Quantitative Assessment of Crack Density Considering Weathering-Induced Uncertainty in Rock Properties

Hyung-Koo Yoon

doi:10.7843/kgs.2025.41.6.195

All Issue

2025 Vol.41, Issue 6 Preview Page Next Page

Quantitative Assessment of Crack Density Considering Weathering-Induced Uncertainty in Rock Properties 입력 물성 변동이 균열밀도 예측에 미치는 영향: 물리 기반 탄력도와 기계학습 중요도 비교

31 December 2025. pp. 195-204

PDF XML

Abstract

Crack density is a key indicator for evaluating the dynamic and elastic behavior of rock and soil materials, and is closely linked to compressional wave velocity (Vp), shear wave velocity (Vs), and porosity (φ). In this study, laboratory weathering test data were expanded into Monte Carlo–augmented datasets by introducing 1%, 2%, and 5% uncertainty into the input properties. The structural sensitivity of the crack density equation and the data-driven sensitivity derived from machine learning methods were jointly analyzed to examine the influence of each input variable. The physics-based elasticity analysis shows that crack density is most sensitive to Vp on a global scale, while Vs exhibits marked local sensitivity under specific conditions; in contrast, the influence of φ remains minimal. The AI-based importance analysis similarly identifies Vp and Vs as the dominant predictors of crack density across all three algorithms, consistent with the physics-based results. Moreover, the relative importance of Vs increases as the uncertainty level (σ) grows, reflecting the nonlinear sensitivity amplification observed in elasticity evaluation. Overall, this study demonstrates that the structural characteristics of the physical model and the empirical sensitivities derived from data-driven methods are complementary, providing insights that strengthen reliability assessment of crack density prediction models and inform management of input material properties.

Keywords

Crack density

Elasticity

Feature importance

Monte carlo simulation

Uncertainty

균열밀도는 암반 및 지반 재료의 동적·탄성적 거동을 평가하는 핵심 지표로, 압축파 속도(Vp), 전단파 속도(Vs) 그리고 간극률(φ)과 밀접하게 연관된다. 본 연구에서는 풍화 실험 데이터를 기반으로 입력 물성에 1%, 2%, 5% 수준의 불확실성을 부여하여 Monte Carlo 기반 증폭 데이터를 구축하고, 균열밀도 산정식의 구조적 민감도와 데이터 기반 민감도를 종합적으로 분석하여 입력 인자의 영향성을 고찰하는 것이다. 우선, 물리식 기반 탄력도를 통해 ε이 입력 변수에 대해 갖는 국소 민감도를 정량화한 결과, 균열밀도는 전역적으로 Vp에 가장 강하게 반응하며, Vs는 특정 조건에서 매우 높은 국소 민감도를 보이는 반면 φ는 기여도가 제한적이었다. 이어서 AI 기반 변수 중요도 분석을 수행한 결과, 세 알고리즘 모두 Vp와 Vs를 균열밀도 예측의 주요 변수로 식별하여 물리 기반 분석과 일관된 결론을 제시하였다. AI 분석에서는 σ 증가에 따라 Vs의 중요도가 상대적으로 크게 증가하는 경향이 관찰되었으며, 이는 탄력도 분석에서 확인된 Vs의 비선형 민감도 확장과 정확히 대응하였다. 본 연구는 물리 모델의 구조적 특성과 데이터 기반 경험적 민감도가 상호 보완적으로 해석될 수 있음을 보여주었으며, 균열밀도 예측 모델의 신뢰성 평가 및 입력 물성 관리에 기여할 수 있다.

References

Ahmed, S., Kaiser, M. S., Hossain, M. S., and Andersson, K. (2024), “A Comparative Analysis of LIME and SHAP Interpreters with Explainable ML-based Diabetes Predictions”, IEEE Access, 13, 37370-37388, https://doi.org/10.1109/ACCESS.2024.3422319.

10.1109/ACCESS.2024.3422319

Alahmad, T., Neményi, M., and Nyéki, A. (2025), “Soil Moisture Content Prediction Using Gradient Boosting Regressor (GBR) Model: Soil-Specific Modeling with Five Depths”, Applied Sciences, 15(11), 5889, https://doi.org/10.3390/app15115889.

10.3390/app15115889

Antonini, A. S., Tanzola, J., Asiain, L., Ferracutti, G. R., Castro, S. M., Bjerg, E. A., and Ganuza, M. L. (2024), “Machine Learning Model Interpretability Using SHAP Values: Application to Igneous Rock Classification task”, Applied Computing and Geosciences, 23, 100178.

10.1016/j.acags.2024.100178

Bae, B., Choi, Y., Yun, K., and Ahn, J. (2024), “Application of Mask R-CNN Algorithm to Detect Cracks in Concrete Structure”, Journal of The Korean Geotechnical Society, 40(3), 33-39, https://doi.org/10.7843/kgs.2024.40.3.33.

10.7843/kgs.2024.40.3.33

Byun, J. H., Lee, J. S., Park, K., and Yoon, H. K. (2015), “Prediction of Crack Density in Porous-cracked Rocks from Elastic Wave Velocities”, Journal of Applied Geophysics, 115, 110-119.

