Cumming-Potvin, D, Wesseloo, J, Jacobsz, SW & Kearsley, E 2018, 'A re-evaluation of the conceptual model of caving mechanics', in Y Potvin & J Jakubec (eds), Caving 2018: Proceedings of the Fourth International Symposium on Block and Sublevel Caving, Australian Centre for Geomechanics, Perth, pp. 179-190, https://doi.org/10.36487/ACG_rep/1815_11_Cumming-Potvin (https://papers.acg.uwa.edu.au/p/1815_11_Cumming-Potvin/) Abstract: The Duplancic conceptual model is the industry accepted model of caving and is the framework within which most results from numerical modelling and cave monitoring are interpreted. The Duplancic conceptual model implies that the damage ahead of the cave back decreases continuously with increasing distance from the cave surface. Evidence from a variety of sources indicates that this may not always be the case and that a discontinuous damage profile may be present. Cumming-Potvin et al. (2016b) describes a physical modelling program which was undertaken to investigate the fracturing and propagation of the cave. The results of these centrifuge tests showed that caving could occur via a series of fractures oriented parallel to the cave surface and that the cave back progressed vertically via ‘jumps’ to the next successive parallel fracture. In Cumming-Potvin et al. (2016a), this caving mechanism was termed ‘fracture banding’. Multiple examples of a similar mechanism of failure were observed in literature. In addition, the patterns in microseismic event location indicate that fracture banding could be occurring in currently operating caving mines. This paper examines evidence from a number of sources in the field of caving mechanics and presents an extended conceptual model of caving. The new model is able to account for the mechanism of fracture banding, along with the continuous style of failure from the Duplancic conceptual model. There are still many unknowns about the fracture banding mechanism and propagation of caves. These include the specific conditions under which the caving mechanism changes and whether the mechanisms lie on a continuum, or if there is a sharp, sudden change. Two conceptual models are presented: one which includes only that which is known about the mechanisms of cave propagation and one which speculates upon the factors involved and the underlying origins of the fractures. Keywords: fracture banding, cave mining, physical modelling, cave monitoring, seismicity, centrifuge