Heart of the Matter
Abstract:The goal of most processes in the manufacturing and service industries is to produce products or services that exhibit little to no variation. Variation in a process is studied by sampling the process. After a particular variation is determined, it must be analyzed and broken down into its individual components. This article discusses various methods for analyzing variation, with an emphasis on the methods of statistical process control (SPC) as promulgated by Walter A. Shewhart, as well as the insights of W. Edwards Deming as elaborated in his philosophy of management. The authors conclude that variation should be considered a strategic management issue because it is at the heart of …
Social Bookmarking
Digg, delicious, NewsVine, Furl, Google, StumbleUpon, BlogMarks, Facebook
You must log in to comment. Please log in or register. Comments FAQ

Excellent article that reinforces the foundations of quality that should be common knowledge throughout the world.
In attempts to broaden application of this knowledge beyond the quality profession, see my recent articles at the Process Excellence Network (PEX), “Putting Deming’s principles into action to transform individuals, communities, and organizations.”
--Tim Clark, 04-15-2013

Excellent article. Clear and understandable. Thanks.
--Robert Wright, 04-09-2013
Average Rating

Out of 3 Ratings
Rate this article
Related Articles
ASQ World Conference on Quality and Improvement Sponsors and Exhibitors
Nifty Fifty
Forward Progress
Featured advertisers
Great article overall, and much needed these days. However, there is one point with which I take issue (and have taken issue with Bill Wortman on it, as well): the statement "A histogram provides 'a deeper understanding of common and special causes.' When analyzing a histogram, variation inside the bell curve is due to chance or natural variation, while other variation is due to special or assignable causes."
This is incorrect. A histogram is a snapshot in time that tells us nothing about special or common causes. It's entirely possible to have a process shifting or trending detectably on a control chart, but if you pile the data in a histogram they will look like a coherent bell curve and may not fail a non-normality test. Similarly, a histogram may show that a process is skewed. If it's skewed, it may still be in a state of statistical control (because that's the way that process works), but then it doesn't meet the "inside the bell curve" requirement.
--Rip Stauffer, 12-11-2013