A variance components analysis fits a nested or crossed random-effects model, by ANOVA or maximum likelihood, to estimate how much variance each level contributes. In a fab, a nested design might sample 3 sites on each of 2 wafers from 10 lots and find that lot-to-lot variance is 70 percent of the total, wafer-to-wafer 20 and within-wafer 10. The biggest component tells engineering where improvement effort will pay off.
Variance components say 65 percent is lot-to-lot. Tuning the within-wafer profile won't move Ppk much.
Also heard as
- components of variance
Related terms
ANOVA
Analysis of variance: splitting total variation into parts from each factor and from error, then testing which parts are real.
Gauge R&R
Study that measures how much observed variation comes from the measurement system's repeatability and reproducibility.
Ppk
Process performance index: a Cpk-style figure computed with the overall standard deviation, capturing long-term variation.
Rational subgroup
Group of measurements chosen so that only routine, short-term variation can occur within it.