Reducing the Number of SPC Charts used: We have many highly capable processes


Daniel J. Rupright

I work in the quality department for a mid to large company. We have plants in Ohio, South Carolina, Arkansas, and recently Pennsylania. In our Ohio plant's SPC department, we have over 1,171 active floor charts. Some have CPK's over 20. Is a guideline for reduction? Our customers' SPC dimensions would still be monitored on our inspection instruction cards done hourly. Does anybody know how to beat this paperwork monster?

Daniel J. Rupright
Defiance Metal Products
Defiance, Ohio 43512

Roger Eastin

Are all these SPC charts required by your customer(s)? I think you can stop, IMHO, doing SPC charts on Cpk's as large as 20 (unless required by your customer). You're trying to swat a gnat with a sludge hammer! Use your money somewhere else. You need to look at your strategy for doing control charts. I would only do them on your worst processes for now (e.g.,Cpk<1.33) and save some trees by not doing all the rest. You may want to audit processes where you consider the cost of nonconforming product too high (internal/external repair costs, etc.), but save your money for those areas that really need it.

Al Dyer

Boy, what a logistical nightmare!

I'd like to add this thought:

If your Cpk's are that high, is there a possibility that you can review your control plans (customer involvement) with the thought of reducing your costs and passing SOME of that savings to the customer?

Kill two birds with one stone.


Al Dyer
Mngt. Rep.
[email protected]

Tommy VanHorne

You mention high Cpk's. Do you have any quality cost data (i.e. internal or external rejections) that support the supposition that "meeting specificaton equates with low rejections".


We are in a similar situation, 1000+ charts in the shop. Currently trying to take SPC off all CPK>10 (for a start!). Have you thiught of gauging the critical dimensions instead, customers tend to go for it if you can prove capability.

In the past we have conducted ppk on several dimensions, even if they are not high lighted as critical. Take the lowest and monitor for a while.

As mentioned above, why waste time and money on very little.


For processes with Cpks that high, is it possible to economize, to cut back on materials, or save time by purposefully and controllably reducing Cpk? DOE comes to mind as a tool, but at >20, you could probably achieve a lot by simple one-factor experiments.

The other thing; is automation feasible? That's a surefire way to eliminate SPC charts.
Top Bottom