I have a stacked table with a Time Point (X) column, a Cumulative Release (Y) column, and several columns of factors from a DOE. We have different specification limits for release at various time points in the dissolution curve, as required by the FDA. In development, I need to fit curves to the data, use models to determine significant factors, and optimize factors using Desirability Functions. I can fit curves with this table, but I cannot use the Desirability Functions because each column can only have one set of specifications saved with it. I can split the table by Time Point and make a new table, but that creates more documentation and review work, as required by our data integrity rules. It would be nice if there was a way to account for these specifications in Fit Model and FDE.
Suggestions
- Allow specification limits to be set conditionally. This may help resolve the problem for non-FDE models.
- In FDE, allow more than one target curve. Curves based on upper spec, target, and lower spec would allow explorations of the variable space for each target curve.
I use JMP Pro 16.2.0