On the performance of modified runs rules charts with estimated parameters. Issue 2 (7th February 2017)

• Record Type: Journal Article
• Authors: Rakitzis, Athanasios C.
• Is Part Of: Communications in statistics. Volume 46:Issue 2(2017)
• Page Start: 1360
• Page End: 1380
• Publication Date: 2017-02-07
• Subjects: Average run length (ARL) -- Estimation effect -- Marginal run-length distribution -- Markov chain -- Runs rules -- Shewhart control charts -- Standard deviation of the run length (SDRL); 62P30; Mathematical statistics -- Periodicals; Mathematical statistics -- Data processing -- Periodicals; Digital computer simulation -- Periodicals; 519.5
• DOI: 10.1080/03610918.2014.1002846
• Languages: English
• ISSNs: 0361-0918

Abstract

ABSTRACT: It is known that Shewhart chart is not sensitive in the detection of small process average shifts. A solution to this problem is the use of stopping rules based on the theory of runs. However, their practical implementation is based on the assumption that the parameters of the process are known, which rarely happens in practice. In this work, the performance of modified runs-rules schemes is studied (using a Markov chain technique) when the parameters of the process have been estimated from a preliminary in-control sample. Guidelines regarding the effective application of these schemes in practice are also provided.