Handling Constraints When Designing Experiments
Presenter: Don McCormack
Working with Linear Constraints and Handling Blocking and Hard-to-Change Factors
The presenter uses tablet processing data to demonstrate how to design an experiment that has linear constraints that eliminate a corner of the design space. He also shows how to determine the values for those constraints and explains when to treat factors as blocking or hard-to-change, depending on the type of variability you want to characterize. The presenter concludes by building a design that incorporates one and two random blocks and a design that addresses process factors that are difficult to change.