Modelling by Experimental Designs

Experimental designs are useful for anyone undertaking scientific research or industrial studies. They are applicable to all disciplines and all industries, from the moment we investigate the relationship between a quantity of interest, y, and variables, x _j , which can modify the value of y. As soon as we take an interest in the function :

you need to think about experimental designs. Their purpose is to optimize the organization of experimental trials, so as to obtain the maximum amount of information with the minimum number of experiments, and the best possible precision on the responses calculated with the model. This objective is achieved by following mathematically established rules and adopting a rigorous approach. There are numerous experimental designs to suit all the cases encountered by an experimenter. Some are more frequently used than others. We will outline the fundamental principles of this new science called experimentation, and review the majority of designs available to experimenters. If they don't find the right design for their study, they can even fashion an original one to suit the requirements of their work.