Climate change can lead to unpredictable slope collapse, which causes human casualties. Therefore, Korea has devoted significant effort to the management of slope disasters. The Ministry of the Interior and Safety of Korea, which oversees the safety of the nation’s people, has allocated a four-year budget of $557 million to investigate, assess, and maintain steep slope sites. However, there have been fatalities caused by steep slope site evaluations based on inadequate knowledge and a single retaining walls and embankments (RW&E) assessment table. Therefore, the assessment table for RW&E-type steep slopes needs to be improved in terms of its accuracy, simplicity, and ease of use. In this study, domestic and global evaluation methods were reviewed, problems associated with the existing RW&E assessment table were identified, and a focus group interview was conducted. The RW&E assessment table was improved through an indicator feasibility survey and analytic hierarchy processing. The improved assessment table was categorized from one to four classifications to reduce the ambiguity of the evaluation: concrete, reinforced soil-retaining walls, stone embankments, and gabions. This study will provide the sustainability of slope safety and serve as a reference for classification and evaluation criteria across all national institutions that conduct RW&E evaluations.

Traffic accidents caused by black ice have become a serious public safety concern due to their high fatality rates and the limitations of conventional detection systems under low visibility. Millimeter-wave (mmWave) radar, capable of operating reliably in adverse weather and lighting conditions, offers a promising alternative for road surface monitoring. In this study, six representative road surface conditions—dry, wet, thin-ice, ice, snow, and sludge—were experimentally implemented on asphalt and concrete specimens using a temperature and humidity-controlled chamber. mmWave radar data were repeatedly collected to analyze the temporal variations in reflected signals. The acquired signals were transformed into range-based spectra using Range-Fast Fourier Transform (Range-FFT) and converted into statistical features and graphical representations. These features were used to train and evaluate classification models, including Extreme Gradient Boost (XGBoost), Light Gradient-Boosting Machine (LightGBM), Convolutional Neural Networks (CNN), and Vision Transformer (ViT). While machine learning models performed well under dry and wet conditions, their accuracy declined in hazardous states. Both CNN and ViT demonstrated superior performance across all conditions, with CNN showing consistent stability and ViT exhibiting competitive accuracy with enhanced global pattern-recognition capabilities. Comprehensive robustness evaluation under various noise and blur conditions revealed distinct characteristics of each model architecture. This study demonstrates the feasibility of mmWave radar for reliable road surface condition recognition and suggests potential for improvement through multimodal sensor fusion and time-series analysis.