What inspired this wish list request?
Currently if you are required to hit minimum power requirements for specific expierement effects you need to generate the design, check the power analysis, evaluate if you met the requirements.... if not you need to return to the design and make changes (add replicates, ect). It would be ideal if we can specify RMSE, alpha, and power in the factor definition table.
What is the improvement you would like to see?
It would be ideal if we can specify RMSE, alpha, and power in the factor definition table and when the design is generated it takes this into consideration and provides designs which meet the expirement confidence requirements.
Why is this idea important?
It is important because the current optimization method is focused on minimizing cost while maximizing learnings. Allowing the user to determine what "costs" should be optimized would be signifigantly more useful and speed up expierement design.