The increasing use of finite elements in civil engineering, and more specifically in geotechnics, allows for the rapid attainment of more precise results. However, the inputs to the various developed models are crucial and failing to use the relevant mechanics can negatively impact the results. Additionally, there are numerous uncertainties governing geomechanical parameters, as well as the potential effect of structural elements, such as piles, added to the soil causing its disturbance. In this complex situation, it is not uncommon for reality to differ from engineers’ predictions. Thus, De Cerenville engineers observed displacements of 80 to 90 mm on sheet piles in one of its projects in Echandens, despite forecasts around 3 cm. In light of this mystery, numerous hypotheses emerged. This thesis aimed to explain on this discrepancy between the initial models and the actual soil behavior. The most rational hypothesis was the excess pore pressure induced by pile driving in the soft clays present in the glaciolacustrine deposits layer. To model the situation, the choice was made to use OPTUM G2, a finite element software gaining notoriety in the geotechnical field. A constituve law was selected to match the expected behavior of the soil layers and implement ideally the available soil properties. Several techniques were used to simulate the excess pore pressure in the finite elements model and evaluated for the adequacy with the displacements data. The study thus concluded on the probable cause of the observed displacements, its implementation in the software used, and commented on the measures to be taken in future projects.