As a result, road icing accident data points with identifiable location coordinates were collected. Among these, temperature, humidity, and precipitation occurrence from Automated Weather Stations (AWS) data were selected for analysis. During the process of correcting meteorological factors using the Inverse Distance Weighting (IDW) method, the optimal Weighting Exponent (p) that minimizes the error was determined and applied. The results showed that accidents occurring in the morning indicated the highest accident occurrence rate. The average temperature at the time of the accidents was -1.4°C, with a humidity level of 85.1%. Precipitation was observed at the time of the accident in 19 cases. CONCLUSIONS : Icing on pavement can occur not only under extreme weather conditions but also under typical meteorological conditions. Typically, icing can occur when the relative humidity is above 70%. Accordingly, for future improvements in the evaluation criteria for icing-prone areas by the Ministry of Land, Infrastructure and Transport, it is possible to incorporate the temperature and humidity ranges that generally lead to icing, taking into account climate characteristics.

The airborne chlorides environment by de-icing salts of the Suam road tunnel in motorway were examined. It was found that higher airborne chlorides were detected inside road tunnel than outside. Therefore, there is a need to appreciate that RC structures inside tunnel might be exposed to salt attack environment.

The airborne chlorides environment by de-icing salts of the Suam road tunnel in motorway were examined. It was found that higher airborne chlorides were detected inside road tunnel than outside. Therefore, there is a need to appreciate that RC structures inside tunnel might be exposed to salt attack environment

Recently, the concern on the deterioration of concrete due to de-icing salts and its counter measurements has been increased. This paper discusses the results from investigation of pre-mature deteriorated existing road-subsidiary concrete structures and the specification of concrete quality.

The road icing forecast and snow detection system using state evaluation algorithm of multi sensor optimizes snow melting system to control spread time and amount of chemical spread application This system operates integrated of contact/non-contact sensor and infrared camera. The state evaluation algorithm of the system evaluates road freezing danger condition and snowfall condition (snowfall intensity also) using acquired data such as temperature/humidity, moisture detection and result of image signal processing from field video footage. In the field experiment, it proved excellent and reliable evaluated result of snowfall state detection rate of 89% and wet state detection rate of 94%.