Penambahbaikan Untuk Pelbagai Skema Carta Kawalan EWMA
Loading...
Date
2003-04
Authors
Tuh, Moi Hua
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The EWMA (Exponentially Weighted Moving Average) control chart which is also known
as a memory control chart is used widely for the monitoring of the quality of a product
which is produced by a manufacturing process. In most manufacturing processes, a large
sample size is easier for the detection of small shifts in the process. However, there are
many situations in which the sample size used in the monitoring of a process is n = 1, i.e.,
the sample consists of an individual unit. Methodologies in the construction of the EWMA
control charts for sample statistics of a subgroup size, n, and individual data are discussed
in this dissertation. Among the control charts that have been suggested by Ng and Case
(1989) such as the EWMASM (EWMA of sample mean), EWMASR (EWMA of sample
range), EWMAID (EWMA of individual measurements) and EWMAMR (EWMA of
moving range of subgroup size two), only the EWMASR and EWMAID control charts are
considered in my study. The main aim and objective of this dissertation is to propose the
use of a Fast Initial Response (FIR) feature for the EWMASR and EWMAID control charts
as well as to study their performances. The performance of the EWMASR and FIR
EWMASR charts in the monitoring of the process standard deviation by changing the
sample size from n=5 to n=10 is also studied. In this dissertation, all the simulation studies
are done using the SAS (Statistical Analysis System) version 6.12 program.
Description
Keywords
Mathematical Sciences , Exponentially Weighted Moving Average