Vol. 9, No. 4

April 2007

PQ Systems

Tidying up the control chart

Quality Quiz: With a video!

MSA with Jackie Graham

Data in everyday life

Bytes and pieces

FYI: Current releases


Send Quality eLine
to a friend!

Just type in your friend's email below:


Sign up
If you received this newsletter from a friend and want your own subscription to Quality eLine, click below.

Subscribe to Quality eLine




Tidying up the control chart:
Put away that eraser!

As tempting as it may be to simply erase out-of-control points on control charts and recalculate the limits, giving in to that temptation is like taking a mulligan in golf—and while mulligans and “do-overs” may be acceptable for kids, they don’t belong in the repertoire of the quality professional.

An “oh-well” approach to out-of-control points, erasing them either physically or mentally from consciousness, means that learning is not taking place. “Analyzing each point that falls outside control limits represents an opportunity to learn a great deal about a process,” says PQ Systems technical support manager Matt Savage, adding that whatever is responsible for the out-of-control point may recur if it is not addressed the first time.

Addressing any out-of-control situation involves an improvement in the process, whether the data point has ensued from special cause or common cause variation. Discovering, for example, that data points fall outside control limits because the process is unstable represents an opportunity to give attention to the causes of instability. When a single point jumps because of a special cause—the electricity has gone off in the plant, for example—it is helpful to see the implications of that cause for the process in the future.

The following chart reflects out-of-control points, with annotation to explain these points. The second chart demonstrates the same data, but with the out-of-control points removed. One can see immediately that chart #2 does not accurately reflect the process, even though it may be a much more attractive, tidy chart.

Chart 1

Chart 2

Comparing these charts, one can see that control limits are tighter if data has been removed (chart 2). Undoubtedly, this presages more out-of-control points in the future, since the data has been artificially manipulated. Control limits are, after all, determined by the data itself, and understanding the process demands authentic data.

One can create a lovely control chart, of course, by making up all the data, eliminating data points that create any challenge to the control limits that one desires to set, and producing a text-book perfect example of a control chart.

In that case, the control chart becomes a work of fiction—just like taking a mulligan.


Copyright 2007 PQ Systems.
Please direct questions or problems regarding this web site to the Webmaster.