Summary: | 碩士 === 國立雲林科技大學 === 工業工程與管理研究所 === 83 === Traditional statistical process control charts assume that
process observation are independent and identically distributed.
In practice , however , autocorrelations and other systematic
time-series effects frequently occur in process observations.
Many research results indicate that even a small autocorrelation
will have significant effects on traditional control chart .
Therefore , we try to investigate the effects of correlated data
to Shewhart and EWMA charts , and compare them to the independ-
ent data . The purpose of this research is to provide a better
understanding on the effects of correlated data to traditional
control charts .
Many authors have provide different methods in dealing with
correlated data . However , due to different viewpoints in
analysis , the practitioners still can't choose a suitable
control chart . Therefore , another purpose of this research
is to perform the sensitivity analysis of Shewhart , EWMA and
SCC charts for correlated data . Simulation is conducted to find
the relationship between the autocorrelation coefficient and
average run length , in order to provide a criterion for pract-
itioners on choosing suitable control charts .
Results show that EWMA chart is suitable for negative correlated
data compared to independent data . In detecting shift of
process mean , SCC control chart is better for large shifts
while EWMA control chart is better for small shifts .
|