region around the optimal

Question

I'm having an issue when trying to explore the region around the optimal solution in an n-dimensional space.

We have OML which finds the optimal solution (very quickly), but I would like to present the user of the model with a region around the optimum, so that they can see how sensitive it is to movement away from that solution in any dimension.

We have tried putting nested ForEach's in the constraint definitions (to do a brute-force solution), but it is very slow. Programming this brute force approach in pure C# is fast, but we lose the ability to keep it dynamically synchronized with the OML.

Is there any input directive or solution meta-data available which would tell us the region surrounding the optimum? Or can someone suggest an alternative?

Answer 1

Is this an LP model? If so, the sensitivity report can give you some of this information. Otherwise, no there is not a built-in way to do that.

Nate

Answer 2

Thanks for the response, Nate.

No it's not an LP model unfortunately.