Finite Element Modeling of Settlements on Spatially Random Soil

Abstract

The effect of a random and spatially correlated soil stiffness on the total settlement under the center of a uniformly loaded flexible-strip footing has been studied. Random field theory has been combined with the finite-element method to perform “Monte Carlo” simulations of the settlement problem with a variable Young's modulus and a constant Poisson's ratio. The soil Young's modulus field has been simulated with a fixed mean, standard deviation, and spatial correlation structure using the local average subdivision (LAS) method. The results of parametric studies have been compared with the deterministic quantities to gauge the effect of the standard deviation, correlation structure, and geometry on the settlement behavior. The results indicate a modest increase in expected settlement for presumptive ranges of soil stiffness variability.