Towards the implementation of a universal control chart and estimation of its average run length using a spreadsheet: An artificial neural network is employed to model the parameters in a special case

A control chart procedure has previously been proposed for which the Shewhart &Xmacr;-chart, the cumulative sum chart, and the exponentially weighted moving average chart are special cases. The rapid and easy production of these charts, plus many others, is proposed using spreadsheets. In addition, for all these novel charts, the average run lengths are generated as a guide to their likely behaviour. The cumulative sum chart is widely employed in quality control and is considered in greater detail. Charts are designed to exhibit acceptable average run lengths both when the process is in and out of control. A functional technique for parameter selection for such a chart is introduced that results in target average run lengths. It employs the method of artificial neural networks to derive apropriate coefficients. This approach may be extended to any of the charts previously introduced.