Research On The Application Of Ant Colony Algorithm In Logistic

Logistics System is a network system, which not only includes logistic lines and logistic node network composed of entities, but also includes computers and communication systems composed of virtual network. Network takes various forms in the background of various problems. Network optimization is widely used to solve these problems in different areas, such as production, distribution, project planning, resource management, and so on.With the complexity and uncertainty of optimizing targets, traditional optimization based on determinacy algorithm has encountered great difficulties. Inspired from biological evolution and Bionics, many intelligent heuristic optimization methods, such as simulated annealing, genetic algorithm, tabu search, neural network algorithms and ant algorithm, are proposed to solve various complex optimization problems (NP-hard problem).As a new bionic optimization algorithm, Ant algorithm features in distributed computing, self-organization, positive feedback, and other characteristics and can be applied in a various applications. But long searching time and partial optimal solution danger might be possible weakness for the algorithm. To overcome these potential shortcomings, the author has proposed a method to initially adjust pheromone to zero to increase random probability as well as expand algorithm search space, in the hope of enhanced overall algorithm optimization capabilities. As for a long search time and the stagnation in sub-optimal solution for problems, this paper integrates the algorithm with 2-opt method to shorten calculating time and accelerate the speed of convergence. Through the TSP Benchmark, lots of simulation experiments have proved effectiveness of pheromone initial zero and 2-opt facilitated ant algorithm for small and medium-sized TSP to obtain the optimal solution, especially Eil51 and Tsp225 problems yield to better solutions.Logistics System transportation and delivery of goods is currently popular in ant algorithm application researches. Transportation is a particular case in linear programming. Ant network interpretation of these problems and the application of adaptive algorithm, various optimal solutions are available with better capabilities compared to traditional method and the smallest element Fu Younger algorithms. The larger set of data scale, the better transportation performance ant algorithm reveals. Goods Distribution is also known as vehicle routing problem (VRP), and in practice, there may not be a feasible VRP solution. To solve the problem, K - TSP solution is employed with priority to achieve feasible solutions. Once Ants have completed all their optimization, pheromone is further strengthened for an ultimate optimal solution or a better answer. Algorithm experiments and simulation results finally confirms that feasible solution priority ant colony algorithm for VRP applications not only shortens the search time, simplifies the solution process, but also produces answers to once"no feasible solution problems".