|
Abstract:
|
In a mixture experiment, several components are mixed to produce a response. Existing literature on mixture designs ignores the addition order of the mixture components. We study the Order-of-Addition (OofA) Mixture experiment, where the response depends on the proportions of each mixture component and their addition order. Empirical study demonstrates that if mixture-order interactions exist, then the optimal mixture proportions identified by traditional models may be misleading. Full designs are proposed for the OofA Mixture experiment, and an algorithm is provided for their construction. Models based on these designs are introduced which include (1) typical mixture model parameters and (2) pairwise order-of-addition effects. Models with mixture-order interactions are also considered. An illustrative example of these methods is shown.
|
