All Issue

2026 Vol.42, Issue 1 Preview Page
Probabilistic Stability Analysis of River Levees Using Machine-Learning-Based Surrogate Model

기계학습 기반의 대리 모델을 이용한 하천 제방의 확률론적 안정해석

Sung-Eun Cho1*
조 성은1*
1Member, Prof., School of Civil and Environmental Engineering & Construction Engineering Research Institute, Hankyong National Univ
1정회원, 한경국립대학교 건설환경공학부 교수
*Corresponding Author
28 February 2026. pp. 137-150
PDF XML Citation
Abstract

Numerical analysis techniques have been widely used to evaluate the stability of river levees by capturing the complex hydraulic and mechanical behavior of geomaterials. However, river levees are designed to maintain extremely low failure probabilities; therefore, probabilistic analyses based on the direct application of Monte Carlo simulation (MCS) with repeated numerical analyses result in excessive computational costs. To address this limitation, this study applies machine learning techniques to the probabilistic stability analysis of river levees. Various machine learning models were trained using input–output data obtained from numerical analyses, and their predictive performance was evaluated using statistical metrics. A Gaussian process regression model, which exhibited the highest predictive accuracy, was selected as a surrogate model for MCS in probabilistic levee stability analysis. The results indicate that the application of a machine-learning-based surrogate model substantially improves computational efficiency compared to conventional probabilistic approaches based on numerical analysis, enabling stable and reliable evaluation of piping and slope failure probabilities and fragility curves.

Keywords
Earthen levee
Machine learning
Probabilistic analysis
Seepage
Slope stability
Surrogate model

수치해석 기법은 지반의 복잡한 수리학적 및 역학적 거동을 분석하는 강력한 도구로 하천 제방의 안정해석에 널리 활용되고 있다. 그러나 하천 제방은 매우 작은 파괴확률 수준을 유지하도록 설계되므로 수치해석을 반복적으로 수행하는 직접적인 Monte Carlo Simulation(MCS)의 적용에 의한 확률론적 해석은 과다한 계산 비용을 유발한다. 이러한 단점을 극복하기 위하여 본 연구에서는 제방의 확률론적 안정해석에 기계학습을 적용하였다. 수치해석의 입력-출력 데이터를 기반으로 다양한 기계학습 모델의 학습을 수행하고 통계적 지표를 사용하여 학습된 모델의 예측성능을 평가하였다. 최종적으로 예측성능이 우수한 Gaussian Process Regression(GPR) 모델을 선정하여 제방의 확률론적 안정해석을 위한 MCS에 활용하였다. 연구 결과, 기계학습 기반의 대리 모델을 적용하면 기존 수치해석 기반의 확률론적 해석에 비해 계산 효율성의 큰 향상이 가능하여, 파이핑 및 사면 파괴 확률과 취약도 곡선을 안정적으로 평가할 수 있음을 확인하였다.

References
1

Ahn, K. H. (2009), Combined Reliability Model for Levee Embankment Integrated Stochastic Characterization of Rainfall Variation, Doctoral Thesis, Kyungpook National University (in Korean).

2

Cho, S. E. (2007), “Effects of Spatial Variability of Soil Properties on Slope Stability”, Engineering Geology, Vol.92, No.3-4, pp.97-109.

10.1016/j.enggeo.2007.03.006
3

Cho, S. E. (2009), “Probabilistic Stability Analyses of Slopes using the ANN-based Response Surface”, Computers and Geotechnics, Vol.36, pp.787-797.

10.1016/j.compgeo.2009.01.003
4

Cho, S. E. (2022), “Applicability of Practical Reliability Analysis to Develop Fragility Curves for Levee”, Journal of the Korean Geotechnical Society, Vol.38, No.11, pp.19-30 (in Korean).

10.7843/KGS.2022.38.11.19
5

Cho, S. E. (2025), “Probabilistic Stability Analysis of Earthen Levees Considering the Hydraulic and Mechanical Properties of Soil”, Journal of the Korean Geotechnical Society, Vol.41, No.1, pp.55-68 (in Korean).

10.7843/KGS.2025.41.1.55
6

Duncan, J. M. (2000), “Factors of Safety and Reliability in Geotechnical Engineering”, Journal of Geotechnical and Geoenvironmental Engineering, Vol.126, No.4, pp.307-316.

10.1061/(ASCE)1090-0241(2000)126:4(307)
7

Fenton, G. A. and Griffiths, D. V. (1997), “Extreme Hydraulic Gradient Statistics in Stochastic Earth Dam”, Journal of Geotechnical and Geoenvironmental Engineering, Vol.123, No.11, pp.995-1000.

10.1061/(ASCE)1090-0241(1997)123:11(995)
8

Fredlund, D., Morgenstern, N. R., and Widger, R. (1978), “The Shear Strength of Unsaturated Soils”, Canadian Geotech J, Vol.15, No.3, pp.313-321.

10.1139/t78-029
9

Gao, K., Cheng, Z., Song, Y., Yin, S., and Chen, Y. (2025), “Machine Learning-Driven Surrogate Model Development for Geotechnical Numerical Simulation”, Geotechnical Research, Vol.12, No.2, pp.71-84.

10.1680/jgere.24.00029
10

GeoStudio (2012), Seepage Modeling with SEEP/W and Slope Stability Analysis with SLOPE/W, Geo-Slope International Ltd, Canada.

11

Ji, J. and Cao, Z. (2024), “Geotechnical Reliability Analysis for Practice”, In Phoon, K. K., Shuku, T., and Ching, J., eds., Uncertainty, Modeling, and Decision Making in Geotechnics, CRC Press, Boca Raton, FL, pp.219-246.

10.1201/9781003333586-8
12

Kang, T. H., Choi, S. W., Lee, C. H., and Chang, S. H. (2024), “A Fundamental Study on the Prediction of Key Operating Parameters using Slurry Shield TBM Excavation Data and Machine Learning Algorithm”, Tunnel and Underground Space, Vol.34, No.6, pp.709-721.

10.7474/TUS.2024.34.6.709
13

Kim, J. M. (2006), “Geotechnical Aspects of River Levee Stability”, Water For Future, Vol.39, No.5, pp.20-27 (in Korean).

14

Kim, J. M., Park, M. C., Cho, W. B., and Han, H. S. (2016), “Influence Assesment Due to Damage of Landside Slope on River Embankment”, J. Korean Soc. Hazard Mitig., Vo.16, No.4, pp.203-213 (in Korean).

10.9798/KOSHAM.2016.16.4.203
15

MathWorks (2025), MATLAB (R2025a), The MathWorks Inc., Natick, Massachusetts.

16

Mitelman, A., Yang, B., Urlainis, A., and Elmo, D. (2023), “Coupling Geotechnical Numerical Analysis with Machine Learning for Observational Method Projects”, Geosciences, Vo.13, No.7, 196.

10.3390/geosciences13070196
17

Mualem, Y. (1976), “A New Model for Predicting the Hydraulic Conductivity of Unsaturated Porous Media”, Water Resources Research, Vol.12, No.3, pp.513-522.

10.1029/WR012i003p00513
18

Nam, M. J., Lee, J. Y., Lee, C. W., and Kim, K. Y. (2017), “Estimating the Compound Risk Integrated Hydrological/Hydraulic/Geotechnical Uncertainty of Levee Systems”, J. Korea Water Resour. Assoc., Vol.50, No.4, pp.277-288 (in Korean).

10.3741/JKWRA.2017.50.4.277
19

Puri, N., Prasad, H. D., and Jain, A. (2018), “Prediction of Geotechnical Parameters Using Machine Learning Techniques”, Procedia Computer Science, Vol.125, pp.509-517.

10.1016/j.procs.2017.12.066
20

Rauter, S. and Tschuchnigg, F. (2021), “CPT Data Interpretation Employing Different Machine Learning Techniques”, Geosciences, Vol.11, No.7, 265.

10.3390/geosciences11070265
21

Richards, L. A. (1931), “Capillary Conduction of Liquids through Porous Mediums”, J Appl Phys, Vol.1, No.5, pp.318-333.

10.1063/1.1745010
22

Sharafati, A., Yaseen, Z. M., and Pezeshki, E. (2020), “Strategic Assessment of Dam Overtopping Reliability Using a Stochastic Process Approach”, Journal of Hydrologic Engineering, Vol.25, No.7, 04020029.

10.1061/(ASCE)HE.1943-5584.0001938
23

Siacara, A.T., Napa-García, G.F., Beck, A.T., and Futai, M. M. (2020), “Reliability Analysis of an Earth Dams Using Direct Coupling”, Journal of Rock Mechanics and Geotechnical Engineering, Vol.12, No.2, pp.366-380.

10.1016/j.jrmge.2019.07.012
24

van Genuchten, M. T. (1980), “A Closed-form Equation for Predicting the Hydraulic Conductivity of Unsaturated Soils”, Soil Science Society America Journal, Vol.44, No.5, pp.892-898.

10.2136/sssaj1980.03615995004400050002x
25

Wu, S., Zhang, J. M., and Wang, R. (2021), “Machine Learning Method for CPTu Based 3D Stratification of New Zealand Geotechnical Database Sites”, Advanced Engineering Informatics, Vol.50, 101397.

10.1016/j.aei.2021.101397
26

Zhang, P., Yin, Z. Y., and Jin, Y. F. (2022), “Machine Learning-Based Modelling of Soil Properties for Geotechnical Design: Review, Tool Development and Comparison”, Archives of Computational Methods in Engineering, Vol.29, pp.1229-1245.

10.1007/s11831-021-09615-5
27

Zheng, G., Zhang, W., Zhang, W., Zhou, H., and Yang, P. (2021), “Neural Network and Support Vector Machine Models for the Prediction of the Liquefaction-Induced Uplift Displacement of Tunnels”, Underground Space, Vol.6, No.2, pp.126-133.

10.1016/j.undsp.2019.12.002
Information
  • Publisher :The Korean Geotechnical Society
  • Publisher(Ko) :한국지반공학회
  • Journal Title :Journal of the Korean Geotechnical Society
  • Journal Title(Ko) :한국지반공학회 논문집
  • Volume : 42
  • No :1
  • Pages :137-150
  • Received Date : 2026-01-29
  • Revised Date : 2026-02-18
  • Accepted Date : 2026-02-23
  • DOI :https://doi.org/10.7843/kgs.2026.42.1.137
Journal Informaiton Journal of the Korean Geotechnical Society Journal of the Korean Geotechnical Society
Archives
: Vol.23~28 (2007~2012)
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close