The uncertainty importance measure of slope stability based on the moment-independent method

It is well known that there are many uncertainties in slope engineering, which have great impacts on slope stability models. However, the uncertainty importance measure is rarely applied in slope stability analysis. In this paper, the moment-independent method of the uncertainty importance measure is first developed to analyze slope stability, in which the low deviation sequence of Sobol is used to simulate the geotechnical parameter samples, and a nonparametric method of the kernel density estimate is employed to estimate the probability density functions of the output responses. In addition, the Gaussian copula is applied to construct the joint distribution of the shear strength parameters in the uncertainty importance measure of the correlation parameters. Two examples are shown to indicate that the impacts of geotechnical parameters on slope stability models are greatly different. The importance sequences obtained by the moment-independent method are in good agreement with those obtained by the variance-based Monte Carlo method. The importance measure indexes clearly state that the uncertainty of the geotechnical parameters significantly affects slope stability models, and the influence of the negative correlation between the shear strength parameters on the uncertainty importance measure should not be neglected.