Peattie, Richard John Reid (2007). THE USE OF SIMULATED FUTURE GRADE CONTROL DRILLING TO QUANTIFY UNCERTAINTY IN RECOVERABLE ORE RESERVES MPhil Thesis, School of Engineering, University of Queensland.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
n01front_peattie.pdf n01front_peattie.pdf application/pdf 173.66KB 8
n02content_peattie.pdf n02content_peattie.pdf application/pdf 4.15MB 9
Author Peattie, Richard John Reid
School, Centre or Institute School of Engineering
Institution University of Queensland
Publication date 2007
Thesis type MPhil Thesis
Supervisor Prof Roussos Dimitrakopoulos
Subjects 260100 Geology
Abstract/Summary Recoverable reserves are that portion of a mineral deposit that has been delineated by a mine plan to be economically minable. At time of prediction, recoverable reserves are estimated based on limited sampling information usually of a high quality into large panels. Subsequent to this, at the time of mining, additional information usually of a poorer quality is available allowing estimation into smaller blocks which are used in ultimate block selection. As recoverable reserve estimation is interested in the prediction of tonnages and grades that will be recovered at the time of mining, the main concern of recoverable reserve estimation is the imprecise selection of ore/waste as a result of the information and support effect. For the selection of a block as ore or waste during mining, a cut-off is usually considered where a block is considered as ore if its grade is greater than the cut-off or waste, if it falls below the cut-off. For perfect selection a cut-off needs to be applied to the real block grades however, as the real block grades are never known, more typically actual selection is employed where a cut-off is applied to the estimated block grades. In this case, the misclassification of blocks as ore or waste can take place as a result of information and support effect resulting in the incorrect selection of blocks. Block selection is at its best at this stage; prior to this during recoverable reserve estimation the potential for misclassification of blocks is far worse due to the reduced data densities existing with exploration drilling and difference in data quality between grade control and exploration sampling. Both of these factors impact on the information and support effect and therefore the ability to accurately select the correct blocks are ore or waste. In reconciling a mine plan it is important to make a distinction between deviations as a result of unpredictable dilution factors caused by mining and geotechnical factors and those factors such as the support and information effect that are predictable. We therefore assume free selection of the selective mining unit (SMU) where access and mining constraints are not taken into account. The widespread use of simulation for recoverable reserve estimation within the mining industry has been limited in the past due to limitations of computational speed and efficiency. This study demonstrates successfully a practical application for the simulating of recoverable reserves directly into SMU’s using a direct block simulation method to overcome these problems. In this thesis, the methodology is demonstrated on the Morila gold deposit, which consists of a geologically defined ore zone within a low grade host rock. Here, the direct block simulation methodology is demonstrated in a production environment and assessed in terms of whether the technique would find general acceptance in the mining industry. The direct block simulation results where contrasted in this thesis to a uniform conditioning estimate demonstrating firstly that the direct block simulation approach is as accurate and as efficient as existing recoverable reserve estimation techniques utilized in the industry. Secondly, that the direct block simulation is a practical alternative to traditional recoverable reserve methods. Finally, that direct block simulation added value by providing additional information not provided by traditional recoverable reserve techniques through a measure of the uncertainty in the recoverable reserves and the incorporation of the information effect. The calculation of recoverable reserves in this study, involved the calculation of an error density map which was added to the individual realisations of the direct block simulation to provide an uncertainty assessment of the recoverable reserves within the deposit, while accounting for the differences in support and information effect between the exploration and blasthole sampling information. The application of the method at Morila utilized known block selection information from mined out area to adjust the estimates to known statistical distributions. Finally, the realisations generated where shown to be comparable to the block selection determined in the mined out areas using the grade control drilling. Recognising the effect of information and support and quantifying and managing its potential impact when planning a mine operation can have a large influence in a project’s feasibility and final economic outcome. This thesis demonstrates a possible methodology for accounting for the expected inaccuracies in the selection of ore and waste block by incorporating the effect of future, as yet unavailable data, into the recoverable reserve estimates. In this way, the effect of support and information effect on the final estimates of recoverable reserves can be mitigated and fully informed mining decision can be made.

Citation counts: Google Scholar Search Google Scholar
Created: Fri, 21 Nov 2008, 14:54:54 EST