Reliability analysis of a strip footing designed against settlement

Abstract

The traditional design of a strip footing for settlement on a sandy soil proceeds by first testing the soil at a limited number of locations to assess its elastic modulus (for example by CPT). Then a simplified settlement design relationship is commonly employed to determine the required footing width. In a real soil, the soil tests may or may not be representative of the average elastic modulus under the footing due to spatial variability. This paper reports on a Monte Carlo investigation in which a spatially variable soil mass is simulated and virtual samples taken at a small number of locations are used to determine the required footing width. The resulting footing is then placed on the simulated soil mass and the actual settlement computed using the finite element method. By repeating this process a large number of times, the reliability of a simplified design approach, based on a limited soil sample, against excess settlement can be assessed. In particular, the paper reports on the reliability of a modified Janbu design approach with respect to the soil’s variance and scale of fluctuation. The overall goal is to investigate the reliability of existing design methodologies with a view towards reliability-based code development.