ABSTRACT
In this paper, probabilistic back analysis of lateritic soil cutting MIDC Zarap is conducted using Bayesian approach for obtaining the slope material parameters such as cohesion c’, friction ϕ’ and unit weight γ. The geometry, input mean of random parameters c’, ϕ’ and γ, and values of other deterministic properties for MIDC Zarap cutting are adopted from Rahul et al. (2023). The normal random fields of c’, ϕ’ and γ are generated for input coefficient of variation COV of 10% and different ratios of the correlation lengths i.e. CLx/CLy of 1,5,10,20, using Karhunen-Loéve (K-L) expansion method in Optum G2 software. The effects of isotropic and anisotropic correlation lengths are studied on the generated random fields and values of output COV. The prior and posterior probability density functions of random variables are then established using Bayesian methodology in MATLAB for different output COVs based on different ratios of the correlation lengths.
