Abstract
Spatial variability of soil properties is an important factor that affects the probability of slope failure. Random finite element method (RFEM) is the most effective method to evaluate the probability of failure of slopes. In this study, the probability of slope failure is calculated using Monte Carlo simulation based on displacement based criteria. The probability of failure of an undrained cohesive slope is studied using RFEM, first order second moment (FOSM) method and FOSM combined with variance reduction theory, called modified FOSM (MFOSM). A comparison among FOSM, MFOSM and RFEM is shown for different values of undrained shear strength and scales of fluctuation. As FOSM does not consider the spatially varying nature of soil properties, it gives a constant probability of failure for varying scales of fluctuation, a parameter that represents the spatial variability of soil properties. In general, MFOSM has similar trend with RFEM, however, MFOSM cannot demonstrate the failure mechanism for spatially varying soils. The use of RFEM and determination of realistic scales of fluctuation, coefficient of variation of soil properties is recommended for slope reliability analysis.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call