The performance of robust multivariate statistical control charts based on different cutoff-points with sustained shifts in mean.
A separate univariate control chart for each characteristic is often used to detect changes in the inherent variability of a process because of ease of computation. Nevertheless, the shortcoming of using separate individual charts would not have detected out of control condition when the characteris...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Asian Network for Scientific Information
2011
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/24745/ http://www.ansinet.com/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | A separate univariate control chart for each characteristic is often used to detect changes in the inherent variability of a process because of ease of computation. Nevertheless, the shortcoming of using separate individual charts would not have detected out of control condition when the characteristics of interests are correlated. The problems get more complicated in the existence of sustained shift in the process mean. In this study, we proposed a robust multivariate control chart which is less sensitive to the sustained shift in mean process. However, the theoretical cutoff-points of the proposed charts are intractable. As an alternative, we proposed two different procedures of empirical cutoff-points. The performance of these two cutoff-points in detecting a step shift in the mean vector is investigated extensively by real example and monte carlo simulation. |
|---|