Stope optimisation uses geotechnical limits, and economic constraints, cut-off grade, head grade, waste inclusion settings, and geometry limits of the orebody to find practical mining shapes directly from the block model. However, this process uses a limited number of deterministic options for the basis of strategic decisions-making. In this webinar we explore a process that incorporates many optimisation frameworks and leverages AI-Assisted workflows to move towards a probabilistic assessment of stope arrangements, enhancing the efficiency of the evaluation process.

Specifically, we will cover.

1) Introduction and methodology
2) Accelerating the creation of stope arrangements
3) Leveraging AI to identify a ‘risk aware’ operating lane
4) Results and key learnings
5) Questions and Answers

This case study uses de-identified data from a metalliferous underground mine. The approach compares stope optimisation outcomes across a range of cut-off grades to identify the most stable operating range and margin curve. Monte Carlo percentile bands are then used to assess uncertainty, with the lane midpoint representing the most stable part of the margin curve.