MAKING MIXTURES ROBUST TO NOISE AND MIXING MEASUREMENT ERRORS

Mixture experiments involve the mixing or blending of two or more ingr edients to form an end product. Typically, the quality of the end prod uct is a function of the relative proportions of the ingredients and o ther extraneous process factors such as heat or time. When some of the process variables are either uncontrollable or difficult to control ( i.e., noise variables) the goal of a mixture experiment should be to f ind the mixture amounts and process settings that lead to a product of high quality that is also robust to the noise. Due to the nature of m ixture experiments this leads to a constrained optimization problem. T his article discusses setting up an appropriate objective function and provides techniques for determining the robust mixture proportions. I t is also shown that under certain conditions mixing measurement error s can be handled in the same way.