Gage Linearity

The gage linearity calculation is an indication of whether the bias error associated with the gage is constant throughout the operating range of the gage. In other words, does the gage have the same accuracy when measuring large parts as when measuring small parts? Gage linearity can be calculated using the below method, which is easily done in our SPC software.

The AIAG (Automotive Industry Action Group; ref. MSA Reference Manual, Feb 1995) method for the gage linearity calculation is as follows:

1. Choose two or more parts throughout the operating range of the gage. Measure the pertinent characteristic using high accuracy gages, such as used in Layout Inspections, to obtain a Reference Value.

2. Using the gage to be studied, have one or more Appraisers measure the part characteristic multiple times. Calculate the average of these measurements for each part.

3. Calculate the Bias for each part as the difference between the Appraiser average and the Reference Value.

4. Using a Scatter Diagram, plot the Reference Value (x-axis) vs. the Bias (y-axis).

5. Use the Coefficient of Determination (R2) to determine if the linear fit is adequate. (Usually a number like 70% or higher provides sufficient linearity). If the fit is linear, then the linearity calculation is:

where:

|slope| is the absolute value of the slope of the Regression line calculated by the Scatter Diagram

PV is the Process Variation, which may be estimated as 6 times the Process Sigma value calculated from a control chart.

Learn more about the Quality Improvement principles and tools for process excellence in Six Sigma Demystified (2011, McGraw-Hill) by Paul Keller, or his online Green Belt certification course ($499).