Practical Effective Measurement System Analysis

Article

Berg, Douglas L.   (1988, ASQC)   Hydra-matic Division of General Motors, Ypsilanti, MI

Annual Quality Congress, Dallas TX    Vol. 42    No. 0
QICID: 3523    May 1988    pp. 822-827
List $10.00
Member $5.00

FOR A LIMITED TIME, ACCESS TO THIS CONTENT IS FREE!
You will need to be signed in.
New to ASQ? Register here.

Article Abstract

With a renewed interest in SPC, concerns have been expressed about the adequacy of the measurement systems employed and the variation they induce in the readings. Traditional evaluations of measurement systems, whether instrumentation, gages, templates, or whatever, have revolved around calibration and traceability. As such, the evaluations have addressed accuracy and linearity, but not variation over time, reading to reading, or system to system. Pioneering work by Mandel in repeatability and reproducibility and subsequent efforts by others are slowly gaining acknowledgement. Putting these concepts in use, however, has caused difficulties for many practitioners. This paper attempts to overcome some of these difficulties by focusing not on the hardware, but rather on the measurement system or measurement process. This goes beyond the gage or other hardware and considers the part, operator, procedure, and environment. It requires a "robust" assessment that reflects actual use. Some simple guidelines focused on the application are proposed to obtain a robust assessment. A judicious selection of the parts can allow the evaluation of the system for accuracy, linearity, stability, repeatability, and reproducibility with greater economy, effectiveness and reduced sample sizes. "Break-in" or "wear" affecting repeat readings can be handled if studied and understood.

The actual evaluation does not require extensive computer resources and a case can be made against using one. Emphasis should be placed on plotting the data to reveal time, operator, machine and part effects as well as accuracy or linearity. Residual analysis is also important. Formal "statistical" analysis utilizing regression and analysis of variance should only follow the graphical review.

Keywords

Gauges,Inspection,Statistics,Gage Repeatability and reproducibility studies (GR&R)


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.