Professional Articles

Cement Handling

Modification of Control Charts for Limestone Quality Under Autocorrelated Condition

Traditional statistical process control charts used for the quality control of minerals assume that observations are independent and normally distributed about some mean value. The validity of the interpretation of out of control condition largely depends on these two assumptions. An assumption of independence of quality related data in mining operations is questionable, as autocorrelation amongst the observations becomes an inherent characteristic in mineral deposits where ore grades are spatially distributed with high interdependence.
This concept led to an examination of another type of control chart, namely the modified Shewhart chart, to capture the autocorrelation among observations while constructing control charts. This paper modifies and extends the existing standard methodology by utilizing the time series analysis approach and by introducing dependence via a third order auto-regressive process (AR (3) Model).
Limestone is the principal ingredient for the production of cement and since there are both large and small suppliers of Limestone to various plants of fluctuating demand, the quality considerations are paramount. And till date, there is no other means of statistical quality control of Limestone for supply to Cement plants except the Shewhart chart and the modified Shewhart chart [2].
The values of the quality characteristic are plotted along the vertical axis, and the horizontal axis represents the samples or subgroups (in order of time), from which the quality characteristic is found [8]. Finally, the ultimate goal of statistical process control is the elimination of variability in the process. It may not be possible to completely eliminate variability in the process, but the control chart is accepted to be an effective tool in reducing variability as much as possible.
However, the conventional Shewhart chart is plotted based on the assumption that the quality characteristics follow the normal distributions and that they are statistically independent.