10.1016/j.jappgeo.2015.02.020

Cao, W., Zhou, A., and Shen, S. L. (2024), “Stability of Unsaturated Soil Slope Considering Stratigraphic Uncertainty and Rotated Anisotropy of Soil Properties”, International Journal for Numerical and Analytical Methods in Geomechanics, 48(2), 359-376, https://doi.org/10.1002/nag.3642.

10.1002/nag.3642

Go, G. H., Woo, H. J., Cao, V. H., Kim, H. W., Kim, Y., and Choi, H. J. (2024), “Sensitivity Analysis Study of Geotechnical Factors for Gas Explosion Vibration in Shallow-depth Underground Hydrogen Storage Facility”, Journal of the Korean Geotechnical Society, 40(4), 169-178.

10.7843/KGS.2024.40.4.169

Hajihosseinlou, M., Maghsoudi, A., and Ghezelbash, R. (2024), “Regularization in Machine Learning Models for MVT Pb-Zn Prospectivity Mapping: Applying Lasso and Elastic-net Algorithms”, Earth Science Informatics, 17(5), 4859-4873, https://doi.org/10.1007/s12145-024-01404-5.

10.1007/s12145-024-01404-5

Heo, Y. G. and Byun, Y. H. (2025), “Permanent Deformation Characteristics of Controlled Low-Strength Material Modified with Lignosulfonate”, Journal of the Korean Geotechnical Society, 41(1), 45-54.

10.1186/s40703-024-00229-x

Kim, E., Noh, S., and Lee, S. S. (2024), “Calculation of Key Blocks’ Safety Ratio based on Discontinuity Analysis”, Journal of the Korean Geotechnical Society, 40(3), 101-108.

10.7843/KGS.2024.40.3.101

Mao, Y., Li, Y., Teng, F., Nanehkaran, Y. A., Azarafza, M., and Zhang, M. (2024), “Predicting the Elasticity Modulus of Sedimentary rocks using Deep Random Forest Optimization (DRFO) Algorithm”, Environmental Earth Sciences, 83(16), 487, https://doi.org/10.1007/s12665-024-11768-y.

10.1007/s12665-024-11768-y

Meng, F., Pei, H., Ye, M., and He, X. (2024), “Stability Analysis of Reservoir Slope under Water-level Drawdown Considering Stratigraphic Uncertainty and Spatial Variability of Soil Property”, Computers and Geotechnics, 169, 106199, https://doi.org/10.1016/j.compgeo.2024.106199.

10.1016/j.compgeo.2024.106199

Mishra, P. and Pandey, R. (2025), “Comparative Analysis of Multiple Linear Regression with L1 and L2 Regularization for Stock Price Prediction”, Aadim Journal of Multidisciplinary Research, 1, 13-24.

10.3126/ajmr.v1i1.82290

Sinha, S., Rao, C. S., Kumar, A., Surya, D. V., and Basak, T. (2024), “Exploring and Understanding the Microwave-assisted Pyrolysis of Waste Lignocellulose Biomass Using Gradient Boosting Regression Machine Learning Model”, Renewable Energy, 231, 120968, https://doi.org/10.1016/j.renene.2024.120968.

10.1016/j.renene.2024.120968

Sun, Y., Zhao, Y., Hao, L., Zhao, X., Lu, J., Shi, Y., and Li, Q. (2024), “Application of the Partial Least Square Regression Method in Determining the Natural Background of Soil Heavy Metals: a Case Study in the Songhua River Basin, China”, Science of the Total Environment, 918, 170695, https://doi.org/10.1016/j.scitotenv.2024.170695.

10.1016/j.scitotenv.2024.170695

Wang, X., Zhao, C., Li, Z., and Huang, J. (2024), “Modeling Risk Assessment of Soil Heavy Metal Pollution Using Partial Least Squares and Fuzzy Logic: A Case Study of a Gully Type Coal-based Solid Waste Dumpsite”, Environmental Pollution, 352, 124147, https://doi.org/10.1016/j.envpol.2024.124147.

10.1016/j.envpol.2024.124147

Yun, H. G., Kim, I., and Baek, S. H. (2025), “Experimental Assessment of Particle Breakage in Weathered Granite Soils under Impact Loading”, Journal of the Korean Geotechnical Society, 41(4), 155-171.

10.7843/KGS.2025.41.4.155

Zhang, J. Z., Jiang, Q. H., Zhang, D. M., Huang, H. W., and Liu, Z. Q. (2024), “Influence of Geological Uncertainty and Soil Spatial Variability on Tunnel Deformation and their Importance Evaluation”, Tunnelling and Underground Space Technology, 152, 105930, https://doi.org/10.1016/j.tust.2024.105930.

10.1016/j.tust.2024.105930

Information

Publisher :The Korean Geotechnical Society
Publisher(Ko) :한국지반공학회
Journal Title :Journal of the Korean Geotechnical Society
Journal Title(Ko) :한국지반공학회 논문집
Volume : 41
No :6
Pages :195-204
Received Date : 2025-11-17
Revised Date : 2025-11-21
Accepted Date : 2025-11-22
DOI :https://doi.org/10.7843/kgs.2025.41.6.195

Journal of the Korean Geotechnical Society ISSN:1229-2427(Print) 2288-646X(Online) 한국지반공학회 논문집

All Issue