The influence of conditional bias in optimum ultimate pit planning

Kriging estimates of geological resources are sometimes conditionally biased because the kriging plan contains too few samples. This may be done purposely in order to match the variance of the estimates to the true block dispersion variance. This work aims at evaluating the influence of conditional bias over the ultimate optimum pit design in relation to its shape, size and NPV (Net Present Value of profits). It has been restricted to two real case studies: a ‘carrot shaped’ gold porphyry in northern Greece and a manto-typeexotic copper deposit located in northern Chile. The methodology applied to each deposit consisted of generating a ‘real block model’ and four resource estimates by the use of conditional simulation and the application of four different kriging plans to the drillhole databases respectively. The estimation models vary in smoothness and conditional bias. Ultimate optimum pits were determined for the ‘real’ and the four different kriging models. Pit comparisons lead to the following conclusions: •In the worst case, conditional bias overestimated the project's NPV by 32% and 5% for the gold and copper deposits respectively. •The overestimation of high grades is more relevant than the underestimation of low grades. This resulted in a tonnage over-extraction of 148% and 1% for the gold and copper deposit respectively. This difference was attributable to the vastly different geometry of the deposits. •The smoothing effect of kriging, without or very little conditional bias, produced open pits that were different to the ‘true’ (ideal) ones and furthermore, underestimated the project's NPV by 10% and 6% for the gold and the copper deposits respectively. •Other factors that influence the optimum open pit design and to some extent control the effect of conditional bias are: the cut-off grade, the orebody geometry and distribution of grades within it and the amount of over-burden. Keywords: conditional bias, ultimate pit design