genetic algorithms and engineering optimization

2
This article was downloaded by: [Central Michigan University] On: 25 November 2014, At: 10:06 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Technometrics Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/utch20 Genetic Algorithms and Engineering Optimization Eric Ziegel a a BP Published online: 01 Jan 2012. To cite this article: Eric Ziegel (2002) Genetic Algorithms and Engineering Optimization, Technometrics, 44:1, 95-95, DOI: 10.1198/tech.2002.s675 To link to this article: http://dx.doi.org/10.1198/tech.2002.s675 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the 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 relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

Upload: eric

Post on 24-Mar-2017

216 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Genetic Algorithms and Engineering Optimization

This article was downloaded by: [Central Michigan University]On: 25 November 2014, At: 10:06Publisher: 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

Genetic Algorithms and Engineering OptimizationEric Ziegelaa BPPublished online: 01 Jan 2012.

To cite this article: Eric Ziegel (2002) Genetic Algorithms and Engineering Optimization, Technometrics, 44:1, 95-95,DOI: 10.1198/tech.2002.s675

To link to this article: http://dx.doi.org/10.1198/tech.2002.s675

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 purposeof the Content. Any opinions and views expressed in this publication are the opinions and views of theauthors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content shouldnot be relied upon and should be independently verified with primary sources of information. Taylorand Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses,damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connectionwith, in relation to or arising 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

Page 2: Genetic Algorithms and Engineering Optimization

BOOK REVIEWS 95

These chapters are heavily integrated with the use of SAS. All SAS code isprovided. There is a lot of discussion about using logistic regression. Jack-kni� ng and bootstrapping are used and are illustrated for validating models.

Using the most recent data mining book, Berry and Linhoff (2000), reportedby Ziegel (in press) as a de� nitive presentation, this book, with its focuson logistic regression, is really more about the data modeling process. Itdoes this part very nicely, but the vast majority of the methodology for datamining is left unexplored in it pages. A better title would have been “DataModeling Cookbook,” and data mining could have replaced data modeling inthe subtitle.

Eric Ziegel

BP

REFERENCES

Berry, M., and Linhoff, G. (2000), Mastering Data Mining, New York: Wiley.Hosmer, D., and Lemeshow, S. (1989), Applied Logistic Regression,

New York: Wiley Interscience.Tukey (1977), Exploring Data Analysis, Reading, MA: Addison–Wesley.Ziegel, E. (in press), Editor’s Report for Mastering Data Mining, by M. Berry

and G. Linhoff, Technometrics, 43.

Genetic Algorithms and Engineering Optimization, byMitsuo Gen and Runwei Cheng, New York: Wiley, 2000,ISBN 0-471-31531-1, xviC 495 pp., $99.00.

Industrial statisticians must have a familiarity with genetic algorithms.There are a number of papers that have appeared in the literature for appliedchemistry, such as Kowar (1998), which claim that experiments designedby genetic algorithms are more ef� cient for � nding optimum chemicalformulations than experiments designed using response surface methodology,such as those that might be selected using the methods in Myers andMontgomery (1995). Trocine and Malone (2000) discussed this topic in apaper presented at the American Statistical Association Annual Meeting for2000.

This book probably offers more material than any statistician would everwant or need to know about genetic algorithms. It also does not discuss theapplication for the design of experiments. The � rst chapter, “Foundations ofGenetic Algorithms,” and the second chapter, “Combinatorial OptimizationProblems,” provide the necessary background and the illustration from a sim-ilar application to show how genetic algorithms can offer a different approachand an alternative solution to the problem of sequential optimization throughdesigned experiments. The third chapter, “Multiobjective Optimization Prob-lems,” and the fourth chapter, “Fuzzy Optimization Problems,” provide theapproach to the multiresponse optimization problem that is described in anumber of applied statistic books, such as Khuri and Cornell (1996).

The rest of the book is devoted to various engineering applications. Theseare listed below:

Reliability design problemsScheduling problemsAdvanced transportation problemsNetwork design and routingManufacturing cell design

Eric Ziegel

BP

REFERENCES

Khuri, A., and Cornell, J. (1996), Response Surfaces, New York: MarcelDekker.

Myers, R., and Montgomery, D. (1995), Optimization Using Designed Exper-iments, New York: Wiley.

