Shewhart, Deming, and Six Sigma

QualityGuru — Tue, 15 Sep 2009 07:27:40 +0000

Another good article by Donald J. Wheeler
This article has focus on selected aspects of the work of Shewhart and Deming and how these compare with a common element of various six-sigma programs, a look at the concept of an operational definition, then turn to what it takes for improvement.
This will lead to a distinction between observational studies and experimental studies.
An Operational Definition

In the pre-publication drafts of Quality, Productivity, and Competitive Position Dr. Deming wrote:
“An operational definition consists of (1) a criterion to be applied to an object or a group of objects, (2) a test of compliance for the object or group, and (3) a decision rule for interpreting the test results as to whether the object or group is, or is not, in compliance.”

This definition closely parallels Dr. Shewhart’s opening statement for his (1939) book Statistical Method from the Viewpoint of Quality Control:
“Broadly speaking there are three steps in a quality control process: the specification of what is wanted, the production of things to satisfy the specification, and the inspection of the things produced to see if they satisfy the specification.”

This idea of an operational definition, which Shewhart and Deming popularized from the work of the philosopher C. I. Lewis, provided the seed for what grew into the Shewhart or PDSA Cycle.

Plan-Do-Study-Act Cycle does form a powerful framework for any improvement effort, it has often been reduced to a checklist to be followed mechanically. This has led to a proliferation of “expanded” PDSA cycles where each of the steps on the checklist are specified in ever increasing detail.
But before we go down this path, I would like to back up and generalize a bit.

In Dr. Deming’s own conversations, when individuals would start telling him about what they or their organization were planning to do, he would invariably have one of two responses for them:
“By what method?” or “How will you know?”
Either one of these questions would generally end the conversation since the individual would have no answer. After discerning this pattern to Dr. Deming’s responses, it finally occurred to me that these two questions corresponded to the last two parts of an operational definition. This realization, in turn, resulted in a generalization of an operational definition to become:

What do you want to accomplish?
By what method will you accomplish it?
How will you know when you have accomplished it?

Whatever you may be doing, until you can answer all three of these questions, you do not have an operational definition but merely a basis for an argument. Shewhart understood this, and in his work he used the concept of an operational definition in the development of the “operation of statistical control.”

Orginal and full text is on SPC Press online

Quality News » Productivity

Shewhart, Deming, and Six Sigma