Abstract
ABSTRACT This paper represents multivariate fragility functions (MFFs) based on sliding displacement of soil slopes for landslide hazard assessment. The MFFs correlate the sliding displacement probability of slopes to earthquake intensity measures (IMs). Conditional distributions are employed to develop MFFs for failure probability analysis of slopes by considering scalar and vector IMs in the absence of any predetermined predictive model for sliding displacement of slopes. Three sliding displacement-based damage states are defined herein as the low, moderate, and high levels of landslide hazard. The results demonstrate that the probability of exceedance from a given damage state strongly depends on the type and the number of IMs chosen for the fragility analysis whereas the predicted failure probabilities resulted from the scalar and multivariate analyses can vary up to three times. Results of this study indicates that the multivariate fragility analyses can lead to landslide damage estimation in more reliable manner, compared with the scalar-valued fragility curves which are commonly used to predict the seismic hazard of earth slopes.
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