All Issue

2025 Vol.41, Issue 4 Preview Page
ChatGPT-driven Numerical Modeling for Geotechnical Applications

지반 해석을 위한 ChatGPT 기반 수치해석 모델링

Taegu Kim1
,
Daehyun Kim2
,
Tae-Sup Yun3*
김 태구1
,
김 대현2
,
윤 태섭3*
1Graduate Student, Dept. of Civil and Environmental Eng., Yonsei Univ.
2Member, Prof., Dept. of Construction, Environment, and Energy Eng., Suwon Univ.
3Member, Prof., Dept. of Civil and Environmental Eng., Yonsei Univ.
1비회원, 연세대학교 건설환경공학과 석사과정
2정회원, 수원대학교 건설환경에너지공학부 조교수
3정회원, 연세대학교 건설환경공학과 교수
*Corresponding Author
31 August 2025. pp. 99-110
PDF XML Citation
Abstract

This study investigates the potential and limitations of automating geotechnical numerical modeling using the large language model (LLM), ChatGPT. Two representative cases were examined: seepage analysis using the finite difference method (FDM) and hydromechanical coupled settlement analysis using the finite element method (FEM). ChatGPT successfully generated and refined code from natural language prompts, demonstrating strengths in code structuring and iterative improvement. However, frequent errors occurred in the implementation of boundary conditions for FDM and in the mathematical formulation for FEM. Both cases required iterative debugging and prompt refinement, with expert intervention essential for ensuring physical validity in FEM. These results suggest that, with sufficient domain knowledge and well-designed prompts, LLMs can serve as effective collaborators in automating geotechnical numerical modeling.

Keywords
ChatGPT
Code generation
Finite element method
Large language model
Numerical modeling

본 연구는 대형언어모델(Large Language Model, LLM)인 ChatGPT를 활용하여 지반공학 수치해석 모델링의 자동화 가능성과 한계를 분석하였다. 대표 사례로 유한차분법(Finite Difference Method, FDM) 기반 침투 해석과 유한요소법(Finite Element Method, FEM) 기반 수리-역학적 연동 침하 해석을 선정하고, 자연어 프롬프트를 통한 코드 생성 및 반복 보완 과정을 수행하였다. ChatGPT는 코드 구조의 체계적 구성과 반복적인 수정 작업의 지원에서 강점을 보였으나, FDM에서는 경계 조건 구현 오류, FEM에서는 수치 불안정성과 약형 구성의 물리적 오류가 반복적으로 발생하였다. 두 사례 모두에서 반복적 프롬프트 보완이 필요하였고, 특히 FEM의 경우 물리적 타당성을 확보하기 위한 전문가의 개입이 필수적이었다. 이러한 결과는 문제에 대한 충분한 이해와 정교한 프롬프트 설계를 전제로 할 때, LLM이 지반공학 수치해석의 자동화에 있어 효과적인 협력 도구가 될 수 있음을 시사한다.

References
1

Benz, T. and Nordal, S. (2010), Numerical Methods in Geotechnical Engineering, Proceedings of the 7th European Conference on Numerical Methods in Geotechnical Engineering, Trondheim, Norway, Taylor & Francis, London, https://doi.org/10.1201/b10551.

10.1201/b10551
2

Borja, R. I. and Alarcón, E. (1995), “A Mathematical Framework for Finite Strain Elastoplastic Consolidation. Part 1: Balance Laws, Variational Formulation, and Linearization”, Computer Methods in Applied Mechanics and Engineering, Vol.122, No.1-2, pp.145-171, https://doi.org/10.1016/0045-7825(94)00720-8.

10.1016/0045-7825(94)00720-8
3

Elgamal, A. (2021), “Guidelines for Geotechnical Finite-element Modeling”, The Open Civil Engineering Journal, Vol.15, pp.424-440, https://doi.org/10.2174/1874149502115010424.

10.2174/1874149502115010424
4

Esfandiari, R. S. (2017), Numerical methods for engineers and scientists using MATLAB, 2nd Edition, CRC Press, Boca Raton, https://doi.org/10.1201/9781315152417.

