In the analysis of mixture experiments it is not unusual to have many
response variables (or physical characteristics of the end product) un
der investigation simultaneously. Generally these responses are correl
ated with one another, and the analyst must compromise by concentratin
g on what he or she considers the most important characteristics, at t
he expense of others, in selecting the optimum formulation. Furthermor
e, in those situations in which some responses are positively correlat
ed while others are negatively correlated, it can be very difficult to
understand the relative importance of the different components. In su
ch situations, a biplot display can help the analyst not only to under
stand the underlying structure of the data better but also to understa
nd the roles played by the different components. Biplots are so named
because both row (mixture formulation) and column (response) informati
on are displayed in a single plot. This graphical technique has been a
pplied successfully to mixture experiments in the photographic develop
ment area and in plastics formulation, as our examples will show.