Efficient and Conservative Estimation of Failure Probability of Strip Footing on Spatially Variable Soil Using Random Finite Element Limit Analysis

The present work deals with a probabilistic analysis of the bearing capacity of shallow foundation on spatially variable cohesive-frictional soils. The spatial variability of soil parameters is modelled by anisotropic random fields. The analysis of both the upper and lower bound of the bearing capacity is carried out using a random finite element limit analysis. The analyses carried out show that the results in the lower bound provide conservative estimation of the results in the mean value, and the upper bound provide conservative estimation of the standard deviation of bearing capacity. The combined approach, consisting of describing the bearing capacity using the probability distribution with the mean value obtained from the lower bound approach and the standard deviation obtained from the upper bound approach, allows for conservative, precise and efficient estimation of the probability of structural failure.

Keywords: random field, scale of fluctuation, random finite element limit analysis, strip foundation