e-book
Modern Engineering Statistics
Statistical methods are an important part of the education of any engineering student.
This was formally recognized by the Accreditation Board for Engineering and Technology
(ABET) when, several years ago, education in probability and statistics became an ABET
requirement for all undergraduate engineering majors. Specific topics within the broad field
of probability and statistics were not specified, however, so colleges and universities have
considerable latitude regarding the manner in which they meet the requirement. Similarly,
ABET’s Criteria for Accrediting Engineering Programs, which were to apply to evaluations
during 2001–2002, were not specific regarding the probability and statistics skills that
engineering graduates should possess.
Engineering statistics courses are offered by math and statistics departments, as well
as being taught within engineering departments and schools. An example of the latter is
The School of Industrial and Systems Engineering at Georgia Tech, whose list of course
offerings in applied statistics rivals that of many statistics departments.
Unfortunately, many engineering statistics courses have not differed greatly from mathematical
statistics courses, and this is due in large measure to the manner in which many
engineering statistics textbooks have been written. This textbook makes no pretense of being
a “math stat book.” Instead, my objective has been to motivate an appreciation of statistical
techniques, and to do this as much as possible within the context of engineering, as many
of the datasets that are used in the chapters and chapter exercises are from engineering
sources. I have taught countless engineering statistics courses over a period of two decades
and I have formulated some specific ideas of what I believe should be the content of an
engineering statistics course. The contents of this textbook and the style of writing follow
accordingly.
Statistics books have been moving in a new direction for the past fifteen years, although
books that have beaten a new path have often been overshadowed by the sheer number of
books that are traditional rather than groundbreaking.
The optimum balance between statistical thinking and statistical methodology can certainly
be debated. Hoerl and Snee’s book, Statistical Thinking, which is basically a book
on business statistics, stands at one extreme as a statistics book that emphasizes the “big
picture” and the use of statistical tools in a broad way rather than encumbering the student
with an endless stream of seemingly unrelated methods and formulas.
This book might be viewed as somewhat of an engineering statistics counterpart to the
Hoerl and Snee book, as statistical thinking is emphasized throughout, but there is also a
solid dose of contemporary statistical methodology.
This book has many novel features, including the connection that is frequently made (but
hardly ever illustrated) between hypothesis tests and confidence intervals. This connection
is illustrated in many places, as I believe that the point cannot be overemphasized.
I have also written the book under the assumption that statistical software will be used
(extensively). A somewhat unusual feature of the book is that computing equations are kept
to a minimum, although some have been put in chapter appendixes for readers interested
in seeing them. MINITAB is the most frequently used statistical software for college and
university courses. Minitab, Inc. has been a major software component of the Six Sigma
movement and has made additions to the MINITAB software to provide the necessary capabilities
for Six Sigma work. Such work has much in common with the field of engineering
statistics and with the way that many engineers use statistics. Therefore, MINITAB is heavily
relied on in this book for illustrating various statistical analyses, although JMP from
SAS Institute, Inc. is also used.
Tidak ada salinan data
Tidak tersedia versi lain