Research On Ant Colony Algorithm And Its Application For Vehicle Scheduling Problem

With the development of the information, logistics, takens as "Third Profit Resource", has been paid more attention and is growing to be faced to a series of opportunities.In logistics, distribution is crucial and directly linked with the customers. Whereas, the optimization of vehicle planning of the distribution system is a key part of the whole system. By the optimization of vehicle planning, the enterprises can improve their nucleus competitiveness and achieve scientific process of logistics.In this thesis, we begin with the introduction of vehicle routing problem related topics, establish the style of the problem is non-full loaded,single deport,close and with time window and build the mathematical model of vehicle routing problem. Introduce the background and details of Ant Colony Algorithm, among the application of Ant Colony Algorithm, we emphasize some existing methods solving vehicle routing problem, build the basic model of Ant Colony Algorithm, analyze the realization process of Ant Colony Algorithm,and according to the realization process, program the scheduling procedures with VB language. Again, we put forward some beneficial suggestions on the solution through analyzing the every parameter in the function of solution.According to the defect of Ant Colony Algorithm, which are slow convergence speed and being able to converge to the local optimum solution and put forward on improved Ant Colony Algorithm, that is GAAA algorithm, it is the core of the research. In the improved algorithm, using the characteristic which is Ant Colony Algorithm apt combined with other approaches, the improved algorithm is based on the combination of Genetic Algorithm and Ant Colony Algorithm to enhance the solution efficiency. First, Genetic Algorithm has the ability of doing a global searching quickly and stochastically, so the algorithm adopts Genetic Algorithm to give information pheromone to distribute.Second, Ant Colony Algorithm has the ability of parallel processing and global searching, so the algorithm makes use of Ant Colony Algorithm to converge to the global optimum solution.Ultimately, realizes the improvement of the algorithm, solves the defect of the Ant Colony Algorithm,improves the efficiency of improved algorithm in solving the vehicle routing problem.Using the development program, optimization simulation designs are made to solve the classical vehicle routing problem. Applied the combined algorithm to the data defined by Solomon and compared with the Ant Colony Algorithm, achieves the fine effect, proves the feasibility, efficiency and universality of this improved algorithm.