This article was downloaded by: [Carnegie Mellon University]On: 09 November 2014, At: 03:32Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

TechnometricsPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/utch20

Problem Solving: A Statistician's GuidePublished online: 12 Mar 2012.

To cite this article: (1996) Problem Solving: A Statistician's Guide, Technometrics, 38:4, 411-411, DOI:10.1080/00401706.1996.10484570

To link to this article: http://dx.doi.org/10.1080/00401706.1996.10484570

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose ofthe Content. Any opinions and views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be reliedupon and should be independently verified with primary sources of information. Taylor and Francis shallnot be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and otherliabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

BOOK REVIEWS 411

overview was completely inconsistent with the useful technical detail of the preceding two SPC chapters. The next industry chapter (5) on pulp and paper processing was written by Frank Sinbaldi. It begins in a similar vein but then proceeds to present numerous control-chart applications from all aspects of the industry and a long list of reasonably current references. The chapter on the steel industry (6) by James Lynn, though not so technically strong. nevertheless is effective again, both with an industry description and the SPC applications. The textile-industry chapter (7) by Roy Johnson is another weak and brief overview, even excluding any references and also having no illustrations with data. The food-industry chapter (8) by John Surak has a reference list but is otherwise no better. A final chapter (9). written by the editor, on discrete parts, offers five pages where often whole books are provided.

This book appears to be a retirement project by the editor. Much of it is out of date, some of it is out of place, and the editor’s decision generally to rely on old friends rather than persons with current technical expertise in the industry vitiated the value of almost all of the industry material. The book would be a bad value at half the price.

REFERENCES

Ford Motor Company (1984), Continuing Process Control and Process Capability Improvemenr, Dearborn, MI: Statistical Methods Office.

Pruett, J., and Schneider, H. (1993), Essentials of SPC in the Process In- dustries, Research Triangle Park, NC: Instrument Society of America.

Ziegel, E. (1994), Review of Essentials of SPC in the Process Industries, by J. Pruett and H. Schneider, Technometrics. 36, 421-422.

Bayesian Methods in Reliability, edited by P. SANDER and R. BADOUX, Dordrecht, The Netherlands: Kluwer, 1991, x + 221 pp., $145.

This is one among generally infrequent reports on books that escaped the notice of the Editor and were not sent by the publisher. We make every effort to review or report all books judged important to the audience for the journal, though there are books that we request that the publishers never send. The book by Martz and Wailer (1982), which is not familiar to me, is the only other book I know that has been devoted to Bayesian methods for reliability application. The interest in Bayesian methods of a petrochemical industry cooperative group for reliability caused me to do a thorough literature search, and I found this book.

Reliability engineers will not find this book very satisfying for show- ing them what Bayesian statistical methods are and how they are used in reliability. The book has only six chapters, which have vastly different lengths, contrasting styles, and no directional flow. The book also is gen- erally focused on mathematics, has minimal applications, makes no use of statistics software, and has nearly one-third of its length, the final chapter, devoted to a special topic, forecasting software reliability. Surprisingly, the Introduction (p. ix) indicates that this is “the core material of a course” organized by several European reliability organizations. One would expect more cohesion and directional intent in this situation.

For statisticians with a background in classical statistics and reliability, there is some nice material in the first couple of chapters. The first chapter is an introduction to Bayesian decision analysis and its value for reliabil- ity situations. This concept is reiterated and extended in many ways in the overview in the second chapter, also about one-third of the book. Em- phasizing the need to use diverse expertise integrated with usually sparse data, it explores many aspects of Bayesian and reliability statistics, in- cluding probability concepts, reliability concepts, non-Bayesian statistics, decision analysis, beliefs as probabilities, Bayesian philosophy, determina- tion of prior densities, empirical Bayes, and numerical methods. These two chapters were a nice mathematical introduction with excellent motivation.

The middle three chapters are all short. Two briefly discuss reliability for nonrepairable and repairable systems, respectively. The third is titled “The Use of Expert Judgment in Risk Assessment.” None would help an engineer actually use Bayesian methods for reliability applications.

