Reliability assessment of reinforced slopes with unknown probability distribution

Abstract

In the proposed study, reliability assessment of the reinforced slope (RFS) is carried out using an efficient and accurate technique of Fourth Moment Normal Transformation (FMNT). The probabilistic analysis is performed using both analytical and numerical methods. FMNT can estimate the probability of failure (Pf) of a RFS with unknown marginal distribution of input random variables. Only the first four moments of any random variable with unknown distribution are required to estimate the Pf of RFS. The use of FMNT with commercially available numerical software packages is precisely explained. The accuracy of the proposed technique, when used with different distributions of random variables, is also illustrated. The present results show considerable efficiency of FMNT in estimating the Pf when used in the analytical domain as well as with a numerical software. A detailed comparison in terms of the efficiency of the proposed formulation is also made with similar literature. FMNT is very useful for the designers to perform the reliability-based analysis of RFS. The present analytical method is also capable of incorporating the pseudo-static seismic forces into calculations.