A tabu search algorithm with a probabilistic neighbor selection criterion for capacitated multicommodity network flow problem

In recent years, network design has been an important issue not only for researchers but also for the industries. This is due to the impact that an efficient distribution network design can have over both costs and the environment. In this article we presented a heuristic solution approach for the multicommodity capacitated network flow (MCNF) problem, which corresponds to a very complex problem within the network design problems. This problem is solved by a Tabu Search (TS) algorithm which obtains improvements in comparison with the used benchmark, especially in small and medium instances. Particularly, this TS algorithm selects the best candidate on the list, through a probabilistic approach based on evolutionary algorithms (TS-PCNS).