This paper presents a probabilistic analysis of the swelling character of Paris plastic clay using a tunnel of the Grand Paris Express project as a case study. Known for its highly plastic behavior, Paris plastic clay originates from the Sparnacian age (Lower Eocene) and constitutes the first Tertiary deposits of the Paris sedimentary basin, often overconsolidated with low permeability.
In addition, this plastic clay exhibits a strong swelling capacity and has already resulted in several accidents in the past. Hence, careful in situ investigations and laboratory tests were conducted to properly analyze Paris plastic clay. Due to the wide dispersion of swelling pressure, probabilistic analyses using finite element modelling of a tunnel laying on plastic clay were carried out. A sensitivity study was also performed to determine the key parameters of the swelling modelling. After that, the monitoring data of in-field swelling pressure was compared with the results of the numerical simulation. Based on these field measurements, Bayesian inverse analysis was employed to update the prior hypotheses so as to improve the posterior results, namely the prediction of the swelling pressure of the plastic clay and the heave of the tunnel raft.