Trocine, L., and Malone, L. (2000), “Experimental Designs and HeuristicSearch Methods,” in ASA Proceedings of Section on Physical and Engi-neering Sciences, pp. 136–141.

Kowar, T. (1998), “Genetic Function Approximation Experimental Design,”Journal Chemical Information and Computer Sciences, 38, 858–866.

Risk Analysis (2nd ed.), by David Vose , West Sussex,U.K.: Wiley, 2000, ISBN 0-471-99765-X, x C 418 pp.,$100.00.

In my report on the � rst edition (Ziegel 1999), noting my BP colleague’sforthcoming book on risk analysis (Koller 1999), I concluded that the Vosebook “would be an excellent supplement to provide greater detail for thequantitative part of the methodology.” My colleague has written a second book(Koller 2000), but in that book he has emphasized examples, not methods. TheVose second edition (2E) is still a � ne complementary book. As I noted in myreport (ibid.), “this is a very complete background book for quantitative riskanalysis.” My comment remains very appropriate. The author has considerablyexpanded the coverage of the technical basis for quantitative risk assessmentin the 2E.

Originally the book bore the title “Quantitative Risk Analysis,” and thesubtitle was “A Guide to Monte Carlo Simulation Modeling.” Here the titlehas a signi� cant simpli� cation, and the subtitle is now “A Quantitative Guide.”As the author has noted in the preface (p. ix), he has “taken a more academicapproach in developing and explaining far more fully the techniques of riskanalysis.” In the year 2001 that is an unusual direction for a second editionof a statistics-related book.

The book has been considerably restructured and reformatted. It is now alarge-format book. Despite that change, the book is 100 pages longer withthe addition of new sections to several chapters and even whole new chapters.There is a new chapter on stochastic processes which includes advanced topicslike renewal theory and mixtures distributions. There is another new chapter,“Quantifying Uncertainty about Model Parameters,” which includes sectionson Bayesian inference, bootstrapping, and maximum entropy. A section onnon-parametric distributions was added to the chapter on determining distri-butions. A section on variable-rate Poisson processes was added to the chapteron time series projections. There are several additional new sections in therisk modeling chapters. A new application chapter was added at the end ofthe book with the title “Animal Import and Food Safety Risk Assessment,” avery relevant topic, with the incidences of animal diseases in Europe.

See Ziegel (1999) for a more complete summary of the book. As this is abook that is focused on the academic side of the risk analysis methodology,it does not include any discussion of the use of the various commerciallyavailable risk analysis software packages.

Eric R. Ziegel

BP

REFERENCES

Koller, G. (1999), Risk Assessment and Decision Making in Business andIndustry: A Practical Guide, Boca Raton, FL: CRC Press.

Koller, G. (2000), Risk Modeling for Determining Value and Decision Making,Boca Raton, FL: Chapman & Hall/CRC.

Ziegel, E. (1999), Editor’s Report on Quantitative Risk Analysis, by DavidVose, Technometrics, 41, 381.

Biostatistical Methods, by John M. Lachin, New York:Wiley, 2000, ISBN 0-471-36996-9, xviiC529 pp., $89.95.

Subtitled “The Assessment of Relative Risks,” the book has a number offeatures that separate it from some of the other biostatistics textbooks thathave been reported here; e.g., most recently, Ziegel (2000) reported the bookby Selvin (1998). First, this is not an introductory book. The author indicatesthat it grew out of the 600 level biostatistics course at George WashingtonUniversity. Second, the author is not a “mathematical statistician” (Preface,p. xvi). He initiated the graduate program in biostatistics from his directorshipof the Biostatistics Center within the School of Public Health. His uniqueperspective is stated in the Preface (p. xv): “Biostatistics is set apart fromother statistical specialties by its focus on the assessment of risks and rela-tive risks through clinical research.” Following his introductory chapter, theauthor offers as the second chapter “Relative Risk Estimates and Tests forTwo Independent Groups.”

Despite the author’s disavowal of training as a mathematical statistician,that issue would not be readily apparent in the textbook that he has written.This is not a book for anyone who is not seeking a high-level and theoreticalapproach to biostatistics. The book basically has two parts. The � rst half uses

TECHNOMETRICS, FEBRUARY 2002, VOL. 44, NO. 1

Dow

nloa

ded

by [

Cen

tral

Mic

higa

n U

nive

rsity

] at

10:

06 2

5 N

ovem

ber

2014