K.J. Siebert
Multivariate analysis procedures have occasionally been applied to brewing problems and appear particularly appropriate and useful. Exploratory Data Analysis methods can reveal how many fundamental properties are represented in a data set and could be used to collectively examine a breweryÕs analytical procedures to find and reduce redundancy. Pattern Recognition procedures can identify beer brands or raw material cultivars and detect adulteration. They also have use in multivariate QA/QC. Empirical modeling has many applications in process control and in discerning relationships between product composition and properties. Multivariate calibration can measure either multiple analytes or single analytes in difficult matrices.