Watch Now 20:47
Presenter: Don McCormack
Creating Three Designs
The presenter gives an overview of design diagnostics that can be used to help guide you in selection which designed experiment for a product, process, or behavior is most suitable for your needs. He suggests that when comparing designs one should consider, for each design, what can be estimated, how precise are the estimations and the effect of leaving things out of the model. He builds and compares a Custom Design, Definitive Screening Design and an augmented Plackett-Burman-like Design for a product with two categorical and seven continuous factors.
Comparing and Choosing Designs
The presenter examines three different designs he created. He performs a power analysis to help determine which factors might be identified as useful for determining the probability of detecting a real signal, identifies what portion of the design space is below a certain variance and plots the prediction variance surface as a function of the design factors.