Modelling the risk of progressive slope failure: A new approach

Abstract

Abstract This paper is concerned with the development of a new, simplified model of progressive slope failure within a probabilistic framework. Moreover, the change in the probability of sliding or overall failure based on ‘conditional’ or observed events can be simulated. Such a simulation can be useful for updating the risk of sliding atvarious stages during the life of a slope. The model requires the definition of an appropriate safety margin representing the event of progressive failure up to a certain position along the slip surface. The probabilities of such events may be calculated if the shear strength parameters are represented by appropriate random fields. In contrast to a previously established progressive failure model, the calculation of joint or bivariate probabilities associated with pairs of events is not required. Therefore, the new model is very convenient for performing the calculations of risk of progressive failure. It works well when compared to the more rigorous bivariate model which has previously been published. The results for an example problem concerning the short-term stability of a slope in saturated clay are compared to corresponding results from the previous, more rigorous model.