1、Zero Acceptance Number Sampling PlansFifth EditionAlso available from ASQ Quality Press:The Handbook of Applied Acceptance Sampling: Plans, Procedures and PrinciplesKenneth S. StephensProcess Quality Control: Troubleshooting and Interpretation of Data, Fourth EditionEllis R. Ott, Edward G. Schilling
2、, and Dean V. NeubauerGlossary and Tables for Statistical Quality Control, Fourth EditionASQ Statistics DivisionFailure Mode and Effect Analysis: FMEA From Theory to Execution, Second EditionD.H. StamatisThe Weibull Analysis Handbook, Second EditionBryan DodsonStatistical Engineering: An Algorithm f
3、or Reducing Variation in Manufacturing ProcessesStefan H. Steiner and R. Jock MacKayIntegrating Inspection Management into Your Quality Improvement SystemWilliam D. MawbyMake Your Destructive, Dynamic, and Attribute Measurement System Work For YouWilliam D. MawbyLean Kaizen: A Simplified Approach to
4、 Process ImprovementsGeorge Alukal and Anthony ManosThe Certified Quality Inspector HandbookH. Fred Walker, Ahmad Elshennawy, Bhisham C. Gupta, and Mary McShane VaughnRoot Cause Analysis: Simplified Tools and Techniques, Second EditionBjrn Andersen and Tom FagerhaugThe Certified Manager of Quality/O
5、rganizational Excellence Handbook: Third EditionRussell T. Westcott, editorTo request a complimentary catalog of ASQ Quality Press publications, call 800-248-1946, or visit our Web site at http:/www.asq.org/quality-press.Zero Acceptance Number Sampling PlansFifth EditionNicholas L. SquegliaASQ Quali
6、ty Press Milwaukee, WisconsinAmerican Society for Quality, Quality Press, Milwaukee 53203 2008 by American Society for QualityAll rights reserved. Published 2008Printed in the United States of America14 13 12 11 10 09 08 5 4 3 2 1Library of Congress Cataloging-in-Publication DataSqueglia, Nicholas L
7、. Zero acceptance number sampling plans / Nicholas L. Squeglia.5th ed. p. cm. ISBN 978-0-87389-739-6 (alk. paper) 1. Acceptance sampling. I. Title. TS156.4.S68 2008 658.562dc22 2008009352No part of this book may be reproduced in any form or by any means, electronic, mechanical, photocopying, recordi
8、ng, or otherwise, without the prior written permission of the publisher.Publisher: William A. TonyAcquisitions Editor: Matt MeinholzProject Editor: Paul OMaraProduction Administrator: Randall BensonASQ Mission: The American Society for Quality advances individual, organizational, and community excel
9、lence world-wide 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 instruction
10、al 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 ou
11、r Web site at www.asq.org or http:/www.asq.org/quality-press.Printed in the United States of AmericaPrinted on acid-free paperThis book is dedicated to my wife Joan; my children Vanessa, Nicholas, and Jacqueline; and my grandson Shane.viiviiCONTENTSList of Figures and Tables . . . . . . . . . . . .
12、. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ixPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiIntroduction . . . . . . . . . . . .
13、 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Attribute Sampling Plans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2Nonstatistical Sampling Pl
14、ans. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2Relationship of c=0 Plans to ANSI Z1.4 Plans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Estimating Potential Savings . . . . . .
15、 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Why Constant Sample Sizes Are Not Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Use of the c=0 Plans Table . . . . . . . . . . . . . . .
16、 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Physically Taking the Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Comments of the AOQL . . . . . . . . . . . . . . . . . . . .
17、 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Background Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Adjustments from MIL-STD-105E/ANSI Z1.4 . . . . . . . . . . . .
18、. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12Sampling Plan “Switching” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12Switching Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
19、 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14Operating Characteristic Curves and Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15Small Lot Supplement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
20、 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33ixixLiST Of figurES aND TabLESfigure 1 Operating characteristic curve . . . .
21、 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4figure 2 Effects of acceptance numbers on OC curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Table 1 c=0 sampling plans. . . . . . . . . . . . . . . . . . . . . . . . . . .
22、. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7figure 3 AOQL curve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Table 2 AOQLs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
23、. . . . . . . . . . . . . . . . . . . . . . . . . . . . 10figure 4 Curves for determining AOQL values. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Table 3 Inspection results from a large receiving inspection department over a one-month period. . . . . . . .
24、. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Table 4 Receiving inspection figures for one month. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Table 5 Receiving inspection figures for one month. . . . . . . .
25、. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Table 6 Receiving inspection figures for one month. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13OC curves for single sampling plans, acceptance number equal to zero (various sample sizes) . . . . . . .
