Probabilistic Stability Design Charts for Shallow Passive Trapdoors in Spatially Variable Clays
Jim Shiau, Suraparb Keawsawasvong
Abstract
Geotechnical engineers are faced with great uncertainty with regard to the stability design of soil structures such as retaining walls, foundations, and slope stability. Instead of using a deterministic approach by replacing the design parameter with a single mean number, Monte Carlo simulations could be used to provide better decisions by considering all possible outcomes under parametric uncertainty. This paper studies probabilistic failures of classical passive trapdoors in shallow depths with spatially random soils. The effects of soil strength variability are investigated for various spatial correlation lengths and trapdoor depths using the latest adaptive finite-element upper-bound limit analysis with second-order cone programming. In addition, probabilistic stability results are compared with those published in renowned literature, and a comprehensive probability of design failure (PF) charts is developed for a practical range of deterministic factors of safety. The extensive results reported in this paper would be of great interest to engineering practitioners because better decisions can be made in the design process and design confidence improved.