Partial factorial designs for multivariate calibration: extension to sevenlevels and comparison of strategy

Multilevel experiments are necessary in chemistry for calibration and mixtu re experiments. This paper presents methods for generating designs for seve n concentration levels (l = 7) using l(2) experiments paying special attent ion to the problem of orthogonality between successive factors. In addition , a method to extend all l level designs to orthogonal arrays with double t he number of mutually orthogonal factors is shown. Some comparisons are mad e between different designs using leverage. Finally, recommendations are ma de in order to select the most appropriate design for each specific situati on. (C) 1998 Elsevier Science B.V. All rights reserved.