An improved double sampling s chart |
| |
Authors: | D He A Grigoryan |
| |
Affiliation: | Department of Mechanical and Industrial Engineering , University of Illinois at Chicago , 842 West Taylor Street, 3049 ERF, Chicago, IL, 60607, USA |
| |
Abstract: | Double sampling (DS) s charts are designed to allow quick detection of a small shift in process standard deviations and to provide a quick response in an agile manufacturing environment. However, current developed DS s charts assume that the sample standard deviations follow a normal distribution. Although valid for relatively large sample sizes, this assumption has limited the application of DS s charts to the monitoring of manufacturing processes with relatively small sample sizes. An improved DS s chart is developed without the normality assumption of the sample standard deviations. The design of the improved DS s chart is formulated as a statistical design optimization problem and solved with a genetic algorithm. The efficiency of the improved DS s charts is compared with that of the DS s charts developed in previous research and with that of the traditional s charts. |
| |
Keywords: | |
|
|