Universität Dortmund, Sonderforschungsbereich 475 - Komplexitätsreduktion in Multivariaten Datenstrukturen, Dortmund
The Desirability Index (DI) is a method for multicriteria optimization in industrial quality control. By design of experiment methods and transferring the multivariate into a univariate optimization problem settings of influence factors are selected that lead to a process with simultaneously optimized quality measures. In this paper a new field of application for the DI is introduced - the field of process control. When a process was designed with the objective of reaching the optimal value of the DI, the DI therefore is the most appropriate measure to monitor this optimally over time. Based on the distribution of the DI control charts for individual measurements are presented and advantages compared to the traditional approaches are pointed out, especially caused by an innovative procedure for the interpretation of out-of-control signals.