Thanks for your response! Sorry that the example was confusing. The example I provided is closest to the way that we are using the ternary plot, without sharing confidential data.
I understand that the way the plot is being used is not typical. The process factor (time) does change inherently change as the reaction progresses - and what we are trying to do is track the reaction's 3 species as time progresses. The 3 species add up to 100.
Thanks for the suggestion. I'm a novice with JMP, and I have briefly looked into the mixture profiler. It seems like mixture profiler need a function or model to use it appropriately, which is something I don't fully know how to implement with this. Would it require a DOE type of analysis?