Abstract: Sandro Macchietto

Imperial College London

Model-Based Design of Experiments: What for, by Whom and for Whom?

The Design of Experiments (DoE) is a rather old subject that can be traced back to the book by Fisher with the same title in 1935. However, it is fair to say it has traditionally been a niche interest area involving a small following in the statistical, mathematical and engineering circles, and pretty much ignored by the majority in the experimental community. Judging by the interest within academic circles, special workshops, conferences, but also the number and range of practical applications published, the subject seems to have recently come out of the closet and started to become a bit more used in practice, if not yet mainstream.

This presentation will discuss some of the reasons for this renewed interest and briefly review some key developments which, in the author’s view and experience, underpin and enable this success. One is identified as the evolution from viewing an experiment as purely a “black-box” to combining DOE statistics with more sophisticated physical representations of the experiment being designed, thus putting the “model-based” in front of DoE. Another, associated with the first, is a better understanding of the relationship between desired performance and evaluation metric(s), leading to the disaggregation of a single “best” design objective into constituent components, and thus to much richer formulations of the design problem that can be tailored to specific situations. A final reason is the substantial improvement in the numerical and computing tools supporting the model-based design of experiments, but also and chiefly in the availability of integrated modelling/solution environments which make the whole technology accessible to a much wider engineering community.

The presentation will illustrate, with reference to examples, some of the new problem formulations that can be used to represents more sophisticated design requirements (including parameter precision, anti-correlation, robustness to uncertainty) and some of the newer solution approaches (including design of parallel experiments, on-line re-design). It will also illustrate some successful applications in a variety of demanding industrial areas, ranging from Fuel Cells, to complex reactor design, to biomedical applications.


 

Aachen Institute for Advanced Study in Computational Engineering Science (AICES)
at RWTH Aachen University, Germany. Email: acces11@aices.rwth-aachen.de
Tel. +49 (0)241 80 99130, Fax +49 (0)241 80 628498