Structural instabilities in open pit walls need to be assessed carefully for both safety and economic issues. Unstable wedges and daylighting blocks may be identified through three-dimensional (3D) image analysis, and determined in a first approximation by a discrete fracture network generation plugged into a polyhedral (rock block) model. To complete this preliminary assessment, a methodology for predicting failure inside the rock matrix is proposed by coupling discrete fracture network (DFN) modelling with the discrete element method (DEM). An identified wedge collapse that occurred in a coal mine is used to set up and to validate the model. In this case, the failure surface is shown to be well predicted by the progressive failure mechanism that develops from the toe of the structure inside the intact rock matrix. The DFN-DEM modelling is shown to be a complementary tool to assess rock slope stability in presence of non-persistent discontinuities.
Predictive models of unstable wedges and blocks daylighting on operational surface mine excavations are important tools for risk management in mining. Discrete fracture networks (DFN) are now commonly used for identifying individual rock blocks which may be unstable within a rock mass. The method using DFN combined with polyhedral modelling and kinematic analysis is thus useful in providing a first and rapid assessment of both the location and likelihood of these potential threats. However, the method does not account for rock mass failure and only blocks that are kinematically free can register as hazards .