MACHINE LEARNING
Geotechnical engineering is a discipline that frequently uses empirical relationships to estimate soil properties and thus can greatly benefit from ML.
The application of ML methods in geotechnical engineering and soil mechanics dates back to the early 1990’s. However, with increasing computational power, ML has gained substantial interest within the geotechnical engineering community and has gradually become an alternative solution for geotechnical problems. In recent years, several studies have investigated the use of ML for piezocone penetration test (CPTu) data interpretations and soil classifications. This technical bulletin will present three applications of ML for CPTu interpretation and site characterization. These include estimation of mine tailings solids and fines, estimation of soil unit weight, and estimation of shear wave velocity from CPTu data.
QUICK LINKS