A Polynomial Time Algorithm for Minimax-Regret Evacuation on a Dynamic Path

Abstract: A dynamic path network is an undirected path with evacuees situated at each vertex. To evacuate the path, evacuees travel towards a designated sink (doorway) to exit. Each edge has a capacity, the number of evacuees that can enter the edge in unit time. Congestion occurs if an evacuee has to wait at a vertex for other evacuees to leave first. The basic problem is to place k sinks on the line, with an associated evacuation strategy, so as to minimize the total time needed to evacuate everyone. The minmax-regret version introduces uncertainty into the input, with the number of evacuees at vertices only being specified to within a range. The problem is to find a universal solution whose regret (difference from optimal for a given input) is minimized over all legal inputs. The previously best known algorithms for the minmax regret version problem ran in time exponential in k. In this paper, we derive new prop- erties of solutions that yield the first polynomial time algorithms for solving the problem.