No abstract

Several control charts for individual observations are compared. Traditional ones are the well-known Shewhart individuals control charts based on moving ranges. Alternative ones are non-parametric control charts based on empirical quantiles, on kernel estimators, and on extreme-value theory. Their in-control and out-of-control performance are studied by simulation combined with computation. It turns out that the alternative control charts are not only quite robust against deviations from normality but also perform reasonably well under normality of the observations. The performance of the Empirical Quantile control chart is excellent for all distributions considered, if the Phase I sample is sufficiently large.

In innovative fast product development processes, such as consumer electronics, it is necessary to check as quickly as possible, using field data, whether the product reliability is at the right level. In consumer electronics, some major companies use the Warranty Call Rate (WCR) for this purpose. This paper discusses extensively the theoretical and practical drawbacks of the WCR. Subsequently, it is demonstrated, using a Weibull failure distribution, that only a few months after product launch, say three months, the warranty data offer the opportunity to estimate the parameters of the failure distribution. Of course, this requires that the warranty data are available in the quality department. Unfortunately, for some companies the field feedback information process from the repair centres to the quality department causes a delay of several months. These companies have to speed up their field feedback information process before they can fully take advantage of the proposed estimation procedure.

SummaryOne of multiple testing problems in drug finding experiments is the comparison of several treatments with one control. In this paper we discuss a particular situation of such an experiment, i.e., a microarray setting, where the many-to-one comparisons need to be addressed for thousands of genes simultaneously. For a gene-specific analysis, Dunnett's single step procedure is considered within gene tests, while the FDR controlling procedures such as Significance Analysis of Microarrays (SAM) and Benjamini and Hochberg (BH) False Discovery Rate (FDR) adjustment are applied to control the error rate across genes. The method is applied to a microarray experiment with four treatment groups (three microarrays in each group) and 16,998 genes. Simulation studies are conducted to investigate the performance of the SAM method and the BH-FDR procedure with regard to controlling the FDR, and to investigate the effect of small-variance genes on the FDR in the SAM procedure.

No abstract