Statistical Engineering

R. Jock MacKay, Stefan H. Steiner

Description

Reducing the variation in process outputs is a key part of process improvement. For mass produced components and assemblies, reducing variation can simultaneously reduce overall cost, improve function and increase customer satisfaction with the product. The authors have structured this book around an algorithm for reducing process variation that they call "Statistical Engineering." The algorithm is designed to solve chronic problems on existing high to medium volume manufacturing and assembly processes. The fundamental basis for the algorithm is the belief that we will discover cost effective changes to the process that will reduce variation if we increase our knowledge of how and why a process behaves as it does. A key way to increase process knowledge is to learn empirically, that is, to learn by observation and experimentation. The authors discuss in detail a framework for planning and analyzing empirical investigations, known by its acronym QPDAC (Question, Plan, Data, Analysis, Conclusion). They classify all effective ways to reduce variation into seven approaches. A unique aspect of the algorithm forces early consideration of the feasibility of each of the approaches. Also includes case studies, chapter exercises, chapter supplements, and six appendices. PRAISE FOR Statistical Engineering "I found this book uniquely refreshing. Don't let the title fool you. The methods described in this book are statistically sound but require very little statistics. If you have ever wanted to solve a problem with statistical certainty (without being a statistician) then this book is for you. - A reader in Dayton, OH "This is the most comprehensive treatment of variation reduction methods and insights I’ve ever seen."- Gary M. Hazard Tellabs "Throughout the text emphasis has been placed on teamwork, fixing the obvious before jumping to advanced studies, and cost of implementation. All this makes the manuscript !attractive for real-life application of complex techniques." - Guru Chadhabr Comcast IP Services COMMENTS FROM OTHER CUSTOMERS Average Customer Rating (5 of 5 based on 1 review) "This is NOT a typical book on statistical tools. It is a strategy book on how to search for cost-effective changes to reduce variation using empirical means (i.e. observation and experiment). The uniqueness of this book: Summarizes the seven ways to reduce variation so we know the goal of the data gathering and analysis, present analysis results using graphs instead of P-value, and integrates Taguchi, Shainin methods, and classical statistical approach. It is a must read for those who are in the business of reducing variation using data, in particular for the Six Sigma Black Belts and Master Black Belts. Don't forget to read the solutions to exercises and supplementary materials to each chapter on the enclosed CD-ROM." - A. Wong, Canada

Sujets

• Business
• Qualité, hygiène, sécurité, environnement
• Operations Management
• Production Management
• Gestion et management
• Economics

Informations

• Publié par: ASQ Quality Press
• Date de parution: 02 janvier 2005
• Nombre de lectures: 1
• EAN13: 9780873891363
• Langue: English
• Poids de l'ouvrage: 10 Mo

