Sampling Theory, Sampling Practices, and Their Economic Impact

Poor sampling, compounded by poor laboratory subsampling, leads to questionable geostatistics, and generates severe conciliation problems between the geological model, the mine, and the plant estimates. These problems also affect the price of commodities and the validity of environmental assessments. The result is a huge money loss for the company involved, evolving later in likely litigation. It is of key importance for geologists, miners, metallurgists, chemists, and environmental specialists to extract maximum information from the available data, as large investments and crucial decisions depend on it. False evaluations lead to devastating scenarios such as:

Abandonment of viable properties,
Exploitation of unprofitable properties,
Mismanagement of viable properties, and
Incompetence in fraud detection.

It is critically important to quantify the heterogeneity of important constituents in any new property. Failure to do appropriate testing leads to invalid sampling and subsampling protocols, excess drilling, and a biased database that would later lead to false geostatistics. The following sequence is part of inescapable practice:

How is the constituent of interest distributed in the material to be sampled?
Conduct Heterogeneity Tests to quantify the sampling characteristics of the constituent of interest.
Optimize sampling protocols and the way they are implemented, according to the results from the Heterogeneity Test.
Implement protocols using valid sampling equipment: 75% of the sampling equipment available on the market will never do the job.
Implement a comprehensive, systematic quality control program to monitor sampling precision and accuracy.

The staggering cost of data irrelevant variability is not easy to detect, quantify, or correct. A strategy for effective management of variability will enable managers to identify and minimize annoying conciliation problems between theoretical models and reality: Your decisions are only as good as your samples!

The course offers simple ways to quantify money losses for a given sampling precision, and it provides a good strategy to prevent catastrophic sampling inaccuracy for which there is no statistical cure. Unless sampling precision and accuracy are clearly connected to economic issues, it is unlikely that any manager would understand the reason for improving sampling protocols and the way they are implemented. At the end of the course, the attendee will be better equipped to present the economic advantages of good sampling to company executives. Therefore, the course is pre-requisite for bank investment: Bankers must listen, and trust the Sampling Theory.