IEEE.org | IEEE Xplore Digital Library | IEEE Standards | IEEE Spectrum | More Sites
Design of experiments (DOE) is an approach to experimental work that aims to produce the best predictions in the most economical way.
The world is noisy and multifactorial. Experiments are necessary to assess reality. But experiments can have a high cost in terms of time, delay and resources. A trade-off between costs and benefits of the experimental campaign must be done. Based on statistics, DOE offers tools for academic and industrial research. It ensures that the quality of the data is adequate to answer the experimenter’s questions.
The presentation shows with two examples how DOE is implemented and the type of insight it can bring.
Keywords: Design of experiments, fractional factorial design, Plackett-Burman design, experimental variance, empirical modeling, factor interactions