Design of experiments

On completion of the course, students will be able to take part in the design, implementation and interpretation of a design of experiments.
This implies being able to:

Define the appropriate type of design of experiments
Choose the factors and the levels at which to study them
Interpret the results of the design qualitatively, quantitatively and statistically
Conclude on a design, be able to extend it if necessary and optimize on the basis of the results.

Teachers are expected to:

Present the method's potential
Define the design of experiments methodology and its conditions of use
Implement the appropriate design of experiments and process the data obtained
Evaluate the relevance of the results obtained

Basic knowledge of statistics and use of R software

Design of experiments (DOE) is an essential methodology for companies committed to continuous improvement. They are used to reduce experimental effort by minimizing costs and study times, while ensuring more reliable information in all areas, from product design to process control.
This course shows how to develop an optimal experimental study strategy in practice, and how to make reliable use of experimental results. It consists of lectures and practical exercises on an experimental simulator.

Methodology
Full factorial designs
Critical analysis of results
Fractional factorial designs
Modeling and optimization
Implementation problems
Modification of the experimental domain: steepest slope and Simplex methods
Second-order designs (centered composite designs and Box-Behnken designs)

During the practical sessions, the implementation of experimental designs using R software will be covered.

Engineering Sciences and Techniques

Les plans d'expériences Gilles et Marie-Christine SADO AFNOR Technique 2002
Introduction aux plans d'expériences - Applications Jacques GOUPY Lee CREIGHTON Dunod 2009
Pratiquer les plans d'expériences Jacques GOUPY Dunod 2005
Design and analysis of experiments D. MONTGOMERY J Wiley 2008
Guide d'autoformation à R