optimizing decisions

[#1]

I am trying to see how one would use JMP to optimize decision-making. I made some progress but figured that there are entire industries based around this theme (!), so maybe no need to invent the wheel here:

I want to decide which is the best conservation strategy given a set of inputs: a facility's resources, the nature of the environmental problem, and things like that (all on a common scale of 1=bad and 5=great). Depending on these factors, the model will project the optimal conservation solution. I have set this up to where a simple nominal logistic model could be run, in which the conservation approach recommended is the Y. I can then use the profiler and input new data to see which conservation approach is projected. For instance, only extremely wealthy facilities/locales can generally mitigate seawater at large scales; if resources are anything below 5 (the highest), mitigation CANNOT be an option. As another example, if the seawater quality is a 1 (lowest), some conservation strategies make no sense at all (e.g., restoration, in which case everything you placed in the ocean would die). It seems like my preliminary results make sense, but I'm wondering if there are entire platforms dedicated to this kind of exercise, albeit for manufacturing, like the optimization and process capacities.

I want to pitch the notion of "we humans are generally bad at decision-making, some I'm going to let the computer do it for me!" Am I making this into something harder than it actually is? I could see making a really convoluted set of conditional arguments as another alternative. 

[#2]

I haven't looked at the JMP data table so my thoughts below are data agnostic. Have you tried any of the tree based modeling methods? Seems to me that 'trees' might be a useful method to play with the inputs under constraints/conditions that are preordained in the real world physical scenarios you could paint. Another method that also comes to mind is not mathematically modeling based but more thought exercise based...scenario planning. I used the process and taught it in an industrial setting for things like capacity planning, resource utilization and such. It helps paint a variety of pictures of alternate realities and allows for things like mutual exclusivity and such. You don't get a mathematical answer per se out of the method...but the scenarios could be inputs into a tree model. Here's a Wikipedia article on scenario planning.Scenario Planning 

[#3]

Peter, thank you for your thoughts. I had indeed envisioned this as a "tree" or a flow-chart, and this scenario planning does certainly appear to have many parallels. I just need to try to find a way to make it more quantitative rather than my gut feelings, specifically by tapping into probabilities: "If my water is characterized by property X, then there is an 80% that option A is the way to proceed." Using the profiler with a nominal logistic model in which the proposed solution is the Y sort of works, but I can't help but feel there is a better way. In any event, I will definitely play around with some of tree options under Fit Model, as well as check out this scenario planning article. I'm sure both will give me some fresh ideas on this topic. Thanks again.