The Optimal Design And Analysis Of Process Control Charts

Control chart is a kind of charts that are used to analysis and determinate whether a process is in-control or not. It use a graph to illuminate the qualityvariations in production process, and help to find the sources of the variations-common causes or special causes. It is a useful statistical method to keep a process in a stable situation. Control chart is one way to deploy quality management and control in a corporation, it is vitally important to make sure produce high-quality productions. In this paper, the corn purpose is to build an effective method to design control charts. At the beginning, in this paper we talk about some background of quality management and statistical process control, and then point out and analyze some defects of Shewhart control that had been used widely. A simple but effective method to aid in design control schemes is presented. Commonly, subgroup size, sampling rate, and control limit are three main parameters in designing a control schemes. Average production length (APL) is introduced as the criterion for measuring control scheme performance. A model is developed with sampling rate and maximum false alarm rate as inputs to determine the optimum subgroup size and control limits as measured by APL.Based on this model, this method is then used to design some other special control schemes, such as skewness distributions, small batches, small shifts, and attribute control charts.Finally, in order to improve the usability of this method, in this paper we present a way to design control charts using the theory of robust design, and then some examples are given to illustrate how to use this method in practice.