Premium
Detecting lack of control in a new, untried process
Author(s) -
Murray Ian,
Oakland John S.
Publication year - 1988
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.4680040407
Subject(s) - cusum , control chart , shewhart individuals control chart , statistical process control , control limits , process (computing) , standard deviation , computer science , process control , control (management) , sensitivity (control systems) , sampling (signal processing) , statistics , engineering , mathematics , ewma chart , artificial intelligence , computer vision , filter (signal processing) , electronic engineering , operating system
In his original work on the use of control charts, Shewhart distinguished between a long running known process and a new, untried process. In the latter case, the control limits are obtained from the sample data. If the process variability increases during the sampling period, the control limits will widen, reducing the sensitivity of the standard deviation and range charts and producing charts which are in control for out‐of‐control situations. Simulations were produced to investigate this problem through the eyes of the operator. Cusum procedures were used to provide the operator with a means of controlling process variability which is more sensitive to change than the control chart alone. The method was applied to data taken from an industrial situation.