Investigation of the homogeneity of solids with a linear regression model

Abstract

Testing of the hemogeneity of solids is usually based on the comparison of the variance of measurements at several sample points with the variance of repeated measurements. In the case of destructive analytical methods, empirical mathematical models are needed because repetitions at the same subsample are impossible. Successful application of such models depends on the adequacy of the mathematical model and the type of distribution observed. Gradients and textures with low or medium local frequency require other tests than stochastic distributions with high local frequency. A simple regression model base on a 2 2 factorial allows conclusions about the type of distribution without great mathematical effort.