Goulet, A, Heinsen Egan, M & Wesseloo, J 2026, 'Advances in probabilistic support design using discrete fracture networks', in D Cumming-Potvin & J Wesseloo (eds), Ground Support 2026: Proceedings of the 11th International Conference on Ground Support in Mining, Australian Centre for Geomechanics, Perth, pp. 1-19, https://doi.org/10.36487/ACG_repo/2625_25
(https://papers.acg.uwa.edu.au/p/2625_25_Wesseloo/)
Abstract: Ground support design in underground excavations is fundamentally influenced by the anisotropic and discontinuous nature of rock masses. This paper presents some recent advances in probabilistic ground support design using discrete fracture network (DFN) modelling. The methodology integrates calibrated DFNs, block identification, stability assessment, and probabilistic risk evaluation to quantify both the likelihood of block occurrence and the probability of block failure. The paper outlines a practical workflow for generating representative DFN models from structural mapping and borehole data, including calibration procedures for fracture orientation, size, and intensity. Thousands of DFN realisations are used to identify tetrahedral blocks around realistic excavation geometries and assess their stability under varying reinforcement and surface support configurations. The approach explicitly captures the influence of rock mass fabric on block formation, support demand, and failure probability, providing a more realistic basis for support design than traditional deterministic methods. The results demonstrate how DFN-based analyses can be used to evaluate support performance probabilistically, compare alternative support standards, and estimate the frequency and magnitude of potentially unstable blocks. The methodology provides a practical framework for moving from deterministic support design towards risk-informed and probabilistic decision-making in underground mining.

Keywords: discrete fracture network, probabilistic design, ground support, block stability, rock mass fabric, risk assessment