How to Perform Continuous Sampling

  • Published on
    27-Jan-2017

  • View
    215

  • Download
    2

Transcript

  • American Society for Quality

    How to Perform Continuous Sampling by Kenneth S. StephensReview by: S. Lynne StokesTechnometrics, Vol. 38, No. 1 (Feb., 1996), p. 82Published by: American Statistical Association and American Society for QualityStable URL: http://www.jstor.org/stable/1268914 .Accessed: 09/07/2014 01:25

    Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

    .JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

    .

    American Statistical Association and American Society for Quality are collaborating with JSTOR to digitize,preserve and extend access to Technometrics.

    http://www.jstor.org

    This content downloaded from 131.155.2.68 on Wed, 9 Jul 2014 01:25:07 AMAll use subject to JSTOR Terms and Conditions

    http://www.jstor.org/action/showPublisher?publisherCode=astatahttp://www.jstor.org/action/showPublisher?publisherCode=asqhttp://www.jstor.org/stable/1268914?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp

  • BOOK REVIEWS

    How to Perform Continuous Sampling (2nd Ed.), by Kenneth S. STEPHENS, Milwaukee: ASQC Quality Press, 1995, xiii + 73 pp., $21 (softcover).

    This small (73 pages) booklet is a revision of Volume 2 of the se- ries The ASQC References in Quality Control: Statistical Techniques. It consists of two chapters and four appendixes. Chapter 1 is a clear de- scription of the basics of acceptance-sampling plans, how they fit into today's quality-control environment, and concepts associated with accep- tance sampling, such as operating characteristic (OC) curves, AOQ, AOQL, and LQL. Chapter 2 discusses the differences between lot-by-lot and con- tinuous sampling, the latter, which is the topic of this book, being for situations "where it is neither convenient nor practical to group product articles in collective lots or batches for the purposes of inspection" (p. 13). This chapter thoroughly covers the basic continuous-sampling plan, CSP- 1, including its history and references, requirements and indications for usage, a flowchart-style implementation diagram, formulas for five com- mon measures of performance, and nomograms and tables for determin- ing appropriate plan parameters to achieve specified performance. Three other sampling plans, which are slight variations of CSP-1, are covered in nearly as much detail. Five other variations on the basic plans, including multilevel plans and those employing tightened and reduced sampling, are discussed in much less detail, although an implementation diagram and set of references is included for each.

    The ASQC Reference Series has as its aim the goal of surveying topics in a practically usable "how-to" form. The foreword for this particular volume further states that the booklet can be used as a "standard techni- cian's manual as well as a standard classroom and course texi and research reference in the area of continuous sampling." Although I agree that this volume would be excellent for a technician who wanted to know how to implement a plan, the when to and why to questions are not answered well enough for the booklet to stand alone as a course textbook. For example a five-step set of instructions and a numerical example are provided for finding the "best" sampling plan parameters for a given AOQL for the CSP-1 plan, but what is meant by "best" is never defined. There is also an uneven level of mathematical expertise assumed for the various parts of the book. Chapter 1 should be accessible to anyone with a basic knowledge of statistics, but Chapter 2 is dense reading because of its use of technical terminology. Appendix A, which provides the derivations for the perfor- mance measure formulas for three of the samplings using a Markov-chain model, assumes a more advanced statistical background.

    This new edition contains a much enlarged and updated reference sec- tion. Nearly 30% of the approximately 170 references are new since the previous (1979) edition. More annotation on the contents of the papers would have been useful. For example, the author comments, after a partic- ularly lengthy list of references on measures of evaluation, that the papers range from "pathological, extreme scenario situations to very practical and informative conditions that shed light on the operation of CSP schemes" (p. 11). It would have been nice to have some guidance on which was which. The new edition also has an appendix that was added to resolve a controversy over an apparent contradiction in some of Dodge's early papers having to do with plan variations in which defective units were removed rather than being replaced.

    The original nomograph from Dodge's articles (1943, 1947) for two of the early sampling plans are reproduced in the booklet, along with several other easier to use nomographs and helpful examples illustrating their use. Tables for choosing parameters for all plans discussed in the booklet are not included, however, and in fact, the author instructs that "Principal tables are MIL-STD-1234(ORD), MIL-STD-1235A, MIL-STD- 1235B, and now MIL-STD-1235C, and should be procured for use with this booklet" (p. 34).

    Finally, one of the most useful aspects of the booklet is a summary table that lists a brief description of each sampling plan and its parameters.

    S. Lynne STOKES University of Texas

    REFERENCES

    Dodge, H. F. (1943), "A Sampling Inspection Plan for Continuous Produc- tion," The Annals of Mathematical Statistics, 14, 127-133.

    (1947), "Sampling Plans for Continuous Production," Industrial Quality Control, 4, 5-9.

    How to Perform Skip-Lot and Chain Sampling (2nd Ed.), by Kenneth S. STEPHENS, Milwaukee: ASQC Quality Press, 1995, xiii + 64 pp.

    This book is Volume 4 of The ASQC Basic References in Quality Con- trol: Statistical Techniques. Stephens is also the author of Volume 2, How to Perform Continuous Sampling. The reader of Volume 4 will quickly note that skip-lot sampling grew out of Harold Dodge's development of continuous-sampling plans.

