Webinar on Introduction to Mixture Design
14/11/19 at 3pm (UK time)
Experimental design is frequently used (at least in well-informed environments) to reveal cause and effect relationships in chemistry and related fields. Designs exist for a host of areas, such as screening, optimization, robustness testing and response surface modelling, and more.
Successful use of experimental design requires that the factors of interest (the process variables) are controllable. For process data, factors are usually varied independently of each other, and designs such as factorial designs, central composite designs, etc., are often encountered. While these designs are versatile and fully deserve their place in an experimenter’s toolbox, they are less suitable for mixture problems.
Mixtures are common throughout a range of industries. Paint, wine, concrete, pellets and gasoline are but some of the mixtures surrounding us. All mixtures have at least one thing in common: Their components sum to 100 %, and cannot be varied independently of each other. An increase in one component necessitates a decrease in another. This constraint makes standard designs at best sub optimal. The situation becomes even worse when additional constraints are imposed. The solution is to employ designs specifically designed for mixtures.
Special care is also needed when analysing the data from a mixture design, as well as when interpreting the model. This webinar will serve as an introduction to mixture design. It is aimed at those having some experience with classical designs, but unfamiliar with mixture designs.
This webinar will take place at 3pm (UK time) and will be presented by Bjørn Grung, University of Bergen.
If you wish to register please contact Christine Stevenson at Christine.email@example.com