Optimal Use of Test Information

Article

Schleusener, Richard D.; Baylis, Jim H.   (1990, ASQC)   Eastman Kodak Co., Windsor, CO

Annual Quality Congress, San Francisco, CA    Vol. 44    No. 0
QICID: 9477    May 1990    pp. 360-367
List $10.00
Member $5.00

This article is not available online. Contact us to receive a scan of the archive, in PDF format.
New to ASQ? REGISTER HERE.

Article Abstract

A common problem in any manufacturing process is deciding what to do with test information. Do you believe the test result? Should you react and adjust your process based on the test result? By how much should you react? Is your ability to make a correction good enough to justify making the correction? This paper outlines a method that accounts for the size of incoming process variability and test variability, and then gives the size of the correction to make to the process that will yield the minimum outgoing variability for that process. The method estimates the variability of corrections made and estimates the projected improvement in outgoing process variability. It will be shown that the alpharain correction procedure is very robust; the correction procedure yields improvements to the outgoing variability even with rough estimates of the adjustment factor. In addition, the alpharain procedure quantifies the impact of making corrections where the ability to correct is not very good. If by adjusting the process, more variability is introduced than was in the process before the adjustment, the best course would be to make no correction at all. This method determines the impact on the outgoing variability taking into account the amount of variability introduced by the correction process. An additional benefit of this technique is in helping a manufacturing or testing organization determine how good test variability needs to be. Often arbitrary limits are set on the level of testing variability allowed because there is not a standard way to dictate what the variability of a test needs to be. Alpharain gives an objective approach to setting limits around testing variability. This paper demonstrates how to estimate the components of variability described above, and how to use those estimates to react to test data from any process. Alphamin does not involve a sophisticated algorithm to produce those estimates. This approach has potential for wide application, and is easy to implement and use.

Keywords

Chemical and process industries


Browse QIC Articles Chronologically:     Previous Article     Next Article

New Search

Featured advertisers





ASQ is a global community of people passionate about quality, who use the tools, their ideas and expertise to make our world work better. ASQ: The Global Voice of Quality.