Three-dimensional slope reliability and risk assessment using auxiliary random finite element method

Auxiliary Random Finite Element Method for Risk Assessment of 3-D Slope

“…Step 2: to N(1−p 0 ) for k = 0, 1, …, m−1, and Np 0 for k = m. Using these SS samples, the preliminary slope failure probability, P f,p , and preliminary slope failure risk, R p , based on the coarse FE model can be calculated as (Xiao et al, 2016) ( )…”

“…Since samples in their close neighborhood may have similar performances, it is reasonable to select a part of samples as the representative samples in small sample space as shown in Figure 1, which is referred as the sub-binning strategy in RCM (Au, 2007) representative sample and is used as input of the fine FE model to recalculate the safety margin of slope stability. By this means, the target slope failure probability, P f,t , and target slope failure risk, R t , based on the fine FE model are calculated as (Xiao et al, 2016) ( )…”

Probabilistic analysis of strip footings based on enhanced Kriging metamodeling

“…For instance, 10 n + 2 simulations are required to compute a failure probability of 10 −n for a coefficient of variation on the failure probability of 10%. In order to overcome the shortcoming of a large number of simulations, some authors have resorted to more efficient probabilistic methods called “variance reduction techniques” (eg, Ahmed and Soubra, Yuan et al, Li et al, Jiang and Huang, Li et al, Xiao et al, Huang et al, Jiang et al, and Van Den Eijnden and Hicks). Although the variance reduction techniques are powerful probabilistic approaches, they remain insufficient when dealing with a small value of the failure probability and a small desired value of the coefficient of variation on this failure probability.…”

Cost-Effective Design of Long Spatially Variable Soil Slopes Using Conditional Simulation

“…In recent years, three dimensional reliability assessments concerning the stability of long 'linear' soil structures have been gaining increasing attention (Spencer and Hicks, 2006;2007;Spencer, 2007;Hicks et al, 2008;Griffiths et al, 2009;Hicks and Spencer, 2010;Vanmarcke, 2011;Li et al, 2013;2015a;2015c;Vanmarcke and Otsubo, 2013;Hicks et al, 2014;Li and Hicks, 2014;Ji and Chan, 2014;Ji, 2014;Li et al, 2016;Xiao et al, 2016;Varkey et al, 2016). The reasons for this are three-fold: (1) the three dimensional nature of soil spatial variability necessitates 3D analysis of geotechnical structures, as this is more realistic than a plane strain analysis which ignores the discrete 3D failure mechanisms generally encountered in practice; (2) the increasing computational power makes 3D analyses possible (Li et al, 2015b); (3) the increasing attention paid by regulatory bodies in asking for rational risk assessments and cost-effective design of important infrastructures, e.g.…”