Computer simulation of grade distributions in iron ore deposits: a tool for grade control planning

"Quality of iron ore is generally determined by several factors which can be correlated or independent. Among these f actors, we generally find per cent iron, silica, magnetite or hardness. Forecasting the short- and long-term fluctuations of these f actors in the future pit is an important feature of mine planning in these deposits. Computer simulation provides a realistic answer to these questions. It allows to generate values for the most important quality features of each shovel load with the proper correlations between them as well as between one load and the next being fully reproduced. Then different sampling schemes f or grade control or blending facilities can be examined on the simulated data. The mechanics of the simulation procedure and its potential applications are described in an example.IntroductionThe purpose of this paper is to illustrate the use of in-situ grade simulation in the planning o f the grade control in an iron ore min e. More precisely, given a production unit of given size, we need to know how many samples have to be collected in that unit so that we ca n recognize its grade with a reasonable precision. Conclusions that we would like to arrive at are of the type: with 4 blasthole samples we ca n get the iron grade of a 10,000 ton bloc k with less than a ± 2% Fe error 95% of the time.The approach t hat we take to solve this problem is simulation. We simulate the grade of samples and the grade of the block that contains these samples. We repeat the simulation for hundred s o f blocks. By comparing the sample grade with the block grade in each case, we determine the actual number o f blocks where the difference sample-bloc k is less than a given limit."