With (6 sigma) improvement projects we try to find the most important process parameters in order to understand, adjust and control a process. Carrying out experiments is essential for this, for which the Design-of-Experiments method (DOE) is extremely suitable.
Until recently, however, a 'relatively' large number of experiments were necessary, certainly if you want to study several process parameters simultaneously. With the invention of the Definitive Screening Designs (B. Jones and Nachtheim), DOE is within reach of every company, even in production environments. In the article (Chemical Engineering Journal), we at Bayer Crop Science (formerly Monsanto) made a comparison between the Definitive Screening Design approach and a classical Response Surface Design (RSM is a kind of optimisation design). The results are astonishing. For the analysis of 6 process parameters, very similar results were obtained with a DSD in only 13 experiments as with the required 90 experiments according to the RSM method.
With the Definitive Screening Designs, we have already achieved good results in the textile, pharmaceutical, biochemistry and metal industries, both in laboratory and production environments. The DSDs bring the 6 sigma approach into a higher gear.