Quality Progress - June 2015 - 52
BY ROGER W. HOERL AND
RONALD D. SNEE
Using statistical engineering principles for problem solving
QUALITY PROFESSIONALS are often
to generate improved results."1 Applica-
involved and political considerations at
faced with solving major organizational
tions of this discipline produce improved
play. Too often, data analysis begins with
problems such as: "Customers are com-
results because statistical engineering is
the data. This can be seen in various data
plaining about the quality of our product
grounded on sound underlying principles
analysis competitions, such as those on
and returning it"; "Our major process is
that address the critical elements of ef-
kaggle.com. Keep in mind, however, that
producing an unacceptable amount of
fective problem solving. In short, the key
the data are not the problem; the problem
defective product"; or "The regulatory
elements of the principles of statistical
is the problem. That is, we should view
agency has identified a major environmen-
engineering are (see Figure 1):
data as a "how," and the original problem
tal problem associated with one of our
* Proper understanding of the problem
trying to be solved as the "what."
How should quality professionals approach such problems, which are clearly
not textbook with one correct answer?
Where should they begin the problemsolving effort? What should be considered? How can the projects be set up for
success? The fundamentals of statistical
* A well-defined strategy for problem
* Evaluation of the pedigree of the associated data and information.
* Integration of sound subject matter
knowledge with data analysis.
* Sequential approaches involving the
Once we are clear on the problem we
are trying to solve and its context, we
can determine the type and amount of
data needed to solve it. Conversely, if
we already have data, clarification of the
problem helps determine how the data can
be best used to solve the problem. "Data
have no meaning in themselves; they are
engineering can provide valuable guidance
testing of existing hypotheses and
meaningful only in relation to a conceptual
for these types of complex problems.
development of new hypotheses.
model of the phenomenon studied,"2 wrote
Statistical engineering has been defined
George Box, Bill Hunter and Stu Hunter.
as: "The study of how to best utilize
Understanding problem context
statistical concepts, methods and tools,
Problem context is everything we know
business solution is not always the best
and integrate them with information
about the problem, including its history,
statistical solution. For example, we may
technology and other relevant sciences
what has been tried before, the technology
determine from initial analysis of exist-
In addition, the best technical or
ing data that they are not appropriate or
sufficient for solving the problem at hand;
additional, better quality data are needed.
Performing sophisticated or detailed
analysis of the current data would simply
waste time at this point.
In other cases, a simple analysis is all
that is needed because the answer is obvious from basic graphs. The bottom line is
that the context of the problem, not statistical metrics, determines the best business
solution and the level of sophistication
Some practitioners have a favorite tool,
whether it is multiple regression, time-series analysis or a nonparametric method.
52 QP * www.qualityprogress.com
Table of Contents for the Digital Edition of Quality Progress - June 2015
Mr. Pareto Head
What’s Your Next Move?
Assessing the Landscape
Change in Flow
Quality in the First Person
One Good Idea
Back to Basics
Quality Progress - June 2015