The main element of uncertainty to be evaluated by the simultaneous stochastic model is the geologic risk, represented by the variability of the grades in each material muck pile. Besides the uncertain nature of the metal content, the classification of the material (ore/waste) is also an element of uncertainty, to be evaluated in each scenario as a discriminatory feature for destination selection. The blocks are set to 40ft x 40ft x 40ft dimensions with a specific gravity of 2.89 (tonnage factor of 12.2 ft3/ton). The simulation was performed in @Risk 8.9 version PALISADE software, by applying for each block 3 simulations, for 10,000 iterations per simulation under specific distributions. This approach requires an understanding of the before and after blasting conditions, discerning between the grade distribution of each block when they are isolated, and the mixed statistics as the blocks are combined and selectivity is no longer possible. The representation of this logic is presented in Figure 61 for the results of muck pile #4, where the block statistics are differentiable on the overlayed grade distribution (on the left), solely considering the geologic aspects of each block. The right-hand-side plot represents the statistics of the same muck pile after the blasting, with all blocks combined and including operational parameters, such as the cutoff grade for material classification. The simulation also differentiates subdivisions in the grade distribution to represent the percentiles of the statistics, where P80 matches the selected cutoff grade to differentiate between ore and waste. From this observation, we can interpret that for 80% of the simulations the muck pile #4 will be labeled as waste, while 20% of the cases may actually be ore. This uncertainty in the correct classification of the material will drive special considerations on what is the correct destination for each muck pile and the economic impact of the decision. Figure 61 - (left) Grade distribution of each block before blasting. (right) Combined statistics after blasting. The output results from the simulation process generate a pool of percentiles {1%, 2.5%, 5%, 10%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 90%, 95%, 97.5%, 99%} that is reduced to key percentiles, henceforth described as scenarios (for the stochastic implementation), all of them assumed to be equally possible to occur (from best to worst case scenario). The mean values for each scenario are displayed in Table 5, where the cells containing
RkJQdWJsaXNoZXIy MTM0Mzk2