26、. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15Table 7 Small lot size supplement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31xixiPrEfaCEFor many years, the acceptable quality level (AQL)
27、1concept was used largely because of the influence of MIL-STD-105 and its revisions. However, in the current business climate of immense worldwide competition and greater demands by customers, more companies are realizing that quality control does not costit pays. As a result, the prevailing wisdom
28、has moved toward zero defects, and AQLs are no longer the rule, but the exception; they are simply not compatible with todays environment. Many companies are striving for zero defects through statistical process control, improved processes, closed loop inspection systems, and other means.The use of
29、sampling plans with zero acceptance numbers is the norm today. The sampling plans in this book (c=0) actually represent a revision in 1963 of similar plans I developed in 1961. Because of the wide-spread use of MIL-STD-105C in 1961, the only way to depart from this standard was to develop a set of p
30、lans that could be compared favorably with the military standard. The c=0 plans were developed and originally implemented in a medium-sized plant that did both military and commercial work. Although the plans were not formally approved, there was no opposition to them.In 1963, MIL-STD-105D was intro
31、duced, and the c=0 plans were updated and revised. This time, the plans were proposed to a large aerospace manufacturer with a staff of resident government quality control representatives. It was necessary to deliver a formal presentation and explain the c=0 plans in great detail. The aerospace manu
32、facturer and the government representatives agreed to accept the plans on a trial basis. While the plans were targeted essentially to the limiting quality (LQ) percentages in the military standard tables, there were departures from these targets in several instances. These special adjustments were n
33、eces-sary to maintain the logic of the c=0 plans. These adjustments were highlighted during the presentation to the aerospace manufacturer.The results of the trial period at this company were excellent. Not only were the savings significant, but there was a significant reduction in assembly problems
34、 as well. A check with the company in 1983 (20 years later) revealed the c=0 plans were still being used.The c=0 sampling plans were presented in a national publication in 1965 (N. L. Squeglia, “Sampling Plans for Zero Defects,” Quality Assurance 4 August 1965: 28). The inquiries and interest genera
35、ted by the article prompted me to write the first edition of this book. Published in 1969, it described the plans in more detail and contained operating characteristic (OC) curves. The continual interest in the c=0 plans and encouragement from Professor N. L. Enrick of Kent State University resulted
36、 in the publication of the second edition in 1981.In 1983, I conducted an informal survey to get some idea of the extent of the savings realized by users of the c=0 plans who had switched from MIL-STD-105D. A few said it was too early to tell, but the majority re-ported a range of savings from 8 per
37、cent to 30 percent, with an average of 18 percent. Of course, the extent of savings is based on the lot sizes and index value (associated AQL) used. The larger the lot and index value, the greater the savings. It is not necessary to implement the plans to determine the savings. The savings potential
38、 can be evaluated from past data as described later, in the section titled “Estimating Potential Savings.”While the hypergeometric distribution was used originally to maximize mathematical accuracy, it is my humble opinion that the most important feature of the plans is the philosophy of zero defect
39、s.The c=0 sampling plans are now in wide use throughout the United States and in other countries. In 1983, the c=0 plans became a part of the Department of Defenses DLAM 8200.2 for use by government Defense Contract Administration Services quality assurance representatives. In 1989, MIL-STD-105E sup
40、er-seded 105D. This revision placed emphasis on the use of 105E as a guide in developing inspection strategies, and it recognized the limitation of the AQL concept. The sampling plans were not changed. As a result of this 1. ANSI Z1.4 changed “acceptable quality level” to “acceptance quality limit”
41、and provided an explanation. The c=0 plans are not AQL plans. For comparison purposes, they are associated with particular ANSI plans. See Table 1; the numbers are index values.xii 1revision, the fourth edition of this book was published in 1994. That edition included important updates and a small-l
42、ot sampling table.In the earliest editions it was necessary to compare the c=0 plans to the MIL-STD plans in order to show the advantages of the c=0 plans. The comparisons provide important information as to their derivation that provided for a smooth transition from the MIL-STD plans to the c=0 pla
43、ns, as discussed in the section titled “Relationship of c=0 Plans to ANSI Z1.4 Plans.” After years of extensive application by government contrac-tors, commercial manufacturing, and service industries, the c=0 sampling plans presented in this book are now considered stand-alone sampling plans, altho
44、ugh the c=0 plans provide equal or greater protection than the MIL-STD plans.In 2000 the Department of Defense declared MIL-STD-105E obsolete and recommended that the c=0 plans from this book be used in its place. Upon cancellation of MIL-STD-105E, the c=0 plans were authorized for use by the Defens
45、e Contract Management Agency/Department of Defense. This correspondence in part reads:Zero Acceptance Sampling Plans By Nicholas L. Squeglia This book gives a number of zero based sampling plans and their corresponding Operating Characteristic (OC) Curves and values. It is the state of the art in ze
46、ro based sampling plans. (DCMA Guidebook)The companies who were using the 105E plans at the time of their obsolescence switched to ANSI Z1.4 (2003), which, ironically, is a virtual copy of the AQL-oriented 105E plans.In the early 2000s, United Technologies issued the c=0 table to their suppliers as
47、part of their supplier Quality Requirements. The table was reprinted with the permission of ASQ.In May 2005, I was awarded ASQs Shainin Medal at the World Quality Congress in Milwaukee for my c=0 sampling plans. The citation read:Nicholas L. Squeglia is credited as one of the most significant contri
48、butors to the effort for driving Zero Defects by developing a set of zero acceptance number sampling plans (c=0) forcing preventive actions, thus saving millions of dollars in military and commercial applications. There was wide-spread acceptance of the plans because they proved to be practical, sim
49、ple to use, and economical.The c=0 plans have been continually gaining in popularity for more than 45 years.Nicholas L. Squegliaxii 1iNTrODuCTiONThe zero acceptance number plans developed by the author were originally designed and used to provide over-all equal or greater consumer protection with less inspection than the corresponding MIL-STD-105 sampling plans. In addition to the economic advantages they offer, these plans are simple to use and administer. Because of these advantages and because greater emphasis is now being placed on zero defects and product liability preventio
