- Western Electric rules
In
Statistical Process Control , the Western Electric Rules aredecision rules for detecting "out-of-control" or non-random conditions oncontrol chart s.Citation
last = Montgomery
first = Douglas C.
publication-date = 2005
title = Introduction toStatistical Quality Control
edition = 5
publication-place =Hoboken, New Jersey
publisher =John Wiley & Sons
isbn = 9780471656319
oclc = 56729567 Locations of the observations relative to the control chart control limits (typically at ±3standard deviation s) and centerline indicate whether the process in question should be investigated forassignable cause s. The Western Electric Rules were codified by a specially-appointed committee of the manufacturing division of theWestern Electric Company and appeared in the first edition of itsStatistical Quality Control Handbook in 1956.Citation
last =Western Electric Company
publication-date = 1956
title =Statistical Quality Control handbook.
edition = 1
publication-place =Indianapolis, Indiana
publisher = Western Electric Co.
oclc = 33858387
page = v Their purpose was to ensure that line workers and engineers interpret control charts in a uniform way.Citation
last =Western Electric Company
publication-date = 1956
title =Statistical Quality Control handbook.
edition = 1
publication-place =Indianapolis, Indiana
publisher = Western Electric Co.
oclc = 33858387
page = 25Motivation
The rules attempt to distinguish
unnatural pattern s fromnatural pattern s based on several criteria:Citation
last =Western Electric Company
publication-date = 1956
title =Statistical Quality Control handbook.
edition = 1
publication-place =Indianapolis, Indiana
publisher = Western Electric Co.
oclc = 33858387
page = 24
#The absence of points near the centerline (identified as a mixture pattern)
#The absence of points near the control limits (identified as a stratification pattern)
#The presence of points outside the control limits (identified as an instability pattern)
#Other unnatural patterns (systematic (autocorrelative), repetition, trend patterns)To achieve this, the rules divide the chart into zones, measured in units of
standard deviation (σ) between the centerline and control limits, as follows:For other control charts based on
skewed distributions, the Handbook recommends:
#Finding the probabilities associated with each zone
#Devising tests that should result in the desired sensitivity to the presence of assignable causesOther unnatural patterns
The Handbook also identifies patterns that require consideration of both the upper and lower halves of the control chart together for identification:
ee also
*
Nelson rules References
Wikimedia Foundation. 2010.