Variability of local stress states resulting from the application of Monte Carlo and finite difference methods to the stability study of a selected slope

In slope stability analysis it is decisive to represent, in the most realistic way, the heterogeneity of the soil characteristics. The Monte Carlo technique is one of the suitable mathematical tools that could be useful for this purpose. The result of this kind of approach is a set of multiple possible realizations of the same system, characterized by different spatial distribution of the numerical values of the geotechnical parameters. The numerical stress states outcomes, obtained through a stability analysis, would depend on the models employed to define the spatial parameters distribution and, further, on the utilized geomechanical models. Accordingly, first of all, we selected a suitable mathematical model, based on the Monte Carlo technique. Then, we integrated it into a commonly used FDM (Finite Difference Method) commercial code (FLAC2D). Finally, we studied the stability of an actual slope, considered as a test case (Lettomanoppello, Chieti, Italy), whose geological and geomechanical characteristics make it suitable for the application of the selected code. In this step, in order to simplify the analyses, we considered only dry conditions. Therefore, we discussed how the stochastic distributions of the local stress states, resulting from hundreds of repeated runs, were correlated to the mechanical parameters distributions obtained by the Monte Carlo application and assigned to each points as inputs. The results of the test application suggest that our approach can be used to identify sectors of the slopes more sensitive to the variability of the input values of the main geotechnical parameters, which would require a more accurate modelling and monitoring. © 2018 Elsevier B.V.