Integration of a neural ore grade estimation tool in a 3D resource modeling package

Ore grade estimation is a key aspect in the evaluation of a mineral deposit. In this paper an alternative approach to currently applied methods of ore grade estimation is presented. This alternative approach involves a modular neural network system integrated in a state of the art 3D resource modelling package. The need for a new method of ore grade estimation comes from the difficulties in applying conventional methods such as geostatistics. These methods require a lot of assumptions, knowledge, skills and time to be effectively applied while their results are not always easy to justify. The aim of the proposed system, called GEMNet II is to provide fast and reliable ore grade estimation, with minimum assumptions and minimum requirements for modelling skills. GEMNet II has been tested on a number of real deposits. The results obtained so far have shown that it can provide with a very fast and robust alternative to the existing time-consuming methodologies for ore grade estimation.