The concept of robustness of statistical procedures is one of the most important subject in Zielinski's papers. In this article the development of the idea of robustness as introduced in Zielinski's papers is presented. The definitions of a supermodel and a robustness function are given. The problem of the robust estimation of a scale parameter in an exponential model and the robustness of tests for comparison of means in two or more populations are described. Robustness in Bayesian statistical models is connected with an unexactly specified prior distribution. Here the following Zielinski's results in Bayesian robustness are presented: the most stable estimator in the Poisson model and the Bayes optimal stopping rule in a homogeneous Poisson process with conjugate classes of priors, the optimal experimental designs in Bayesian linear models under variation in the prior, an upper bound for the Kolmogorov distance between the posterior distributions in terms of that between the prior distributions.
Wprowadzenie zmian mające na celu poprawę jakości technologicznej wyrobów należy rozpocząć od analizy sytuacji zastanej. Analiza ta powinna obejmować wszystkie elementy procesu technologicznego mające wpływ na jakość wyrobu. W poniższym opracowaniu zaproponowano wprowadzenie liczbowego wskaźnika SWJ do ilości oceny jakości technologicznej oraz wskaźnika WPJ do oceny poprawy jakości. Dane do analizy pochodzą z banda przemysłowych.
EN
The paper presents the problem of technological quality of bearings analyzed by means of FMEA and Taguchi method. The work attempts to find an answer to a question how the information put in FMA sheet may be used to assess. Quality. The way this data may provide insight into quality level of production process has been analyzed with focus on identifying recommended change in the future. To describe a quality level two coefficients have been created: SWJ (average value of quality) and WPJ (coefficient of quality improvement). To identify direction of changes Pareto analysis has been applied. At the and of the paper Taguchi method has been referred to. The method may be used to improve quality through selecting optimal process parameters.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.