Mixture design model with uncertain response and its application

In the application of mixture experiments, it's hard to model the outputs with the form of persons' description words mathematically. This paper models the outputs by fuzzy numbers and proposes a mixture model for such kind responses. The parameters of the proposed model are estimated by iterative least square method, and the goodness of fit of the model is evaluated by the average closeness degree. Finally, the procedure is illustrated by an example.

Automated summary

This paper proposes a mixture model for the outputs in mixture experiments, which are hard to model mathematically using description words. The model uses fuzzy numbers and the parameters are estimated by iterative least square method. The goodness of fit is evaluated by the average closeness degree. An example is provided to illustrate the procedure.