From the uncertainty set to the solution and back: the two stage case

Robust optimization approaches compute solutions resilient to data uncertainty, represented by a given uncertainty set. Instead, the problem of computing the largest uncertainty set that a given solution can support was, so far, quite neglected and the only results refer to the single stage framework. For that setting, it was proved that this problem can … Read more

Robust optimization: from the uncertainty set to the solution and back

So far, robust optimization have focused on computing solutions resilient to data uncertainty, given an uncertainty set representing the possible realizations of this uncertainty. Here, instead, we are interested in answering the following question: once a solution of a problem is given, which is the largest uncertainty set that this solution can support? We address … Read more

Efficient approaches for the robust network loading problem

We consider the Robust Network Loading problem with splittable flows and demands that belong to the budgeted uncertainty set. We compare the optimal solution cost and computational cost of the problem when using static routing, volume routing, affine routing, and dynamic routing. For the first three routing types, we compare the compact formulation with a … Read more

Using mixed-integer programming to solve power grid blackout problems

We consider optimization problems related to the prevention of large-scale cascading blackouts in power transmission networks subject to multiple scenarios of externally caused damage. We present computation with networks with up to 600 nodes and 827 edges, and many thousands of damage scenarios. Citation CORC Report TR-2005-07, Columbia University Article Download View Using mixed-integer programming … Read more