Συνδυασμένα διαγράμματα για τη μέση τιμή και τη διασπορά
Control charts are widely used in monitoring or surveillance of processes in a variety of industries. These graphical displays are designed to allow a practitioner to determine whether a process is in-control or out-of-control by taking samples at specified sampling intervals and plotting values of some statistics on a graphical interface which includes decision lines called control limits. The vast majority of control charts are designed to monitor a single process parameter, such as the mean or the variance, but it is often desirable to monitor the mean and the variance simultaneously, since both may shift at the same time and since a change in the variance can affect the control limits of the mean chart. In this thesis we present an overview of control charts for joint monitoring of mean and variance. More specifically, in Chapter 1 we briefly present the general theory of control charts. We focus mainly on Shewhart, CUSUM and EWMA control charts for variables. In Chapter 2 we present elaborately the combined control charts and their performance. In these control charts a single statistic is plotted for monitoring both the process mean and standard deviation. In addition, we compute the average run length of the combined control charts. Finally, in chapter 3 we compare the control charts based on the Average run length (ARL) and we propose which control chart is most suitable for detecting an out of control signal for various shifts in process mean and/or standard deviation.