    The first edition of this volume contained explanatory material, tables, and curves for skip-lot and chain-sampling plans. This second edition in- cludes an updated reference section, a description of An Attribute Skip-Lot Sampling Program (ANSI-ASQC S1 1987), and a new appendix "to re- solve a controversy over an apparent anomaly in several of Dodge's earlier papers" (p. xi).

    This volume contains sufficient material to enable a user to learn how to use skip-lot and chain sampling. Because in the case of skip-lot sampling a user typically superimposes a set of skipping rules over a referenced plan from, for example, MIL-STD-105E (1989) or ANSI/ASQC Z1.4 (1981), copies of documents such as these should be available to the practitioner. In particular ANSI/ASQC S-1 (1987) is a method for using skip-lot sampling in conjunction with ANSI/ASQC Z1.4 (1981).

    Chain sampling is especially useful when cost or other factors dictate small sample sizes, making cumulation of results of samples from sev- eral lots the only reasonable option for determining acceptability. The chain-sampling plans developed to handle situations such as this are more complicated, naturally, than a single-sampling plan.

    The mathematical content of this book is sufficient to use the material contained therein. Those wishing to delve more deeply can consult the extensive list of references.

    The book provides some history of skip-lot sampling and chain sam- pling. Regarding skip-lot sampling, although the book cites references by A. Endres and C. E. Stine, each of whom worked for me at the time he wrote his reference papers, the book does not mention that the U.S. Army was apparently the first to superimpose skipping rules over a referenced sampling plan. This began in the early sixties.

    Richard M. BRUGGER Consultant

    REFERENCES

    ANSI/ASQC Standard S1 (1987), An Attribute Skip-Lot Sampling Pro- gram, Milwaukee, WI: American Society for Quality Control.

    ANSI/ASQC Standard Z1.4 (1981), Sampling Procedures and Tables for Inspection by Attributes, Milwaukee, WI: American Society for Quality Control.

    MIL-STD-105E (1989), Sampling Procedures and Tables for Inspection by Attributes, Washington, DC: U.S. Department of Defense.

    Fundamentals Concepts in the Design of Experi- ments, by Charles R. HICKS, New York: Saunders Col- lege Publishing, 1993, xii + 509 pp., $66.50.

    This is the fourth edition (first edition was 1964) of a book that has been one of my standard references since graduate school in the mid-70s. The audience is practitioners in the field with some background in statistical inference. As a textbook, it would be appropriate for senior undergradu- ate or first-year graduate students. It does not use matrix notation, so the prerequisites are algebra and familiarity with statistical inference, with fa- miliarity and understanding of analysis of variance (ANOVA) tables being especially useful. The text comprises an introduction; review of statistical inference; single-factor (variable) experiments, with and without restric- tions on randomization; factorial experiments including nested, 2k, 3k, and fractional factorials; regression; and several miscellaneous topics.

    The topics that Hicks covers better than most include nested designs; random, fixed or mixed models; and generating the correct expected mean squares and formulating the correct F tests for various designs. I have always appreciated that Hicks introduces both nested designs and random

    TECHNOMETRICS, FEBRUARY 1996, VOL. 38, NO. 1

    82

    This content downloaded from 131.155.2.68 on Wed, 9 Jul 2014 01:25:07 AMAll use subject to JSTOR Terms and Conditions

    http://www.jstor.org/page/info/about/policies/terms.jsp

    Article Contentsp. 82

    Issue Table of ContentsTechnometrics, Vol. 38, No. 1 (Feb., 1996), pp. 1-94Front Matter [pp. 74-74]Bayesian Analysis of Ordered Categorical Data from Industrial Experiments [pp. 1-10]Empirical Bayes Estimation of the Reliability of Nuclear-Power-Plant Emergency Diesel Generators [pp. 11-24]Prediction in Repeated-Measures Models with Engineering Applications [pp. 25-36]Multivariate Process Analysis with Lattice Data [pp. 37-49]Application of Nonparametric Binary Regression to Evaluate the Sensitivity of Explosives [pp. 50-58]Standard-Error Estimates for Rates of Change from Indirect Measurements [pp. 59-68]An Algorithmic Approach to Constructing Supersaturated Designs [pp. 69-73]Book ReviewsReview: untitled [pp. 75]Review: untitled [pp. 75-76]Review: untitled [pp. 76-77]Review: untitled [pp. 77-78]Review: untitled [pp. 78-79]Review: untitled [pp. 79-80]Review: untitled [pp. 80]Review: untitled [pp. 80-81]Review: untitled [pp. 81]Review: untitled [pp. 82]Review: untitled [pp. 82]Review: untitled [pp. 82-84]Review: untitled [pp. 84]Review: untitled [pp. 84]Review: untitled [pp. 84-85]Review: untitled [pp. 85-86]Review: untitled [pp. 86]Editor Reports on New Editions, Proceedings, Collections, and Other BooksReview: untitled [pp. 86-87]Review: untitled [pp. 87]Review: untitled [pp. 87]Review: untitled [pp. 87]Review: untitled [pp. 87-88]Review: untitled [pp. 88]Review: untitled [pp. 89]Review: untitled [pp. 89]Review: untitled [pp. 89-90]Review: untitled [pp. 90]Review: untitled [pp. 90-91]

    Letters to the EditorComment on Zhou and Mathew (1994) [pp. 92-93]

    Back Matter [pp. 94-94]