Simulations of Deposits

The regionalized variable $z(/boma{x})$, which is the subject of geostatistical structural analysis, is interpreted as one possible realization of the random function $Z(/boma{x})$. It is possible to generate other realizations $z_s(/boma{x})$ of this random function $Z(/boma{x})$ by the process of simulation. This simulated realization $z_s(/boma{x})$, will differ from point to point from the real realization $z(/boma{x})$, but, statistically, will show the same structure of variability (same histograms and variograms); moreover, at the experimental data locations $/boma{x}_i$, the simulated values will be equal to the true values, i.e., $z_s(/boma{x}_i) = z(/boma{x}_i), /forall /boma{x}_i /in$ data set. The simulated and actual realization will, thus, have the same concentrations of rich and poor values at the same locations. The two realizations can be viewed as two possible variants of the same genetic phenomenon characterized by the random function $Z(/boma{x})$. The simulated realization, however, has the advantage of being known at all points and not only at the data points $/boma{x}_i$.

Various methods of extraction, stoping, blending, stockpiling, etc., can then be simulated on this numerical model $/{z_s(/boma{x}), /boma{x} /in D/}$ of the deposit or, zone $D$. Using feedback, the effect of these simulated processes on recovery, fluctuations of ore characteristics, and on the running of the mine and mill in general can be used to modify certain parameters of the mining project.

The techniques for the simulation of a regionalization $z(/boma{x})$ can be extended to simulations of coregionalizations, i.e., to the simulations of the simultaneous spatial distributions of the various characteristic variables (thicknesses, grades, density) of a mineralization.