I have performed a gage analysis using standard sample size of 30 parts, 3 operators and 3 repeats. The result from Minitab is attached.
The 95% CI was calculated separately using modified large-sample (MLS) method* for two-factor balanced random ANOVA (crossed) model. The total gage R&R (a) = 0.00674 (95% CI: 0.00484, 0.04732) and the part to part variability (b) = 0.17594 (95% CI: 0.08250, 0.59008).
I will like to obtain the %contribution which is (a/(a+b))*100. The point estimate for %contribution is 3.6895% ~ 3.69%. How do I calculate the 95% CI for the %contribution?
* as described in Design and Analysis of Gauge R&R Studies: Making Decisions with Confidence Intervals in Random and Mixed Anova Models by Richard Burdick et al.
A quick and simple approach would be to use a bootstrapping / resampling approach.
Thank you. I had a looked at your suggestion. An article is linked.
But the problem still remained.
Having determined the upper and lower limits of the 95% CI estimates for sigma repeatability, reproducibility and total gauge (regardless of the method; MLS or bootstrapping) how do I determine the 95% CI limits for %study var. .
Maybe I should have asked how do I determine the 95% CI for %study var to be precise.
http://www.cob.calpoly.edu/~eli/pdf/TQM-14-3.pdf
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA