Journal of Intelligent Geotechnical Engineering and Foundations

Articles

Vol. 1 (2025)

Machine Learning Prediction on the Ground Bearing Capacity of a Rigid Footing Resting on Ground with Graded Soil Particles using Discrete Element Method

Submitted
November 18, 2025
Published
2025-12-10

Abstract

Compared to the continuum-based methods, the discrete element method (DEM) may reflect the effects of granular characteristics (such as the mean particle size d50, coefficient of uniformity Cu and coefficient of curvature Cc) on the behavior of soil mass. Hence, in this work, the rolling resistance linear (RRL) contact model in combination with the circular particles in the DEM, as a simplified approach of reflecting actual shapes of soil particles, is utilized to assess the effects of granular characteristics on the ultimate bearing capacity pu of the ground. Furthermore, a large amount of DEM simulations of ground with various granular characteristics are conducted, so that a dataset of relating the pu to the granular characteristics for machine learning (ML) can be achieved. Based on the dataset of DEM simulations, three ML algorithms, namely multiple linear regression (MLR), artificial neural networks (ANN) and extreme gradient boosting (XGBoost), are applied to training the prediction models of pu. The research results reveal that the footing rotations and the asymmetric failure pattern of ground in the DEM simulations are largely attributed to the granular characteristics of soil mass. For a ground composed of soil particles with larger d50 and larger Cu, the pu of the ground generally tends to be higher and the thickness of developed shear band tends to be greater. For a ground composed of soil particles with larger d50 but smaller Cu, the asymmetric failure pattern of ground is more legible and the predicted pu tends to be smaller. It was also found that among the investigated ML algorithms, the MLR algorithm may provide an explicit model for the pu, while ANN and XGBoost algorithms offer higher prediction accuracy.

