|کد مقاله||کد نشریه||سال انتشار||مقاله انگلیسی||ترجمه فارسی||نسخه تمام متن|
|506897||865062||2016||7 صفحه PDF||سفارش دهید||دانلود کنید|
• A data fusion method is proposed as a tool for the exploration phase of mining.
• Raman analysis is used to extract mineralogical data from drillcore samples.
• Data Envelopment analysis (DEA) is carefully introduced to analyse the data.
• The results of DEA modelling are compared with Principal Component Analysis.
• The results of DEA modelling are compared to laboratory analysis for validation.
The exploration of mining has often been limited by time-consuming methods of analysis. This paper introduces Data Envelopment Analysis (DEA) as a new tool for the exploration phase of mining. DEA is a non-parametric method for data fusion, and it is used alongside with the on-site Raman analysis. Ten meters of halved rock drillcore from the Kittil mine (Suurikuusikko deposit) were pulverised and homogenised, thus ensuring that each meter had a representative sample. These 10 samples, one for each meter, were subsequently measured with a grid measurement (32×32 measurement each) using the Raman setup. All the data points were analysed using the point-count method. After identifying the frequency at which potentially valuable minerals appear in the samples, this information was analysed using DEA. The study ends by presenting an efficiency score for each meter of drillcore. These efficiency scores enable geologists to judge more rapidly which parts of the drillcore must be logged more carefully. In addition, Principal Component Analysis (PCA) is discussed as an alternative for producing similar results to DEA.
Journal: Computers & Geosciences - Volume 93, August 2016, Pages 96–102