Fundamental Design of Experiments

Experimental design is the science of optimizing test information given fixed resources. The origin of modern design of experiments (DOE) started in the field of agriculture and is associated with the work of Sir Ronald Fisher in the early 1920’s. Systematic application of DOE to industrial problems started thirty years later with the works of George Box at Imperial Chemical Industries and Genichi Taguchi at the Institute for Statistical Mathematics and the Electrical Communication Laboratory. Jack Kiefer’s development of optimal designs in the 50’s and 60’s allowed DOE to expand beyond its mathematical foundation to the world of computer generated designs.

Though DOE has come a long way since the early 60’s, many of its concepts can be traced back to ideas from these four people. This talk will focus on fundamental DOE concepts as they can be applied to fermentation processes and media design. Topics to be discussed will include factorial designs, split plot experimentation, mixtures, robust designs, and nonlinear design.