Research, Based On Ant Colony Algorithm Tobacco Distribution Optimization Scheduling

The cigarette of our country has a monopoly in the particularity dealt exclusively and enormous volume of consumption to the cigarette, And the cigarette vehicle deployment question provides and delivers the key question in the system as the tobacco logistics, have important research and using value. Papers for the tobacco distribution logistics distribution center path optimization problems, based on ant colony algorithm to improve the distribution of the optimized scheduling method.Paper describes the mechanism of ant colony algorithm theory and principles of the system, from distributed computing, self-organization, positive and negative feedback has done a detailed introduction. Papers will be single-loop path of the choice of distribution reduced to a single traveling salesman problem (TSP) is solved, combined with the basic principle of ant colony algorithm with the model, a single-loop distribution of ant colony optimization problem through compare selected simulation results the basic ant colony algorithm in the initial parameters, the basic ant colony algorithm for the slow convergence and easy into the shortcomings of local optimum, an improved ant colony algorithm. Improved Algorithm for the max-min ant algorithm from the use of restrictions on the scope of the idea of pheromones, the shortest path as a result of inhibition and increased the gap between the longest path of the amount of information arising from the phenomenon of search stagnation; At the same time, the introduction of pheromone-proliferation strategy and to improve the overall situation of pheromone update mechanism, to effectively control the phenomenon of precocious puberty, so that faster convergence algorithm to approximate the global optimal solution, the issue of three TSP simulation test shows that the proposed algorithm in the search capabilities and performance have been improved significantly.Papers will also be improved ant colony algorithm applied to multi-loop vehicle scheduling problem, the establishment of a multi-loop mathematical model of the vehicle scheduling problem, the problem with the TSP analysis of the similarities and differences compared to the approximate solution of the model proposed acquisition strategy, given the multi-loop vehicle scheduling problem of ant colony algorithm to improve the implementation of specific steps, and through a city in the tobacco distribution point 20 of the examples of solving the model, the validity of the algorithm in multi-loop applications, vehicle scheduling problem of the validity.