Its NPV is only 17% lower than the best possible NPV for each scenario and 59% higher than the deterministic plan. This approach minimizes the risk of sending an extracted material to the wrong destinations while producing optimal production schedules. The second stage decisions re-evaluate block destinations based on a set of realization. In the first stage, extraction periods are obtained using an e-type model and remain fixed for all realizations. Next, a two-stage stochastic production scheduling approach is used as a replacement for deterministic mine planning. Unlike conventional cosimulation, the integration of inverse transform sampling helps to reproduce an inequality constraint in the bivariate relationship and increases the correlation coefficient. Overall, the proposed algorithm demonstrates simulations of similar quality as the conventional approach, which can be observed through histogram and variogram validations. Results are assessed based on validation of marginal distribution, spatial correlation and bivariate relationship. The proposed algorithm is applied to a real case study from an Iron deposit with a sharp inequality constraint between Iron and Silica. This study proposes an algorithm based on a hierarchical cosimulation framework integrated with inverse transform sampling, which is designed to model variables within thresholds derived from a linear inequation. Therefore, this work's motivation is to propose a way to model datasets containing inequality constraints between variables and generate production schedules considering geological uncertainty. The poor reproduction of this feature can lead to an overestimation of the secondary variable, affecting the processing cost. For example, it is common to deal with challenging datasets containing bivariate complexities among variables, such as inequality constraints. Nevertheless, there are still many limitations in commercially used stochastic geostatistical algorithms, particularly multivariate methods. Integration of stochastic geostatistical realizations into mine planning can help to minimize the risk of not meeting the production targets. However, traditional mine planning does not allow the risk management associated with geological uncertainty due to using a single mineral resource model as an input. In that regard, geostatistics and mine planning as disciplines are critical parts of the mining business. The modern mining industry employs plenty of exploration data digitization and utilizes computational resources for future forecasting and production scheduling.

Download the new Diamond Cut 11.00.0