Correcting for dilution of rakes processed sequentially from multiple mines at BHP Billiton Iron Ore Newman Joint Venture operations

Abstract

At BHP Billiton Iron Ore operations, run of mine ore from the several Newman Joint Venture (NJV) mines is railed in 12 kt rakes to Port Hedland. Rakes are processed sequentially, with the lump and fines components from each rake sampled after crushing and screening. The grade of ore from the individual mines varies around consistently different means. Although a discrete sample is attributed to each rake, the sample contains some material from the following and/or the preceding rakes due to dilution by ore from the neighbouring rakes during the dumping, screening and crushing processes. Individual rake grades at the port are therefore biased according to the mine sources of adjacently processed rakes. Dilution does not affect the grade of the overall NJV product or the accuracy of port stockpile grades, however it does limit the ability to use port rake grades to review the quality performance of individual mines and provide meaningful lump–fines algorithms and port–mine biases for individual mines, information that is useful when scheduling ore for mining. A non-linear regression model has been used to estimate the average dilution from the preceding and following rakes. It has been shown that the lump and fines samples for 12 kt rakes are each diluted within the Port process with about 8˙3% of the sample coming from the preceding rake and ∼3˙6% from the following rake. This can, bias the reported rake grade by up to 1% iron. The results of this analysis have been incorporated into the daily grade scheduling system to provide more accurate individual mine grade data and consequently better decision making.