The article presents methods for the study of experimental factors. The use of Ishikawa diagrams as tools for identifying relevant experimental variables is discussed. The construction and evaluation of two-level factorial and fractional factorial designs are presented in detail. Statistical properties of such designs are discussed. It is shown how the use of t-statistics and confidence intervals as well as how cumulative Normal probability plots can be used to identify significant variables. The methods are illustrated by examples from chemical experiments.
ISBN för värdpublikation: 978-0-444-64166-3