Network interdiction with asymmetric cost uncertainty

We study a shortest-path interdiction problem in which the interdictor acts first to lengthen subset of arcs, and an evader second select shortest path across network. In this problem, cost for evader’s arc consists base if is not interdicted, plus additional that incurred interdicted. The aware costs when action taken, but does know values are uniformly distributed within given (arc-specific) intervals. evader, on other hand, observes exact value costs, due actions. interdictor’s thus maximize expected minimum attainable by evader. provide partitioning algorithm computing optimal solution leveraging bounds gleaned from Jensen’s inequality as proposed earlier maximum-flow problem. also several algorithmic strategies accelerating convergence demonstrate their effectiveness randomly generated instances.