Among the numerous practical application of Statistical Process Control (SPC) is the monitoring of the number of defecting (non-conforming) items that are produced from a manufacturing company or the number of the incidents of a disease in a specic area. The basic aim is to detect any kind of change (e.g., an increase in the number of non-conforming items) quickly and accurately. In such cases, the available data are usually discrete (counts) and for process monitoring, the ordinary control charts for attributes (e.g., np-, p-, c- and u-charts) are used in practice. The main assumption is that the data come from the Poisson or the binomial distribution.
Due to technological advancements and automation progress, many processes are now characterized by a low fraction of non-conforming items. That kind of processes are known as high-yield processes and it is very common to have an excessive number of samples with zero non-conforming items. Consequently, due to the inherent over-dispersion of data, the ordinary schemes for attributes cannot be eactively used because of the high false alarm rates and the low statistical power in the detection of changes in the parameters of the process.
In this talk, new control charts, which are suitable for the monitoring of high-yield processes, are proposed and studied. Instead of the ordinary Poisson and binomial distributions, we assume that a proper parametric model for the process is the zero-inaged Poisson (ZIP) or the zero-inated Binomial (ZIB) distribution. We provide the Markov chain methodology for the theoretical study of each chart as well as aspects of their statistical design. Also, numerical comparisons between the dievent control charting techniques are given. Finally, the practical application of the proposed techniques is discussed.
(joint work with Prof. Philippe Castagliola and Prof. Petros Maravelakis)