On the rapid development of new products through empirical modeling with diverse data-bases

摘要

This research addresses the development of industrial products with specified final properties. There are basically three major degrees of freedom to control the final product properties: the selection of the raw materials, the selection of the ratios in which to use the selected materials, and the process conditions under which to manufacture the product. A new approach to modeling and optimization is presented that simultaneously takes into account all of these degrees of freedom for the development of new products. The approach involves building multi-block Partial Least Squares (PLS) models that pull together diverse industrial databases on previously made products and on the properties of the component materials used to make these products. The resulting models are then used in an optimization framework to select raw materials from much larger databases (including materials never previously used), to select the ratios in which to combine them and to select conditions under which to process them in order to yield a product with specified end properties at a minimum cost. The methodology has been successfully used for the development of several industrial products.