The closest i can come to thinking about this is to suggest the Taguchi methods as a basis for this. Perhaps a web search on Taguchi will help.

JMW
www.viscoanalyser.com
Eng-Tips: Pro bono publico, by engineers, for engineers.

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

Really not in my area, but our process engineers use Aspentech HYSYS to model the plant and review changes, different senarios, model expansion...etc; before making any changes in the field.  They report the model is right on the money...but like software, remember, garbage in - garbage out (be prepared to spend time inputing existing conditions and parameters).

http://www.aspentech.com/includes/product.cfm?IndustryID=1&ProductID=274

i sincerely doubt there is a software that is able to make conditional decisions and works better at that then a human being. if you come across such a software, please inform me because i'd very much like to take a look at that.

but perhaps i missunderstood and you are referring to an economical decision as done with LP models. if this is the case, you should look into PIMS (also by aspentech) or aspen.refsys (if it is avaiable by now). but these tools just calculate the optimal solution of a system of linear equations.