Files in this item

FilesDescriptionFormat

application/pdf

application/pdf9215875.pdf (12MB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:The development of performance measures and diagnostic procedures for X(bar) and S(2) charts
Author(s):Rahn, Gregory Ernest
Doctoral Committee Chair(s):Kapoor, Shiv G.
Department / Program:Industrial and Enterprise Systems Engineering
Discipline:Industrial Engineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Statistics
Industrial Engineering
Abstract:In today's competitive marketplace, companies must continually strive for improved quality and productivity of its products and processes just to maintain its competitive position. For this reason, quality systems were developed to improve quality through the design and manipulation of process resources over time. When examining a specific portion of a process, called a QC window, statistical process control methods, such as Shewhart charts, are responsible for the observation, evaluation, and diagnosis of processes to aid in the decision and implementation of corrective action.
This dissertation develops methods to improve the evaluation and diagnostic elements of the QC window. To enhance the evaluation capability of Shewhart control charts, performance measures in terms of operational characteristics, are defined and developed for the X chart, S$\sp2$ chart, and both charts combined. These performance measures are analyzed with respect to process shifts in both mean and variance. In addition, performance is enhanced by: (a) enforcing four rules on the S$\sp2$ chart to improve sensitivity to variance shifts (b) modifying the control limits on both the X and S$\sp2$ charts to capitalize on the distinct relationships between rules as well as quantifying the tradeoff between the probability of a false alarm ($\alpha$) and the probability of a false indication of control ($\beta$).
To support the diagnostic capability of Shewhart charts, a sequential diagnostic procedure is developed that estimates the mean and variance of a process given observed control chart performance. Once the current state of the process is identified, the decision and implementation of corrective action can then be performed to position the process at its desired state. By improving the evaluation and diagnostic capability of control charts, the work presented in this thesis expands the role of control charts, thus enabling them to reach their full potential.
Issue Date:1992
Type:Text
Language:English
URI:http://hdl.handle.net/2142/19024
Rights Information:Copyright 1992 Rahn, Gregory Ernest
Date Available in IDEALS:2011-05-07
Identifier in Online Catalog:AAI9215875
OCLC Identifier:(UMI)AAI9215875


This item appears in the following Collection(s)

Item Statistics