REFERENCE

Martz, H., and Waller, R. (1982), Bayesian Reliability Analysis, New York: John Wiley.

Problem Solving: A Statistician’s Guide (2nd ed.), by C. CHATFIELD, London: Chapman and Hall, 1995, xvi + 325 pp., $36.95 (softcover).

The first edition in 1988 received a generally favorable review from Donnelly (1990). His primary criticism was that it had an inappropriate title because the topics that we generally allocate to problem solving in the United States were not covered. Otherwise, he gave the book good marks. commenting that “if you are interested in a good book on data analysis, then this is your cup of tea” (p. 349). Donnelly emphasized that “the thrust of the data analysis part of the text is on what Chatfield calls the initial examination of data” (p. 349).

Chatfield has increased the number of pages by more than 25%, but he modestly claims (Preface, p. vii) only to have “substantially rewritten and extended the discussion in Part One,” which was labeled, “General Principles Involved in Tackling Real-Life Statistical Problems,” the part of the text that Donnelly liked. It has expanded from 83 pages to 125 pages, which is most of the increase of the size of the book. Accentuating the strengths that were described by Donnelly, about two-thirds of the addition occurs for the three sections “Analyzing the Data.” Chattield maintains a strict separation between discourse (most of Part One) and illustration (here Part Two, labeled “Exercises”).

No use of statistical computing is presented in the illustrations. There remains in addition the rather unique Part Three, labeled “Appendices.” Most of this part is Appendix A, ‘A Digest of Statistical Techniques,” of which there are 15, the basic material from several different statistics courses, all presented compactly in 50 pages. Perhaps this book works as leisure reading, but otherwise I am not sure that this is my “cup of tea.”

REFERENCE

Donnelly, P. (1990), Review of Problem Solving: A Statistician’s Guide, by C. Chatfield, Technometrics, 32, 349-350.

Stochastic Modeling and Geostatistics, edited by J. YARUS and R. CHAMBERS, Tulsa, OK: American As- sociation for Petroleum Geologists, 1994, x + 379 pp., $149.

Book 3 in the association’s series on computer applications in geology, this very handsome volume would be suitable as a coffee-table book. It is beautifully printed on high-gloss 8$ x 1 1 inch pages with typesetting in two columns. There is considerable USC of black and white production graphics, some in almost every paper, generally actual graphical displays of geological characteristics, and even several color graphics pages. This, of course, results in a very high price for the book.

Geostatistics generally has advanced over the past 30 years with little help from statisticians. The two editors are petroleum-company geologists, including one from my own company, who have no particular education or background in statistics, and that statement probably is equally appropri- ate to nearly every one of the 50 or so authors for the 25 papers that have been collected here. In the Preface (p. ix), the authors state, “Feostatisti- cians are geologists, petrophysicists, geophysicists, or even engineers by training.” Because computers handle most of the statistical aspects of geo- statistics, geology types do not need nearly as much understanding of statistics as statisticians must have of geology to be successful practition- ers. The ultimate statistical treatise on spatial statistics (Cressie 1991) is not even listed among the books in the bibliography, “Selected Readings in Statistics.” Value from geostatistics in petroleum geology, however, is not the glory of the methodology but using the tools as part of the pro- cess of finding petroleum reserves. The editors and their contributors have done a fantastic job in showing their professional colleagues the value that geostatistical methods can offer to the petroleum industry.

There are a first chapter overview on stochastic methods for reservoir characterization, the objective for petroleum geologists who use geostatis- tics, and two brief essay chapters in a section “Getting Started.” There are four chapter topics for the second part, “Principles”-semiovariograms, sample support, vertical variability measures, and upscaling procedures. The Introduction (p, 1) notes that the contributors to this section were asked “to minimize detailed mathematical and theoretical discussions,” and in fact those were practically nonexistent here.

TECHNOMETRICS, NOVEMBER 1996, VOL. 38, NO. 4

Dow

nloa

ded

by [

Car

negi

e M

ello

n U

nive

rsity

] at

03:

32 0

9 N

ovem

ber

2014