Metodologias de sequenciamento de lavra com a utilização de multivariáveis do modelo de blocos simulado para avaliação de riscos e incertezas

Abstract

The currently advanced mining sequencing work is based on a conventional model, where the geological block model is based on a single sample of iron grade and its contaminants, resulting in a deterministic product of the sequencing of the mining fronts. In recent years the Geostatistical Simulation has been presented as the main tool to produce multiple equiprobable models of the same mineral deposit, which can be studied and investigated as a panel of possibilities, where the variability and uncertainty of the simulations can provide another tool for the company's decision making. The present study has the main objective of using the simulated model of the Sapo Mine, containing for each block 50 simulations of the Fe2O3, Al2O3 and P content variables. This model will be used in mine sequencing works, presenting new possibilities and methodologies for the evaluation of mining plans for the short-term and long-term horizons. In short, the strategy outlined for the elaboration of this new methodology \ dynamic sequencing work will be composed of three steps: I- Perform a conventional sequencing using the content estimative originally calculated by ordinary kriging, II- Calculate 50 sequences based on each of the simulations undertaken in the block model, obtaining 50 equiprobable sequencing scenarios and III- Direct the sequencing in the areas with greater geological knowledge, using a higher economic weight in the smaller variance blocks. The sequencing techniques applied showed complementary and more robust results than those of traditional methods. The incorporation of risk analysis and geological uncertainties in the construction process of mine sequencing is possible and is now indispensable in the strategic assessment of mine planning projects. With the accomplishment of step II it was possible to prove that all simulations were successful in maintaining the final production target of Pellet Feed - 23.5Mt year, as well as to identify that the months of October, September and December presented moderate risk in relation to the mine total movement maximum capacity, since about 30% of the sequencing exceeded their stipulated limit for these periods. The adoption of the Economic Weight for Variance in mining sequencing work enabled the planner to force mining sequencing in areas of greater geological knowledge in the early periods, thus giving the geology team more time to improve geological drilling surveys in the areas of greatest spacing of the survey meshes. One of the main results was the evidence of the change in behavior of Variance and Standard Deviation over time, it was possible to decrease both more sharply in the first half of the year, thus reducing the geological uncertainty of the period.