Abstract
In the analysis of mixture experiments it is not unusual to have many response variables (or physical characteristics of the end product) under investigation simultaneously. Generally these responses are correlated with one another, and the analyst must compromise by concentrating on what he or she considers the most important characteristics, at the expense of others, in selecting the optimum formulation. Furthermore, in those situations in which some responses are positively correlated while others are negatively correlated, it can be very difficult to understand the relative importance of the different components. In such situations, a biplot display can help the analyst not only to understand the underlying structure of the data better but also to understand the roles played by the different components. Biplots are so named because both row (mixture formulation) and column (response) information are displayed in a single plot. This graphical technique has been applied successfully to mixture experimen...
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have