Abstract
When geotechnical reliability analysis are performed using the stochastic finite element method, a proper estimation of the spatial distributions of soil properties is important. In this paper, to give the best linear unbiased estimator of the unknown space parameter of soil properties, the local estimation technique, which is called “kriging”, is incorporated in the stochastic finite element method so that the estimation is made based on the in-site sampling data. This paper discusses the applicability of this proposed approach using the first-order Gaussian approximation by comparing the results with those obtained by the conditional simulation for the same problem.