10.1201/9781315152417
5

Fukuchi, T. (2016), “Numerical Analyses of Steady-state Seepage Problems Using the Interpolation Finite Difference Method”, Soils and Foundations, Vol.56, No.4, pp.608-626, https://doi.org/10.1016/j.sandf.2016.07.003.

10.1016/j.sandf.2016.07.003
6

Kashefi, A. and Mukerji, T. (2023), “ChatGPT for Programming Numerical Methods”, Journal of Machine Learning for Modeling and Computing, Vol.4, No.2, pp.1-74, https://doi.org/10.1615/JMachLearnModelComput.2023048492.

10.1615/JMachLearnModelComput.2023048492
7

Kim, J., Tchelepi, H. A., and Juanes, R. (2011), “Stability and Convergence of Sequential Methods for Coupled Flow and Geomechanics: Fixed-stress and Fixed-strain Splits”, Computer Methods in Applied Mechanics and Engineering, Vol.200, pp.1591-1606, https://doi.org/10.1016/j.cma.2010.12.022.

10.1016/j.cma.2010.12.022
8

Kumar, K. (2024), “Geotechnical Parrot Tales (gpt): Harnessing Large Language Models in Geotechnical Engineering”, Journal of Geotechnical and Geoenvironmental Engineering, Vol.150, No.1, 02523001, https://doi.org/10.1061/JGGEFK.GTENG-11828.

10.1061/JGGEFK.GTENG-11828
9

Logg, A., Mardal, K. A., and Wells, G. N. (2012), Automated Solution of Differential Equations by the Finite Element Method: The FEniCS book, Springer, Berlin, https://doi.org/10.1007/978-3-642-23099-8.

10.1007/978-3-642-23099-8
10

Na, S. H. and Kim, S. J. (2024), “Dynamic Thermo-hydro-mechanical Modeling of a Saturated Porous Medium Using One-dimensional Finite Element Analysis”, Journal of the Korean Geotechnical Society, Vol.40, No.6, pp.79-86, https://doi.org/10.7843/kgs.2024.40.6.79.

10.7843/kgs.2024.40.6.79
11

Shin, H. and Kim, J. W. (2020), “Numerical Formulation of Axisymmetric Shell Element and its Application to Geotechnical Problems”, Journal of the Korean Geotechnical Society, Vol.36, No.12, pp.27-34, https://doi.org/10.7843/kgs.2020.36.12.27.

10.7843/kgs.2020.36.12.27
12

Sim, J. U., Jeong, S. S., and Lee, J. H. (2015), “Numerical Analysis of Self-supported Earth Retaining Wall with Stabilizing Piles”, Journal of the Korean Geotechnical Society, Vol.31, No.5, pp.35-46, https://doi.org/10.7843/kgs.2015.31.5.35.

10.7843/kgs.2015.31.5.35
13

Sun, W., Ostien, J. T., and Salinger, A. G. (2013), “A Stabilized Assumed Deformation Gradient Finite Element Formulation for Strongly Coupled Poromechanical Simulations at Finite Strain”, International Journal for Numerical and Analytical Methods in Geomechanics, Vol.37, No.16, pp.2755-2788, https://doi.org/10.1002/nag.2161.

10.1002/nag.2161
14

Wu, H. and Zhu, L. (2025), “Using Large Language Models to Automate Code Generation for PDE Solvers”, Proc. of the AI4X 2025 International Conference, Singapore.

15

Wu, S., Otake, Y., Mizutani, D., Liu, C., Asano, K., Sato, N., Saito, T., Baba, H., Fukunaga, Y., Higo, Y., Kamura, A., Kodama, S., Metoki, M., Nakamura, T., Nakazato, Y., Shioi, A., Takenobu, M., Tsukioka, K., and Yoshikawa, R. (2025), “Future-proofing Geotechnics Workflows: Accelerating Problem-solving with Large Language Models”, Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards, Vol.19, No.2, pp.307-324, https://doi.org/10.1080/17499518.2024.2381026.

10.1080/17499518.2024.2381026
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 :4
  • Pages :99-110
  • Received Date : 2025-07-07
  • Revised Date : 2025-08-18
  • Accepted Date : 2025-08-19
  • DOI :https://doi.org/10.7843/kgs.2025.41.4.99
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