Abstract
In semiconductor manufacturing it is useful to design control charts to be sensitive to anticipated assignable causes. Process knowledge can be used to develop summaries for run-to-run control that are more sensitive to problems than traditional approaches. As the number of measurements recorded from a run increases, appropriate summaries derived from process knowledge become more important. Control charts based on linear contrasts are generated from partitions of a standard analysis of variance model. These contrasts simplify the development and interpretation of charts sensitized for specific assignable causes. The use of EWMA control charts for these statistics is described.
Original language | English (US) |
---|---|
Pages (from-to) | 261-272 |
Number of pages | 12 |
Journal | Quality and Reliability Engineering International |
Volume | 14 |
Issue number | 4 |
DOIs | |
State | Published - Aug 1998 |
Keywords
- Average run length
- Exponentially weighted moving average
- Statistical process control
ASJC Scopus subject areas
- Safety, Risk, Reliability and Quality
- Management Science and Operations Research