|
|
Exploring traditional, innovative, and revolutionary issues in the minerals,
metals, and materials fields. |
VISIT THE
JOM COVER GALLERY
JOM HOME PAGE
TABLES OF CONTENTS
HTML-ENHANCED ARTICLES
MATERIAL MATTERS
MATERIALS RESOURCE
CENTER
JOM TECHNICAL DIRECTORY
TMS WEBCAST HOME
PROFESSIONAL PREFACE
SUBJECT INDEXES
TECHNICAL EMPHASIS
CALENDAR
AUTHOR'S KIT
ABSTRACT SUBMISSION FORM
BOOK REVIEWS
ADVERTISING INFORMATION
RESERVE A CLASSIFIED
SUBSCRIPTIONS
SINGLE ISSUE PURCHASES
LETTER TO EDITOR
|
|
05/6/2004 -
Statistics and Probability for Engineering Applications (2003)
by W.J. DeCoursey
ISBN 0-7506-7618-3. Newnes, Amsterdam. 2003. Softcover & CD-ROM. 396 pages. $59.99.
| REVIEWED BY: | R. Sundaresan, International Advanced Research Centre for Powder Metallurgy and New Materials
|
The book is aimed at students, particularly engineering students, and practicing engineers. The general scheme of the book consists of the description of the topics, generously interspersed with worked out examples and problems. At appropriate points, examples and problems take up application of Excel worksheet. The first three chapters define terms and introduce basic concepts of probability, variability, and related statistical functions. Grouped frequencies and graphical descriptions are introduced in the fourth chapter.
The next two chapters dwell at length on probability distributions, first of discrete variables, then of continuous variables. The ultimate aim of statistics as a tool for the engineer is most often the prediction of reliability. This calls for defining probabilities, and hence, the prediction of a random variable. This is realized through probability functions and cumulative distribution functions. These take the form of either binomial or Poisson distribution in the case of discrete variables. When the variable is continuous, the distribution takes the form of a probability density function. These can be t-, F-, and chi-squared distributions, in simple uniform or exponential functions or more complex Weibull, beta, or gamma distributions. In these computations, Excel can be directly used with its many useful statistical functions.
Normal distribution is the most important probability distribution function. Transformation of variables to yield a normal distribution is covered in the next chapter. Sampling and combination of variable are taken up in the eighth chapter, while statistical inferences for the mean, variance, and proportion are dealt with in the next two chapters.
The first ten chapters have been written with the primary aim of teaching statistics to the layman. Major applications of statistics are taken up subsequently. Chapter 11 introduces design of experiments. Design of experiments is of importance to industrial experiments where interfering factors may have considerable influence on the outcome. Since in normal operation more than one governing factor is changing, perhaps randomly, evaluation of the effects of these factors is difficult. Planned experiments at a lower scale can be employed in a series of small changes to the operating conditions. These changes can be one factor at a time or of a factorial design. Examples in the chapter illustrate the immensity of factorial design when the factors involved increase and the need for experimental design that is evolutionary is brought out. Minimizing bias by randomization (using Excel to obtain random numbers!) and by blocking are detailed with examples. In the next chapter, the concept of analysis of variance is introduced to enable analysis of data obtained from the factorial experiments designed. The author makes it clear that such strategies employed in industrial experiments are just introduced here and would need further study for full understanding and application.
The final chapter is a list of reference books for further study. The strength and application area of each book is highlighted briefly. There are useful appendixes of statistics tables of normal probability; t-, chi-squared, and f-distributions; and notes on Excel.
The book comes with a CD that contains the complete book in portable document format, spreadsheets for direct use on Excel with many of the problems cited in the book, and a useful statistics software package, Dataplot, maintained by the National Institute of Standards and Technology (NIST). The reviewer was not able to access the homepage of NIST for detailed documentation, but the program was eminently useful as a single program covering the various statistical computation and plotting requirements.
The book fills a major requirement in today’s engineering. Manufacturing and quality control employ statistics, and engineers who are laymen to statistics are required to use statistics tools. Clearly the need for understanding basic statistics is minimal in such cases, but the need to appreciate the methodology is immense. This book, with its chapters written progressively starting from basic definitions in such a way for understanding the application of statistics clearly, fulfils that need extremely well. The worked out examples and the problems set for each of the sections are down-to-earth, everyday problems. The book achieves the objective of teaching the use of Excel spreadsheets for solving such problems. This volume would thus be ideal for all undergraduate students of engineering and a handy reference book for engineers in the area of manufacturing and quality control.
|
| | The Minerals, Metals and Materials Society's JOM Book Review Program pages are
maintained with News Update 1.1 |
|
|
