Coupling Geotechnical Numerical Analysis with Machine Learning for Observational Method Projects

In observational method projects in geotechnical engineering, the final design is decided upon during actual construction, depending on observed behavior of ground. Hence, engineers must be prepared to make crucial decisions promptly, with few available guidelines. this paper, we propose coupling numerical analysis machine learning (ML) algorithms for enhancing decision process projects. The proposed methodology consists two main computational steps: (1) data generation, where multiple models are automatically generated according anticipated range input parameters, and (2) analysis, parameters model results analyzed ML models. Using case study Semel tunnel Tel Aviv, Israel, demonstrate how can contribute success through computation feature importance, which assist better identifying key features that drive failure prior project execution, providing insights regarding monitoring plan, as correlative relationships between various tested, (3) instantaneous predictions construction.