Evaluation of the reserve of a granite deposit by fuzzy kriging

This study describes a method for estimating the reserves in an ornamental granite deposit. The method is adapted to the existing optimised mining method and provides a more realistic estimate of commercial quality than conventional methods. Using a batholith geology-mining map, a description of fracturing in the mining area, and underground information provided by continuous-core boreholes, an estimate of reserves was made that featured the following innovations: 1) the use of fuzzy set theory to assign qualities to each block, and 2) the inclusion of fracture anisotropy at the deposit level in the fuzzy mathematical formulation. These innovations enabled quality distributions in the rock mass to be assessed more realistic, by reducing the estimation error caused by assigning a single quality to the primary block and by improving the estimate of the distribution of commercial qualities defined by fracturing. This method produced a satisfactory estimate of the commercial qualities of one of the world's most important granite deposits (located in northwest Spain). The results have been stored in a database–linked to a graphic representation of the blocks (similar to a geographical information system)–which, by providing automatic access to information in regard to the location of the best quality commercial blocks, ultimately optimises mining planning.