Many of the current multiple-response optimization approaches fail to account for uncertainties, resulting in misleading quality estimates that lead to poor product design. This study proposes a Bayesian decision theoretic approach to the modeling and optimization of multiple-response systems that accounts for the correlation among the responses, the variability of the predictions, and the uncertainty of the model parameters. A Bayesian model averaging approach is also proposed to account for response-model uncertainty. The approach is applicable to many types of quality criteria and characteristics.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados