Dataset: steel_rough.csv
Source: R.S. Peruchi, P.P. Balestrassi, A.P. de Paiva, J.R. Ferreira,
M. d. S. Carmellozi (2013). "A New Multivariate Gage R&R Method for
Correlated Characteristics," International Journal of Production Economics,"
Vol. 144, pp. 301-315
Description: Results of a 2-Way Random Effects Model (aka Gage Repeatability
and Reproducibility). The study had 12 parts, 3 operators, and 4 replicate
measurements. There are the following variance components:
Model y = mu + alpha_i + beta_j + (ab)_ij + e_ijk
V(P) = V(alpha)
V(Repeatbility) = V(e)
V(Reproducibility) = V(beta) + V(ab)
V(MS) = V(Repeatability) + V(Reproducibility)
V(Total) = V(P) + V(MS)
%R&R = 100*sqrt(V(MS)/V(T))
Signal-to-Noise Ratio: SNR = sqrt(2V(P)/V(MS))
Variables/Names
Part (part)
Replicate (replicate)
Operator (operator)
Arithmetic Average Roughness (roughave)