Design of Experiments (CESPE Academy)

CA15

Engels

Description

Design of experiments, or DOE, is a key tool for product and process improvement and innovation in pharmaceutical domain. This course motivates the standard and routine use of a fully flexible approach to the design of experiments, named optimal design of experiments, by showing its industrial application in a variety of case studies covering a wide range of practical situations. The increasing computing power has made optimal experimental design a key tool for scientists and researchers in the 21st century.
The course will cover completely randomized experiments, blocked experiments and split-plot types of experiments, applied to screening experiments and response surfaces experiments. This training will conclude with a discussion of definitive screening designs and OMARS designs, which allow an efficient combination of screening and response surface experimentation. Throughout the course, the introduction of new theoretical concepts, demonstrations with the JMP software and exercises are intertwined.

Programme

Link to Course Programme

Day 1 – Introduction to Design of Experiments

Linear regression analysis
- Quantitative & categorical factors
- Main effects, interaction effects, quadratic effects
- Selecting, visualizing, and exploiting models
Optimal design of experiments
- Precision of estimation and prediction
- D- and A-optimal design
- I-optimal design
- Orthogonal design
- Randomization
D- and A-optimal screening experiment

Day 2 – DoE for screening, optimization and mixtures

I-optimal response surface experiments
- Combining quantitative and categorical factors
- Dealing with constraints on the levels of the factors
Blocking
- Importance of blocking and modeling its data
- Blocked screening experiment
- Blocked response surface experiment
Mixture experiments
- Dedicated regression models
- Visualizing data and models
- I-optimal design of experiments for mixtures

Day 3 – Budget-constrained design of experiments

Split-plot types of experiments
- Hard-to-change factors
- Multi-stage experimentation
- Restricted randomization
- Modeling data in the event of restricted randomization
- Split-(split-)plot experiments
- Strip-plot experiments
Combining screening and response surface experiments in one
- Definitive screening designs
- OMARS designs

Remarks

Additional information

All sessions will be taught by Peter Goos. Peter Goos is a full professor at the Faculty of Bio-Science Engineering of the University of Leuven and at the Faculty of Business and Economics of the University of Antwerp, where he teaches various introductory courses on statistics and probability. His main research area is the statistical design and analysis of experiments. Besides a number of articles in top scientific journals in marketing, transportation, quality, operations research and statistics, Peter published numerous books and received the Brumbaugh Award, the Youden Award, the Shewell Award and the Lloyd S. Nelson Award of the American Society for Quality, the Ziegel Award and the Statistics in Chemistry Award from the American Statistical Association, and the Young Statistician Award of the European Network for Business and Industrial Statistics.
One week prior to the course, the registered participants will receive further information (directions, parking availabilities, course material).
The participants will receive an evaluation form to provide feedback on this training.
By registering for this course, the participants consent to the sharing of their name, email, and job information with the course instructor/provider for the purposes of tailoring and then sharing educational course-related information.
For further information or any questions related to the course, please contact Mark.Gontsarik@UGent.be

Annullation/Cancellation

Registration is possible up to one week before the day of the respective session.
Registrations can only be cancelled by email. Cancellations earlier than 2 weeks before the course start will be fully refunded. In case the cancellation occurs later than two weeks before, but earlier than one week before the start of the course, half of the registration fee will still be due. In case of cancellations less than one week before the start of the course, the full registration fee will still be due. For UGent PhD students, the price of the registration fees for the Academic (UGent postdoc or professor) category will be considered when calculating the cancellations fees.
The organization reserves the right to reschedule or cancel the training in case of insufficient number of participants.