Published March 2025 | Published
Journal Article Open

LS-DEM Guided Analysis of Geotechnical Tests: Exploring Strength Anisotropy and Stress Dependency

  • 1. ROR icon ETH Zurich
  • 2. ROR icon California Institute of Technology
  • 3. Effectum Medical AG
  • 4. Swiss Federal Laboratories for Materials Science and Technology Empa
An error occurred while generating the citation.

Abstract

Reliable interpretation of model tests in geotechnical engineering often is hampered by the limitations of traditional laboratory element testing, especially under low-stress conditions and unconventional stress paths. This paper presents a pragmatic hierarchical multiscale numerical approach that combines the level set discrete-element method (LS-DEM) and continuum-based analysis to improve the interpretation of scaled geotechnical tests. LS-DEM enables high-fidelity simulations of soil behavior, overcoming challenges such as boundary effects and metrological limitations that make accurate assessments under specific conditions in standard laboratory tests difficult. The virtual LS-DEM specimens can be calibrated reliably using traditional laboratory tests, such as oedometer and triaxial tests. This approach was demonstrated and implicitly validated through its application to a study of displacement-dependent earth pressure on retaining walls. Key findings include the identification of additional kinematic constraints under plane strain conditions as the primary factor behind the high peak strength observed in scaled tests, and the observation that the strength of granular materials is negligibly affected by changes in stress level. In addition, the observation of pressure-dependent elastic parameter trends, consistent with previous studies, further validates LS-DEM as a reliable tool for quantitatively capturing the behavior of granular materials. By reducing reliance on semiempirical scaling laws and providing a robust framework for informing continuum-based models, this LS-DEM–based hierarchical approach effectively bridges the gap between small-scale laboratory experiments and large-scale geotechnical applications. Ultimately, this methodology enhances the design and analysis of geotechnical structures with greater confidence and accuracy, providing a practical and effective tool for addressing complex geotechnical engineering challenges.

Copyright and License

© 1996-2025 American Society of Civil Engineers. All rights reserved, including rights for text and data mining and training of artificial intelligence or similar technologies.. This work is made available under the terms of the Creative Commons Attribution 4.0 International license, https://creativecommons.org/licenses/by/4.0/.

Acknowledgement

The authors thank Dr. Balz Friedli, Dr. Dominik Hauswirth, and Dr. Marc Kohler from ETH Zurich, as well as Dr. Jason Marshall and Dr. Reid Kawamoto from Caltech, for their valuable discussions and insights. Special thanks are given to Prof. Johan Clausen for generously sharing the Abaqus implementation of the Mohr–Coulomb constitutive law and to Dr. Reid Kawamoto for providing his original LS-DEM code. The numerical simulations were performed on the Euler cluster, operated by the High Performance Computing group at ETH Zurich. This work was partially supported by the Swiss Federal Roads Office and the Swiss Federal Office of Transport (Research project AGB 2015/029).

Data Availability

Some data, models, or code generated or used during the study are available in a repository online in accordance with funder data retention policies. A data set containing the experimental wall test results is available in Perozzi and Puzrin (2023). Other data, models, or code that support the findings of this study are available from the corresponding author upon reasonable request, including the results of the virtual element tests.

Files

perozzi-et-al-2025-ls-dem-guided-analysis-of-geotechnical-tests-exploring-strength-anisotropy-and-stress-dependency.pdf

Additional details

Created:
March 12, 2025
Modified:
March 12, 2025