Design of experiments or DOE is a key tool for product and process improvement and innovation. However, experimenters often have to deal with a mismatch between standard experimental designs, such as factorial and fractional factorial designs, central composite designs, and the features of their problems. This course motivates the standard and routine use of a fully flexible approach to design of experiments, named optimal design of experiments, by showing its application in ten case studies covering a wide range of practical situations. The increasing computing power and the availability of user-friendly software for the tailor-made design of experiments has made optimal experimental design a key tool for researchers, engineers and statistician in the 21st century. This course will demonstrate the usefulness of optimal design of experiments in a wide variety of contexts, and discuss the analysis of the data produced by the experiments. Throughout, the introduction of new concepts, JMP demos and exercises are intertwined. The JMP software will be available to all participants.
The course will deal with screening experiments, response surface experiments and mixture experiments, and show how to take into account practical complications such as constraints on the factor levels, the need for blocking, the availability of covariate information concerning the experimental units, and difficulties to randomize the experimental tests. Blocked experiments, split-plot experiments and strip-plot experiments will therefore be key topics.
While the concepts and theory will be explained in detail, this is a hands-on course aimed at applying optimal experimental design and analyzing data, rather than a course on optimal design theory. Therefore, the Spring School 2018 complements the Summer School 2018, where the emphasis is more on the theory rather than on the application.
