Levkovitch, V, Flatten, A, Bushkov, V, Selivanov, D & Beck, D 2023, 'Using observations and measurements to establish a relation between modelled movement and Probability of Failure in a large open pit', in PM Dight (ed.), SSIM 2023: Third International Slope Stability in Mining Conference, Australian Centre for Geomechanics, Perth, pp. 707-720, https://doi.org/10.36487/ACG_repo/2335_48
(https://papers.acg.uwa.edu.au/p/2335_48_Beck/)
Abstract: In late 2021, the potential to extend a very large open pit was investigated. The mine has a high-resolution structural and hydrogeological model, permitting construction of a large-strain, discontinuum, closely coupled hydromechanical finite element model with a much better than inter-ramp scale representative elementary volume. In models of this type, the hydrogeological parameters and constitutive model are also non-linear, so that conductivity and the Biot coefficient are expressed as functions of the modelled plastic strain tensor.
Additionally, because extreme seasonal variations occur in surface water flows, the models included simulation of the flow of near-surface water and a novel method for capturing the inflow of this water into the dilated rock mass, joints and faults, thus permitting estimation of active water pressure. Active water pressure in the damaged and dilated rock mass and rock mass defects is a major driver of instability in slopes, but it is not often simulated, especially in models of this scale (30 pit stages, over 100M degrees of freedom).
The addition of this mechanism for water-driven instability and the high-resolution model setup permitted very close calibration of modelled and measured movements, enabled measurement of hydraulic head and failures using many years of data, and facilitated a statistical analysis of the correlation of failure and modelled movement and strain. This was undertaken using a cell evaluation method (CEM; Beck & Brady 2001), resulting in a true probability function using model outputs with an estimate of error and resolution.
The CEM has previously been used to compute the relation between seismic event probability and stress and energy criteria for underground mines, but this is the first known use of its application for an open pit.
The project resulted in a comprehensive validation of the model for the mine, a useable probabilistic criterion for slope failures ranging from small to medium, and a hypothesis ranging from the extrapolation of early slope stability data to the prediction of larger failures. There are also lessons regarding the necessary resolution of hydromechanical assumptions and inputs for high-quality prediction of slope instability, particularly the need to capture the influence of rock mass and rock defect dilation and degradation on flow and pressure.

Keywords: feasibility study, large open pit slope, hydromechanical coupling, non-linear model, open pit stage, Probability of Failure, active pressure, Factor of Safety, slope failure, pore pressure, near-surface hydrology, LR4, freeze thaw, melt water