A case study on process optimization using the gradient loss function

This paper focuses on studies where the goal is to optimize the key quality characteristics of a specific product or process. In order to do that, the process parameters must be adjusted in such a way that deviations from target are minimized while robustness to noise and to process parameter fluctuations are maximized. Since this is a multi-response, multi-objective problem, the optimum solution is a compromise. In this paper we present a gradient loss function which captures all the desired objectives, and propose a general methodology to perform the optimization analysis. This methodology is then carried out experimentally on a cosmetic product and the results are discussed.