Band 16, Heft 2

In this paper we provide a method of finding exact designs for single sample acceptance sampling plans. The method uses an algorithm that allows determination of plans based on the hypergeometric, binomial or Poisson distributions. Designs can be found for specified OC curves of either Type A or Type B.

This article presents the method how to design a R control chart by variables, assuming that the quality characteristic that is being controlled follows a Γ-distribution. The use of an R -control chart shall be complementary to the application of a control chart for the process average. The problem is to find a coefficient that allows the estimation of the process standard deviation by means of the sample range and determines the values of the control chart limits. The sample size and the sampling interval, on the other hand, may be taken from the design of the complementary chart. (González, Viles, Economic Quality Control 15: 109-118, 2000)

In this paper, we extend Johnson's work [Johnson, T. Journal of Quality Technology 24: 211-215, 1992] of formulating the relationship between process capability indices (PCI) and expected squared error losses. Moreover, a comparison is made between the expected squared error loss of PCI under the specified capability index value.

Process control by means of a p -chart or np -chart becomes almost impossible, if the quality level of the process in question is very good. However, if the quality characteristic is measurable then by narrowing the given tolerance limits and using accordingly changed gauges one can easily control the process by means of an attribute control chart. In this paper a method is developed to determine an optimal (in a certain sense) way to select the new narrowed tolerance tolerance limits if the distribution of the quality characteristic can be sufficiently well approximated by a normal distribution. This paper is the outcome of engineering research aiming at obtaining a useful result which can be applied at shop-floor level. A majority of modern methods (including ISO 9000 ideology) try to prevent defects by searching for defects. In contrast, the idea here is to increase sensibility of the method by defining a “high virtual defective level”, which is used for defect prevention without actually searching for defects.

PISA stands for “Programme for International Student Assessment” performed, by the OECD and consisting of a cyclic evaluation of the new generations' basic competences. The first evaluation took place in 2000 and the results were published on 4 December 2001. According to the official statements an overall of 32 countries participated in this first study. The publication of the results in Germany caused some great excitement, as Germany the “country of arts and science” was ranked at the lower edge not so far from Brazil. Politicians, teachers, scientists and any other people expressed their opinions, looked for those responsible, for the bad state and demanded immediate actions for improving the education in Germany. Bavaria, the German state which pretends to have the best educational system, at least in Germany, claimed that the Turkish and Jugoslavian students in Germany had caused the bad grade. Other being more cautious demanded better teachers, another educational system and better pupils. The nationwide discussion arouse my curiosity and thus, I tried to get some details on PISA and found them e.g. on the homepage of the OECD, the renowned Max-Planck-Institut Berlin (http://www.mpib-berlin.mpg.de/pisa) and other national institutions in the participating countries. The study has an ambitious aim and is essentially a statistical one. Therefore, a thorough evaluation of the quality of the study and its results should stand at the beginning of any discussion. This paper looks at PISA exclusively from a statistical point of view and makes an attempt to evaluate its quality. It is an attempt, because only very limited information on the applied statistical procedures were available despite of the fact that hundred of pages describing the study are made available through internet.

In this paper we present a mathematical model of a generic manufacturing system. The quality of the manufactured product is captured through one or more of its critical design/process parameters. Managers often face the dilemma of which product/process is to be singled out for improvement in quality with the limited capital outlay on hand. In this paper, we use the critical process parameters along with the standard production variables in a mathematical programming framework, to identify the process to be targeted for improvement. Consistent high quality, results in higher rewards in a perfectly competitive market place and also requires higher amounts of the employed resources, including capital. The optimal investment to be made in achieving higher rewards depends upon the product characteristics. We consider alternative systems, which differ in their costs of quality, since the underlying process parameters follow different probability distributions. Our experiments provide some important insight into the optimal investments in quality, and the accompanying qualityproductivity trade-offs. We demonstrate that 100% process conformance or 100% use of the productive resources do not result in maximum net profits. Our model also reinforces the notion that consistent high quality ultimately translates into a corresponding gain in productivity and higher profits or net revenues.