Cadangan Carta Kawalan Teguh Purata Bergerak Berpembera Teksponen Untuk Varians Proses
Loading...
Date
2005-03
Authors
Fern Fern, Lim
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In this study, we suggest an approach for the monitoring of the process variance of the data
which contain outliers. The chart which is usually used in the monitoring of the process
variance for contaminated data is the RQ chart. However, this chart is less sensitive in the
detection of outliers involving small variance. To overcome this sho11coming, the
EWMASRQ chart is proposed. The EWM.ASRQ chart is a good alternative in the
monitoring of the process variance for contaminated data, i.e., those that contain outliers.
Outliers consist of observations whose values are larger or smaller than most of the data in
a process. The presence of the outliers indicate the occurrence of out-of-control signals
which must be detected by a control chart. The control limits of the EWMASRQ chart are
based on the interquartile range estimator, hence they are less affected by outliers. Thus, the
distance of the limits of the EWMASRQ chart will not be stretched by the presence of
outliers in a process. This makes the proposed EWMASRQ chart sensitive in the detection
of outliers. The performance of the EWMASRQ chart will be evaluated for four different
cases that will be discussed later. Examples that show how the EWMASRQ chart is used in
real situations will also be given.
Description
Keywords
Mathematics , Process control -- Statistical methods