A Study On Approach Optimization For Logistics Distribution Based On Hybrid Algorithm System Of Ant Colony And Artificial Immune

As the globality of China,s logistics industry as well as the objective conditions of constantly changing, China,s modern logistics industry has been growing rapidly. First, the issue of logistics overview on the origin of logistics and related concepts and the development of the trends of domestic logistics and overseas logistics, including the concept and characteristics the third parties and the fourth parties and also made a brief introduction about Modern Logistics Information Systems and Technology. Analyze conventional means of implementation of the logistics system and a number of important issues of logistics links which often encountered in the planning and designing. This paper also analyzes and explores several important issues about logistics industry, and gives some solutions. Simulation results show a reference value. The second chapter describes the basic principles of the ant arithmetic, and makes a detailed description of the ant behaviors which is involved in a relative arithmetic, improves the ant arithmetic according to distribution of the unique characteristics and uses this way to solve the problems of the path optimization algorithm of logistics and distribution. Ant algorithm is the principal basis of the principle of positive feedback information and an organic combination of heuristics,ηij ant algorithm is the key of the design. When the group is too large, it is very difficult to find the optimal solution in a short time, and if the information produced by a random path changes too fast, it is very easy to happen that the search stagnated. In order to control the rate of the information changing, a new way that choosing the next visited city has to be used. This method made the full-rate"γ"increase as the number of vehicles which traverse the city. It could enhance the diversity of the path chosen so as to avoid falling into the optimum. Pheromone updated strategy is one of the key steps in the ant algorithm. Rapid updating of information will lead to a local optimal algorithm or even fall into stagnation; on the other hand, slow updating of information will slow the pace of convergence and can not search for the most optimal route. This paper uses two methods to updating the information: one of them is the global information-updated; another is partial information-updated. The simulation results show that under the same conditions as the basic parameters"α", the basic ant algorithm has the same result with the improved ant algorithm. But the improved algorithm greatlyreduced the number of the evolution, and the rate of convergence has been increased by nearly 50%. Integrate this improved algorithm and 3-opt algorithm will greatly improve the accuracy of the solution. It also showed a workable solution to solve large-scale VRP problems.The third chapter describes the basic idea of the artificial immune algorithm (AIA) and ant algorithm. AIA has very excellent overall-situational random search capability, but a low capability in using the information in feedback system. When it comes to solve a certain extent inactive redundant iteration, it has a very low efficiency. The ant algorithm has parallel and distributed searching capability through the ways that accumulate the information and update convergence to the optimal path. But the initial lack of information will cause a slower searching pace. The chapter learned of the proposed algorithm artificial immune algorithm and ant algorithm, it uses artificial immune algorithm to create a distribution of information and uses ant algorithms to find optimal solutions, so that this way will complement each other's advantages and have significantly improved in convergence and optimization. The mixed method based on ant algorithm and artificial immune algorithm, and the basic idea is to take full advantage of rapid process randomness and global convergence of AIA algorithm in order to find a feasible solution, and after the process, using ant colony algorithm. That means ant colony algorithm used the former feasible result of AIA to create initial information-distribution, and then take full advantage of parallel algorithms, positive feedback, solution to improve efficiency. The simulation results show that this new method is superior to ant algorithm in convergence and superior to artificial immune in capacity optimization. This new method is better in the speed of convergence and the capability of optimization.In this paper, we discuss logistics and delivery route optimization problems using evolutionary algorithm (improved ant algorithm) and mixed method of ant algorithms and artificial immune algorithm, and use the example of simulation amount the ant algorithm, artificial immune algorithm and the mixed method. It has achieved good results that finding optimal path to the logistics of the delivery vehicle for the goal line. In the future, we will focus on combining ant algorithms with the other algorithms to find the optimal distribution logistics.