文档介绍：Measurement System Analysis
(Measure Phase)
Acknowledgement
The examples quoted in this module are from
1) Six Sigma Project # 4454-01
“Reduction of Nitrogen Consumption for Reflow”
R. Thiruselvam (Asia Pacific Disc Drive Operations)
2) Six Sigma Project # 4454-02
“Reduction of AOI False Rejects”
R. Thiruselvam (Asia Pacific Disc Drive Operations)
Scope of Module
Components of Gage Variation
Effect of Measurement Variability
Assessing Measurement Variability
Measurement System Analysis for Attributes
Components of Variation
Components of Gage Variation — Accuracy
Accuracy is the agreement between the average of a set of measured
values and the true value of the characteristic being measured.
Bias is the difference between the average of a set of measured values
and the true value of the characteristic being measured.
x
–
Bias
True Value
Observed Value
Example 1
An engineer chose five “golden units” that represented the expected
range of measurements. Twelve random measurements were made
on each part. Historical process variation was found to be 12 mm.
Data can be found in the Measurement System file.
Mean of the sixty measurements x = mm
Mean value of the five standards = 6 mm
Bias = x –= – 6
= – mm
= % of process variation
–
–
Components of Gage Variation — Linearity
Linearity is the difference in bias throughout the expected range of
measurements.
1
2
Components of Gage Variation — Linearity
Linearity may be obtained via the following procedure:
a) For each case, compute the Error (ei), . true value – measured value
b) For each part, compute the Mean Error (ei), . iei n
c) Determine the slope for best fit line of Mean Error vs True Value
d) Linearity = Slope × Process Variation
–
Example 2
Using the data set from Example 1,
Master Mean Error
2 –
4 –
6 –
8
10
Best Fit Line : Mean Error = – + Master