Control Chart For Monitoring Tow-parameters Weibull Distribution

Statistical quality control system has been widely used in various industrial processes. In the tradition of quality inspection stage, quality assurance is to rely solely on the test or inspection to achieve. Therefore, quality assurance looks like afterward quality assurance and uneconomic quality assurance. Now, the research of statistical quality control focus on control chart technique, in which the Shewhart control chart, CUSUM control chart and EWMA control chart, is commonly used tools of monitoring the process mean or variance drift. The traditional Shewhart control chart strictly speaking only for normal distribution process.For non-normal processes must develop a new control technology. For a long time, people realize that, when the samples from the non-normal distribution, conventional control chart influences the monitoring capability significantly. The Weibull distribution is a kind of important continuous and non-normal distribution, a lot of skewness distribution can be expressed the Weibull distribution approximately. Now, the Weibull distribution is widely used in the product quality of life. It has theory value and practical significance of researching for the Weibull distribution process control chart method.In this paper, the first part outlines the research status at home and abroad and introduces the related theory. The second part is the main content of this paper, in this part we identify with the parameters related to statistics and then using computer simulation method to be validated by comparing. For the Weibull distribution shape parameters, known firstly, the diagram method of two-sided control charts and one-sided control charts is given, and then computer simulation is carried out and the previous research works are compared with. Comparison results show that the control chart we used either two-sided control charts or one-sided control charts, control chart performance was superior to the control chart proposed by Pascual. For the scale parameter of the control chart, firstly give a control limits show, and then set up a control chart and a simulation study. After study on single parameters, we study on a joint innovatively. The above is the study about known parameters. When the parameters unknown, we use method of parameter estimation and discuss the control effect of the sample amount. Simulation results show that, when parameters unknown, our methods of constructing the shape parameter control chart performance are still better than Pascual’s control chart performance. The difficulty of this article is, we cannot find the unbiased control chart for joint control charts, because of it is almost impossible to search the false alarm rate c...