Extrait

H1212CH00 FM.qxd 3/31/05 10:29 AM Page i
Statistical EngineeringH1212CH00 FM.qxd 3/31/05 10:29 AM Page ii
Also Available from ASQ Quality Press
Applied Statistics for the Six Sigma Green Belt
Bhisham C. Gupta and H. Fred Walker
Applied Data Analysis for Process Improvement: A Practical Guide to Six Sigma Black
Belt Statistics
James L. Lamprecht
Design of Experiments with MINITAB
Paul Mathews
Failure Mode and Effect Analysis: FMEA From Theory to Execution, Second Edition
D. H. Stamatis
Statistical Quality Control Using Excel, Second Edition
Steven M. Zimmerman, Ph.D. and Marjorie L Icenogle, Ph.D.
The Desk Reference of Statistical Quality Methods
Mark L. Crossley
Improving Performance Through Statistical Thinking
ASQ Statistics Division
Six Sigma for the Shop Floor: A Pocket Guide
Roderick A. Munro
Defining and Analyzing a Business Process: A Six Sigma Pocket Guide
Jeffrey N. Lowenthal
To request a complimentary catalog of ASQ Quality Press publications, call 800-248-1946,
or visit our Web site at http://qualitypress.asq.org.H1212CH00 FM.qxd 3/31/05 10:29 AM Page iii
Statistical Engineering
An Algorithm for Reducing Variation
in Manufacturing Processes
Stefan H. Steiner and R. Jock MacKay
ASQ Quality Press
Milwaukee, WisconsinH1212CH00 FM.qxd 3/31/05 10:29 AM Page iv
American Society for Quality, Quality Press, Milwaukee 53203
© 2005 by
All rights reserved. Published 2005
Printed in the United States of America
12 11 10 09 08 07 06 05 5 4 3 2 1
Library of Congress Cataloging-in-Publication Data
Steiner, Stefan H., 1964–
Statistical engineering : an algorithm for reducing variation in manufacturing
processes / by Stefan H. Steiner and R. Jock MacKay.
p. cm.
Includes bibliographical references.
ISBN 0-87389-646-7
1. Production management. 2. Manufacturing processes. I. MacKay, R. Jock.
II. Title.
TS155.S773 2005
658.5—dc22
2004030676
No part of this book may be reproduced in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise, without the prior written permission of the publisher.
Publisher: William A. Tony
Acquisitions Editor: Annemieke Hytinen
Project Editor: Paul O’Mara
Production Administrator: Randall Benson
ASQ Mission: The American Society for Quality advances individual, organizational, and
community excellence worldwide through learning, quality improvement, and knowledge
exchange.
Attention Bookstores, Wholesalers, Schools, and Corporations: ASQ Quality Press books,
videotapes, audiotapes, and software are available at quantity discounts with bulk purchases for
business, educational, or instructional use. For information, please contact ASQ Quality Press at
800-248-1946, or write to ASQ Quality Press, P.O. Box 3005, Milwaukee, WI 53201-3005.
To place orders or to request a free copy of the ASQ Quality Press Publications Catalog, including
ASQ membership information, call 800-248-1946. Visit our Web site at www.asq.org or
http://qualitypress.asq.org.
Printed on acid-free paperH1212CH00 FM.qxd 3/31/05 10:29 AM Page v
To Anne Marie, Erik, and Emily—S.H.S
To Samm—R.J.M.H1212CH00 FM.qxd 3/31/05 10:29 AM Page viH1212CH00 FM.qxd 3/31/05 10:29 AM Page vii
Contents
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxi
Chapter 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Truck Pull 2
1.2 Engine Block Leaks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Camshaft Lobe Runout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 Sand Core Strength 7
1.5 Crankshaft Main Diameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.6 Paint Film Build . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.7 Refrigerator Frost Buildup 9
PART I SETTING THE STAGE 11
Chapter 2 Describing Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1 Language of Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Causes of Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3 Displaying and Quantifying Process Variation . . . . . . . . . . . . . . . . . . . . . . 18
2.4 Models for Variation and the Effects of Causes . . . . . . . . . . . . . . . . . . . . . 23
Chapter 3 Seven Approaches to Variation Reduction . . . . . . . . . . . . . . . . . . . 29
3.1 Fixing the Obvious Based on Knowledge of a Dominant Cause . . . . . . . . 30
3.2 Desensitizing the Process to Variation in a Dominant Cause . . . . . . . . . . . 32
3.3 Feedforward Control Based on a Dominant Cause . . . . . . . . . . . . . . . . . . 33
3.4 Feedback Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.5 Making the Process Robust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.6 100% Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.7 Moving the Process Center 39
viiH1212CH00 FM.qxd 3/31/05 10:29 AM Page viii
viii Contents
Chapter 4 An Algorithm for Reducing Variation . . . . . . . . . . . . . . . . . . . . . . 41
4.1 The Statistical Engineering Variation Reduction Algorithm . . . . . . . . . . . 41
4.2 How to Use the Algorithm Effectively . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Chapter 5 Obtaining Process Knowledge Empirically . . . . . . . . . . . . . . . . . . 51
5.1 Question, Plan, Data, Analysis, and Conclusion (QPDAC)
Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5.2 Examples 58
5.3 Summary 64
PART II GETTING STARTED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Chapter 6 Defining a Focused Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.1 From a Project to Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.2 The Problem Baseline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
6.3 Planning and Conducting the Baseline Investigation . . . . . . . . . . . . . . . . . 76
6.4 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
6.5 Completing the Define Focused Problem Stage . . . . . . . . . . . . . . . . . . . . . 86
Chapter 7 Checking the Measurement System . . . . . . . . . . . . . . . . . . . . . . . . 89
7.1 The Measurement System and its Attributes . . . . . . . . . . . . . . . . . . . . . . . 90
7.2 Estimating Measurement Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
7.3 Estimating Measurement Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
7.4 Improving a Measurement System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7.5 Completing the Check the Measurement System Stage . . . . . . . . . . . . . . . 102
Chapter 8 Choosing a Working Variation Reduction Approach . . . . . . . . . . 105
8.1 Can We Find a Dominant Cause of Variation? . . . . . . . . . . . . . . . . . . . . . . 106
8.2 Can We Meet the Goal by Shifting the Process Center Without
Reducing Variation? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
8.3 Can We Reduce Variation by Changing One or More Fixed Inputs
Without Knowledge of a Dominant Cause? . . . . . . . . . . . . . . . . . . . . . . . . . 110
8.4 Does the Process Output Exhibit a Strong Pattern Over Time? . . . . . . . . . 112
8.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
PART III FINDING A DOMINANT CAUSE OF VARIATION . . . . . . . . . . . . . . . 115
Chapter 9 Finding a Dominant Cause Using the Method of Elimination . . . 117
9.1 Families of Causes of Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
9.2 . . . . . . . . . . 121
9.3 Implementing the Method of Elimination . . . . . . . . . . . . . . . . . . . . . . . . . 124H1212CH00 FM.qxd 3/31/05 10:29 AM Page ix
Contents ix
Chapter 10 Investigations to Compare Two Families of Variation . . . . . . . . . 131
10.1 Stratification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
10.2 Comparing Two Time-Based Families . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
10.3 Comparing Upstream and Downstream Families . . . . . . . . . . . . . . . . . . 138
10.4Assembly and Component F . . . . . . . . . . . . . . . . . . . 141
10.5 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
Chapter 11 Investigations to Compare Three or More Families
of Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
11.1 Multivari Investigations: Comparing Time- and
Location-Based Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
11.2 Comparing Families Defined by Processing Steps . . . . . . . . . . . . . . . . . 163
11.3 Comparing Component Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Chapter 12 Investigations Based on Single Causes . . . . . . . . . . . . . . . . . . . . . . 179
12.1 Group Comparison: Comparing Parts With Binary Output . . . . . . . . . . . 179
12.2 Investigating the Relationship Between Inputs and
a Continuous Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Chapter 13 Verifying a Dominant Cause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
13.1 Verifying a Single Suspect Dominan