Response surface designs minimize cost by reducing the number of experimental formulations required to study a particular product characteristic. The response surface designs presented by Mullen and Ennis (1979) worked with a maximum of 5 variables. However, researchers are often faced with the question of how to handle a problem which has a larger number of variables. We recently had the opportunity to consider the problem of prioritizing 15 variables in an industrial blending project which led to considering the theoretical aspects of various screening techniques.
This article appears as:
Mullen, K. and Ennis, D. M.(1985). Fractional factorials in product development. Food Technology, 39(5), 90,92,94,97-98, 100, & 102-103.
Colleagues can request this journal article here:
Fractional factorials in product development
Not a Colleague? Click here to join for free!