References

  1. Azami A, Pietruszczak S, Guo P (2010) Bearing capacity of shallow foundations in transversely isotropic granular media. Int J Numer Anal Methods Geomech 34(8): 771-793. https://doi.org/10.1002/nag.827
  2. Behera RN, Patra C (2018) Ultimate bearing capacity prediction of eccentrically inclined loaded strip footings. Geotech Geol Eng 36:3029-3080. https://doi.org/10.1007/s10706-018-0521-z
  3. Bergstra J, Bengio Y (2012) Random search for hyper-parameter optimization. J Mach Learn Res 13(2).
  4. Bhandari A, Han J (2009) DEM study of a shallow foundation under vertical loading. In: Contemporary Topics in Ground Modification, Problem Soils, and Geo-Support. pp 465-472. https://doi.org/10.1061/41023(337)59
  5. Chen X, Wu Y, Yu Y et al. (2014) A two-grid search scheme for large-scale 3-D finite element analyses of slope stability. Comput Geotech 62:203-215. https://doi.org/10.1016/j.compgeo.2014.07.010
  6. Chen X, Wang D, Yu Y et al. (2020) A modified Davis approach for geotechnical stability analysis involving non-associated soil plasticity. Géotechnique 70(12): 1109-1119. https://doi.org/10.1680/jgeot.18.P.158
  7. Chen X, Tang J, Cui L et al. (2023) Stability and failure pattern analysis of bimslope with Mohr-Coulomb matrix soil: From a perspective of micropolar continuum theory. J Cent South Univ 30(10): 3450-3466. https://doi.org/10.1007/s11771-023-5452-z
  8. Chen X, Liu Z, Cui L, Tang J (2022) An automatic identification method for width of shear band of sand in PFC simulations. Chin J Geotech Eng 44(S2): 179-182.
  9. Dong Z, Li P, Yao M, Li X, Zhao L, Wang L (2024) Comparison and analysis of denoising method in TBM key tunnelling data. Georisk Assess Manag Risk Eng Syst Geohazards: 1-14.
  10. Fathipour H, Payan M, Chenari RJ et al. (2022) General failure envelope of eccentrically and obliquely loaded strip footings resting on an inherently anisotropic granular medium. Comput Geotech 146: 104734. https://doi.org/10.1016/j.compgeo.2022.104734
  11. Fu Z, Chen S, Liu S (2016) Discrete element simulations of shallow plate-load tests. Int J Geomech 16(3): 04015077. https://doi.org/10.1061/(ASCE)GM.1943-5622.0000588
  12. Gao Z, Lu D, Du X (2020) Bearing capacity and failure mechanism of strip footings on anisotropic sand. J Eng Mech 146(8): 04020081. https://doi.org/10.1061/(ASCE)EM.1943-7889.0001814
  13. Huang N, Lu G, Xu D (2016) A permutation importance-based feature selection method for short-term electricity load forecasting using random forest. Energies 9(10): 767. https://doi.org/10.3390/en9100767
  14. Jiang M, Konrad JM, Leroueil S (2003) An efficient technique for generating homogeneous specimens for DEM studies. Comput Geotech 30(7): 579-597. https://doi.org/10.1016/S0266-352X(03)00064-8
  15. Kimura T, Kusakabe O, Saitoh K (1985) Geotechnical model tests of bearing capacity problems in a centrifuge. Géotechnique 35(1): 33-45. https://doi.org/10.1680/geot.1985.35.1.33
  16. Kingma DP, Ba J (2014) Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
  17. Krabbenhoft S, Damkilde L, Krabbenhoft K (2014) Bearing capacity of strip footings in cohesionless soil subject to eccentric and inclined loads. Int J Geomech 14(3): 04014003. https://doi.org/10.1061/(ASCE)GM.1943-5622.0000332
  18. Krishnan K, Chakraborty D (2022) Probabilistic study on the bearing capacity of strip footing subjected to combined effect of inclined and eccentric loads. Comput Geotech 141: 104505. https://doi.org/10.1016/j.compgeo.2021.104505
  19. Li J, Tian Y, Cassidy MJ (2015) Failure mechanism and bearing capacity of footings buried at various depths in spatially random soil. J Geotech Geoenviron Eng 141(2): 04014099. https://doi.org/10.1061/(ASCE)GT.1943-5606.0001219
  20. Liu Z, Chen X, Cui L, Tang J (2024) Porosity field measurement technique for shear band width in direct shear and biaxial discrete element numerical experiments. Rock Soil Mech 45(S1): 742-750.
  21. Liu Z, Chen X, Cui L, Xu Z, Yan Z (2025) Numerical analysis of the effect of the gradation characteristics of soil particles on the ground bearing capacity of a rigid footing via the discrete element method. J Cent South Univ, In Press.
  22. Loukidis D, Chakraborty T, Salgado R (2008) Bearing capacity of strip footings on purely frictional soil under eccentric and inclined loads. Can Geotech J 45(6): 768-787. https://doi.org/10.1139/T08-015
  23. Lundberg SM, Erion GG, Lee SI (2018) Consistent individualized feature attribution for tree ensembles. arXiv preprint.
