DoE, Great if Results aren't Qualitative and a Model Can Be Made

LINK-> (Using Design of Experiment Methods for Screening and Optimization of Quantitative Factors when Results are Qualitative) details the sequential use of traditional DOE (vertices generations and optimality selection) to generate a small set of trial experiments from the number of factors.  The best result from a set of experiments is used as the starting point for the next series of experiments.  It will eliminate unnecessary ones and optimize the others (screening and optimization) .  Its primary use is for situations when quantitative results are to difficult, impossible or time consuming to obtain or when a mathematical model of the system cannot be determined or it is not clear how to weight multiple responses.