Exact and Heuristic Algorithm for Multi-constrained Optimal Path Problem

The Multi-Constrained Optimal Path (MCOP) problem is an abstraction and extension of the QoS Routing, and plays an important role in different applications. The paper gives an exact algorithm and a heuristic algorithm based on Tabu Search for solving MCOP, respectively. For exact algorithm, a tighter bound is proposed on the basis of Branch and Bound algorithm so that it satisfies a large number of requests, and by introducing aspiration criteria at the appropriate time in H_MCOP, we propose a novel heuristic algorithm named TS_MCOP, based on Tabu Search. Computational experiments shows that TS_MCOP improves in view of success probability, optimality, average cost deviation compared to original algorithms by 0.8%, 16%, and 20%.