Activity Number:
|
385
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 15, 2002 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Physical & Engineering Sciences*
|
Abstract - #301212 |
Title:
|
The MEWMA Control Chart When Different Variables Have Different Sample Sizes
|
Author(s):
|
Keunpyo Kim*+ and Marion Reynolds, Jr.
|
Affiliation(s):
|
Virginia Polytechnic Institute and State University and Virginia Polytechnic Institute and State University
|
Address:
|
, Blacksburg, Virginia, 24061,
|
Keywords:
|
Average Run Length ; Measurement Costs ; MEWMA ; Multivariate Quality Control ; Statistical Quality Control
|
Abstract:
|
The multivariate exponentially-weighted moving average (MEWMA) control chart is one of the major tools for multivariate monitoring. It is usually assumed that the same sample size is used for each variable. But when multiple variables are monitored for a process, different variables may have different measurement costs. MEWMA control charts with different sample sizes based on different measurement costs are considered to improve the performance of a process. The average run length (ARL) is computed using simulation and compared for some selected sets of sample sizes. When different variables have significantly different measurement costs, using different sample sizes can be better than using the same sample size for each variable.
|