Application of generative AI to automate numerical analysis and synthetic data generation in geotechnical engineering

Purpose This study explores the integration of generative artificial intelligence (AI) into numerical analysis workflows in geotechnical engineering to address the challenges of generating synthetic datasets. This study aims to create a framework that allows practitioners with limited programming skills to automate complex simulations, enabling the development of extensive data sets for AI and machine learning applications. Design/methodology/approach The study proposes a seven-step methodology using the finite element method and Python programming to auotmate numerical modelling. Generative AI, specifically ChatGPT, is used as a virtual assistant to guide practitioners through automation. The methodology is validated through a pilot study predicting excavation-induced ground displacement in Sydney’s Hawkesbury Sandstone. Findings Integrating generative AI into numerical analysis workflows accelerates data generation and improves the quality of synthetic datasets. The pilot study indicates that the generated datasets closely align with real-world measurements, confirming the robustness and reliability of the proposed framework. Research limitations/implications The study’s accuracy may be affected by assumptions in numerical analysis and input parameter quality. Future research should explore more complex geotechnical conditions, such as 3D effects, to further validate and enhance the methodology. Practical implications This framework provides an efficient solution for geotechnical practitioners to generate extensive datassets for AI training, reducing reliance on experienced programmers. It streamlines workflows and enhances data-driven decision-making in geotechnical engineering. Originality/value The paper introduces a novel integration of generative AI into numerical analysis workflows, offering an innovative approach to automate synthetic data generation. It serves as a valuable tool for advancing AI applications in geotechnical engineering, particularly for those with limited programming experience.