Application of stochastic fields in geotechnical problems

Abstract

This work focuses on the probabilistic analysis of slope stability and rigid shallow footing problems. For this purpose, mathematical models based on the Finite Element Method (FEM), Montecarlo (MC) method and Local Average Subdivision (LAS) procedure are studied. The LAS procedure is used to generate random fields, which properly represent the associated uncertainties in the properties of the materials. The FEM focuses on the numerical response of the problem in terms of displacements and stresses. The plasticity of the soil can be included via a visco-plastic algorithm beside a Mohr-Coulomb law. The LAS and MEF procedures are implemented in the framework of a MC analysis, where each MC execution requires several simulations of the problem at hand. This permits to quantify the failure probability of the system and report the most probable settlement to occur in the case of shallow foundations. After many executions of the numerical model, it is suggested that at least 4000 and 500 simulations are needed for the slope stability and shallow foundation problems, respectively, in order to obtain stable and reliable values. The obtained results show that the failure probability of the slope is relatively low and equal to 0.18, while the expected settlement of a shallow rigid foundation is around 1.96 cm.