Practical issues in the construction of control charts in mining applications

In recent years the importance of quality has become increasingly apparent due to global competition. According to Banker et al. (1998), quality is the most important strategic issue of top management. It is primarily concerned with consumer’s view of value desired and value received. In general, the term quality addresses two aspects: quality of design and quality of conformance. In the mineral industry, quality control procedures are primarily concerned with the quality of conformance which aims at meeting consumer desired specification limits. The problem of quality control is of decisive importance in the mining industry due to inhomogeneous formation of ore deposits. Particularly in the cases of ore deposits which are formed under complex geological phenomena, the process of mineralization largely affects the grade distribution. This leads to formation of deposits which are usually characterized by wide variation of ore types and grades. The task of a quality control practitioner is to homogenize the different ore qualities during extraction in order to satisfy consumer’s quality requirements. To facilitate their operations, the use of statistical quality control techniques including control charts has been proved to be a major contributing factor in achieving required quality and improving productivity. Through the application of control charts, various assignable causes of quality variation are identified. The Shewhart X — and R charts are still the most popular control charts in use today for continuous monitoring of quality variation, due to their simplicity and effectiveness. The main purpose of the application of these charts is to identify the root causes of quality variation based on which corrective actions are taken to remove irregular grade fluctuation. In order for the Shewhart X — and R charts to function properly, the underlying assumptions are that the sample observations are assumed to be normally distributed and statistically independent. Even though these assumptions are important but their existence is questionable in mining applications. A theoretical justification for normality is based on the Central Limit Theorem. According to the Central Limit Theorem, the distribution of the sample averages of n independent observations will approach normality as the number of observations in a sample increases, even if the individual observations are not normally distributed. In this context, Shewhart observed that many individual observations are nonnormal, although the distribution of sample means of size four will in many cases Practical issues in the construction of control charts in mining applications