| dc.description.abstract | Traditionally, the minimum cost transshipment problems were simplified as linear cost problems in order to reduce problem complexity. In practice, the unit cost for transporting freight usually decreases as the amount of freight increases. Hence, in actual operations the transportation cost function can usually be formulated as a concave cost function. Great efforts have been devoted to the development of solution algorithms. However, they were confined to specical transportation networks. Besides, their methods were focused on local search algorithms or traditional heuristics. Recently, researchers began to use advanced neighborhood search algorithms to solve concave cost bi-partite transportation network problems to enlarge search area and find near-optimal solutions. This type of research, however, neglected flow transfers in transportation networks. Recently, there has been research adopting the genetic algorithm (GA), the ant colony system algorithm (ACS) and the particle swarm optimization algorithm (PSO) for solving concave cost network problems, and obtaining better solutions than some neighborhood search algorithms do. The harmony search (HS), a global search algorithm, has led to good results in many applications. In some applications, HS was even more effective than GA. Since there has not yet been any research applying HS to minimum concave cost network flow problems, we employ HS, coupled with the techniques of PSO, ACS and TA, to develop one global search algorithms for efficiently solving minimum concave cost network flow problems.
In the solution method, we take the harmony search as the foundation, use a global search which is harmony memory consideration and a local search which is Pitch Adjusting to compose the new feasible solution, we also join the velocity update rules in PSO, the pheromone update rules in ACS, TA and the concave cost initial solution algorithm, we develop a analogous harmony search which is fitting the minimum cost transshipment problems with concave costs. Finally, to evaluate our algorithms we designed a network generator to create a sufficient number of problem instances. The C++ computer language was used to code all the necessary programs and the test was performed on personal computers. To evaluate our algorithm, we also tested the recently designed TA, GDA, GA, ACS and PSO that solve minimum concave cost network flow problems. The results show that the developed algorithms performed well in the tests. | en_US |