Design of Experiments, DoE

Today’s development processes are characterized by complex products and manufacturing processes, as well as by ever more demanding time and cost targets. Design of Experiments (DoE) represents a major building block to meet this pressure to improve efficiency in the field of experimental design. Central applications are attempts to understand complex technologies and processes and, at the same time, to improve or optimize them. In practice, it is still the cases that, despite the existing knowledge about possible methods, “traditional” but mostly inefficient techniques like “One factor at a time” experiments are extensively applied in industrial practice. Negative examples also contribute to this, where the most “basic knowledge” is that DoE was useless, because no significant effects could be found. It is not enough to know only the methodology for a successful use of DoE in practice. The necessary conditions, which must be in place parallel to the actual experiment design, must also be considered. Successful DoE programmes use 80% of the time on planning and securing the experimental conditions especially with regard to variance reduction, and only 20% on statistical analysis. This seminar provides both the methodological foundations of the design of experiments as well as the overall understanding of the planning and evaluation of statistically planned experiments.