Identification of Predisposing Factors for Mining-Induced Ground Collapses Using Data Analysis and 3D Numerical Modelling

The research aims to improve the methodology used to identify areas at risk of ground collapse or subsidence within mining districts in Lorraine that were formerly worked using the abandoned room-and-pillar mining method.

The research is divided into two parts. The first focuses on identifying the predisposing factors for sudden ground collapses and progressive subsidence through data analysis, including machine learning techniques. The second focuses on the development of coupled hydro-mechanical numerical models, primarily in 3D, as well as meta-models.