The purpose of this study is to evaluate the rock in the extreme regions by conducting a field study and laboratory tests on three rocks of diorite, andesite, grano-diorite around Sejong Station in Antarctica. The King George Island, the research area, is mostly covered by glaciers and partially exposed bedrock along the coast. Around the coastal area and Sejong-bong, andesites, diorites, grano-diorites are distributed and were measured rebound values using Silver Schmidt hammer. This hammer, unlike conventional Schmidt value’s R, calculates Q values using input and output energy. As a result of field study, the average Q value of diorite was estimated 76, which is high compared others, and andesite was estimated 67, which is low compared others, grano-diorite was estimated 72, which is widely scattered. Freeze-Thawing test was performed based on ASTM C-666, KS F 2456. The temperature range of freeze-thawing test is from -20 ℃ to 20 ℃ referred to the published papers, and all rocks are completely saturated without humidity. The temperature holding time was set to 2 hours for temperature inside rock to -20 ℃ when the atmosphere temperature is -20 ℃. The freeze-thawing test was carried out every 20 cycles for porosity, absorption, and slaking durability. The laboratory tests were performed 200 times in total. As a result, up to 100 cycles, the porosity and absorption were not significantly different. Since then, they increased slightly. However, the slaking tended to increase gradually from the 0 cycle. In order to accurately assess the weathering of the three rocks, continuous freeze-thawing tests should be conducted. Acknowledgement : This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MIST (2019R1F1A1048854)

Urban pothole management conventionally relies on reactive, complaint-driven maintenance, which inevitably suffers from operational delays and inefficient resource distribution. To catalyze a paradigm shift toward proactive maintenance, this study proposes a comprehensive data-driven framework that integrates machine learning, explainable AI (XAI), and Geographic Information System (GIS) spatial clustering to predict localized pothole risks. Utilizing a dataset of 21,397 spatial points (13,104 occurrences and 8293 non-occurrences) collected from Gwangju Metropolitan City (2022–2024), 19 diverse spatial, environmental, and geometric variables were extracted. To ensure strict statistical rigor and mitigate data leakage, dual-track encoding and Phi-K correlation analyses were applied to capture complex non-linear dependencies. Among the six predictive models evaluated, the Random Forest algorithm demonstrated superior performance, achieving an accuracy of 90.47% and an ROC-AUC of 0.957. Subsequent XAI interpretation using Shapley Additive exPlanations (SHAP) empirically revealed that physical scale and spatial vulnerabilities synergistically accelerate pavement fatigue. Furthermore, Density-Based Spatial Clustering of Applications with Noise (DBSCAN) successfully translated these predictive probabilities into 14 highly localized hotspots. Ultimately, the findings of this study provide a highly actionable, evidence-based guideline for local governments to abandon uniform budget distribution and transition to a preemptive, concentrated allocation of municipal resources directed exclusively at high-risk zones.