Application Of Correlation Analysis To Drilling Programs A Case Study

INTRODUCTION The goal of any development drilling program is the drilling of the exact number of holes necessary to adequately determine grade and dimensions of an orebody, no more and no less. This cost of development drilling represents a large part of the total development budget. Any technique which eliminates unnecessary drilling costs produces beneficial results, both to the company and to the person who can claim the savings. This high development cost factor has produced a quantity of research to develop techniques, for efficient control of drilling programs. Recent mining literature is full of articles describing methods for measuring and increasing the effectiveness of these expensive projects. The method described in this paper was developed during the latter part of a major development drilling project at the Christmas Mine of Inspiration Consolidated Copper Company in Gila County, Arizona. It was used to test the need for drilling individual holes but could be expanded to control an entire project. This technique was used at Christmas as a practical means for evaluating adequacy of drilling on an individual hole basis, and it was possible to eliminate several previously approved holes. Application of this method involves an evaluation of geological conditions present and a good bit of common sense. Projects involving large disseminated, rather homogeneous orebodies of the porphyry-copper type would be the most likely candidate for application of the technique. Discontinuous orebodies would not be amenable to this technique. The method is based on the measurement of the linear correlation between the spatial distribution of two sets of data. The Pearson product-moment correlation coefficient is used. This is a standard statistic commonly used to measure linear correlation between two sets of data. The statistical significance of the correlation coefficient at a 95% confidence level is the main criterion for controlling the project. For purposes of reference we will call our technique the "correlation method" in this paper.