Abstract
In this paper, Gamma(λ, 2) distribution is considered as a failure model for the economic statistical design of X̄ control charts. The study shows that the statistical performance of control charts can be improved significantly, with only a slight increase in the cost, by adding constraints to the optimization problem. The use of an economic statistical design instead of an economic design results in control charts that may be less expensive to implement, that have lower false alarm rates, and that have a higher probability of detecting process shifts. Numerical examples are presented to support this proposition. The results of economic statistical design are compared to those of a pure economic design. The effects of adding constraints for statistical performance measures, such as Type I error rate and the power of the chart, are extensively investigated.
| Original language | English |
|---|---|
| Pages (from-to) | 645-654 |
| Number of pages | 10 |
| Journal | Computers and Industrial Engineering |
| Volume | 43 |
| Issue number | 3 |
| DOIs | |
| State | Published - Sep 2002 |
Bibliographical note
Funding Information:Financial assistance from the Natural Science and Engineering Research Council (NSERC) of Canada for the support of this collaborative research is gratefully acknowledged. The authors also acknowledge the King Fahd University of Petroleum and Mineral for facilitating this project. The valuable suggestions of the referee are greatly appreciated.
Keywords
- Economic design
- Economic statistical design
- Gamma in-control times
- Variable sampling intervals
ASJC Scopus subject areas
- General Computer Science
- General Engineering