Sequential design of mixture experiments for the development of new products

This paper presents a new sequential approach to mixture DOE that simultaneously selects new materials and ratios in which to combine them with the objective of achieving a set of specified final product properties. In each sequentially designed experiment the approach selects new materials and formulation ratios through an optimization based on mixture-property partial least square (PLS) models that account for similarities among the properties of different materials (Muteki et al. 1, 2). The approach is useful for the case where one has existing experiments some of which may not be too far from those required for the desired new product. The sequential DOE methodo logy is demonstrated on an industrial polymer blending problem. Copyright © 2007 John Wiley & Sons, Ltd.