Future of machine learning in geotechnics
Citation
Phoon, K.K., Zhang, W. (2023). Future of machine learning in geotechnics. Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards, 17(1): 7-22. Link to paper
Abstract
This perspective paper outlines a data-centric agenda for the next stage of machine learning in geotechnical engineering. Rather than focusing only on algorithm novelty, it emphasizes three core elements: data centricity, practice-centered deployment, and explicit geotechnical context. The authors discuss challenges such as poor-quality field data, explainability for site characterization, and integration with engineering workflows. They also identify future opportunities including digital twins, meta-learning, and ML systems that become operationally indispensable. The framework provides strategic guidance for aligning ML research with real geotechnical decision needs.