Towards Deterministic Fragmentation Models for Surface Mines

Fragmentation prediction is important in surface mining to optimize downstream processes such as loading, hauling, comminution, and beneficiation processes. Traditional approaches, such as the Kuz-Ram model, image-based fragmentation analysis, and numerical simulations, rely heavily on large datasets acquired after blasting. Moreover, due to the inherent limitations in each method there is an underestimation of fine and coarse fragment sizes, when applying these models without prior knowledge regarding the fragmentation distribution. These drawbacks are restricting their use only in post-blast parameter prediction. In response to these inefficiencies, this study presents a deterministic fragmentation prediction methodology based on a single-blasthole model. The model simulates the formation of the damage zones around a blast hole based on the number of cracks within a distance from the centre of the blast-hole and correlates them with an estimated fragment size in each domain. A fragmentation distribution from the literature is employed to validate the proposed approach and the results are compared with those generated using a modified Kuznetsov model. This approach aims to enhance blast design decisions by enabling pre-blast fragmentation prediction. Based on this, blast designs could be estimated depending on the desired fragmentation before blasting, giving the opportunity for a tailored to local rock properties optimization.