Control Charts Under Linear Trend,
Abstract
The performance of control charts is usually evaluated by assuming a step change in the process mean. However, it is more appropriate to evaluate the performance of control charts by assuming a drift in the mean for processes where a gradual drift models the shift in the mean more accurately. Three major methods for computing the average run length (ARL) of control charts assuming a step change in the mean are reviewed. Generalizations of these methods for computing the ARL of control charts assuming a drift in the mean are then examined. Average run length; Cumulative sum; Exponentially weighted moving average; Integral equation; Linear drift; Markov chain; Normal distribution; Statistical process control.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1992
- Accession Number
- ADP007209
Entities
People
- F. F. Gan
Organizations
- National University of Singapore