  24. Lyu Y, Chen X, Tang J et al. (2023) Application assessment of hybrid smoothed finite element method for geotechnical deformation and stability analysis. J Cent South Univ 30(3): 919-933. https://doi.org/10.1007/s11771-023-5285-9
  25. Lyu Y, Chen X, Tang J et al. (2024) An implicit stabilized node-based smoothed finite element method for ultimate bearing capacity analysis of strip footing. Eng Anal Bound Elem 160: 52-64. https://doi.org/10.1016/j.enganabound.2023.12.018
  26. Pan Q, Chen X, Zhang J, Li F, Zhang D, Tang C (2024) Preface for machine learning and artificial intelligence in geotechnics: Opportunities and challenges. J Cent South Univ 31: 3819-3822. https://doi.org/10.1007/s11771-024-5842-x
  27. Rattez H, Shi Y, Sac-Morane A et al. (2022) Effect of grain size distribution on the shear band thickness evolution in sand. Géotechnique 72(4): 350-363. https://doi.org/10.1680/jgeot.20.P.120
  28. Roy N, Shree K (2024) Machine learning prediction tool for seismic bearing capacity of strip footings in rock mass. Transp Infrastruct Geotech 11(2): 900-919. https://doi.org/10.1007/s40515-023-00312-3
  29. Shiau JS, Lyamin AV, Sloan SW (2003) Bearing capacity of a sand layer on clay by finite element limit analysis. Can Geotech J 40(5): 900-915. https://doi.org/10.1139/t03-042
  30. Sundararajan M, Taly A, Yan Q (2017) Axiomatic attribution for deep networks. In: International Conference on Machine Learning. PMLR, pp 3319-3328.
  31. Tang J, Wang X, Chen X et al. (2022) Geotechnical Strain Localization Analysis Based on Micropolar Continuum Theory Considering Evolution of Internal Characteristic Length. Int J Geomech 22(8): 06022016. https://doi.org/10.1061/(ASCE)GM.1943-5622.0002462
  32. Tang J, Chen X, Cui L, Liu Z (2023) Strain localization of Mohr-Coulomb soils with non-associated plasticity based on micropolar continuum theory. J Rock Mech Geotech Eng 15(12): 3316-3327. https://doi.org/10.1016/j.jrmge.2023.02.029
  33. Tang J, Chen X, Cui L, Xu Z, Liu G (2024) Geotechnical analysis involving strain localization of overconsolidated soils based on unified hardening model with hardening variable updated by a composite scheme. Int J Numer Anal Methods Geomech 48(8): 2237-2256. https://doi.org/10.1002/nag.3730
  34. Thapa A, Adhikari K, Shrestha D K, et al. (2025) Deep Learning-Assisted Computational Modeling of Granular Soil Behavior for Foundation Design Optimization in Kharke Khola Region. Int. J. Geoinformatics 1(2): 25-45. https://doi.org/10.46610/IJGST.2025.v01i02.003
  35. Toyosawa Y, Itoh K, Kikkawa N, Yang J, Liu F (2013) Influence of model footing diameter and embedded depth on particle size effect in centrifugal bearing capacity tests. Soils Found 53(2): 349-356. https://doi.org/10.1016/j.sandf.2012.11.027
  36. Veiskarami M, Shokoohi MA (2023) A Lower Bound Estimate of the Bearing Capacity of Foundations on Inherently Anisotropic Sands Implementing the Fabric Tensor. Geotech Geol Eng 41(2): 831-845. https://doi.org/10.1007/s10706-022-02308-1
  37. Wang D, Chen X, Yu Y et al. (2019) Stability and deformation analysis for geotechnical problems with nonassociated plasticity based on second-order cone programming. Int J Geomech 19(2): 04018190. https://doi.org/10.1061/(ASCE)GM.1943-5622.0001339
  38. Wu P, Chen J, Huang J, et al. (2026) Calibration of Mesoscopic Parameters for Lunar Soil Using a Machine Learning-Based Discrete Element Method. J. Aerosp. Eng. 39(1): 04025110. https://doi.org/10.1061/JAEEEZ.ASENG-6267
  39. Wu Y, Zhou X, Gao Y et al. (2020) Bearing capacity of embedded shallow foundations in spatially random soils with linearly increasing mean undrained shear strength. Comput Geotech 122: 103508. https://doi.org/10.1016/j.compgeo.2020.103508
  40. Xiao Y, Zhao M, Zhao H et al. (2018) Finite element limit analysis of the bearing capacity of strip footing on a rock mass with voids. Int J Geomech 18(9): 04018108. https://doi.org/10.1061/(ASCE)GM.1943-5622.0001262
  41. Yin Z, Wang P (2021) Micro-mechanical analysis of caisson foundation in sand using DEM: particle shape effect. Appl Ocean Res 111: 102630. https://doi.org/10.1016/j.apor.2021.102630
  42. Zhang P, Yin Z, Jin Y (2021) State-of-the-art review of machine learning applications in constitutive modeling of soils. Arch Comput Methods Eng 28(5): 3661-3686. https://doi.org/10.1007/s11831-020-09524-z
  43. Zhang P, Yin Z, Jin Y (2022) Machine learning-based modelling of soil properties for geotechnical design: review, tool development and comparison. Arch Comput Methods Eng: 1-17. https://doi.org/10.1007/s11831-021-09615-5