A Study On The Issue Of Vehicle Route Optimization Upon Ant Colony Algorithm, Artificial Fish Swarm Algorithm And Hybrid Swarm Intelligence Algorithm

Logistics distribution is directly linked to customer in logistics activities. In the cost of logistics, distribution costs account for a very high percentage. Vehicle routing greatly influence logistics services, costs and benefits in delivery. It is an important area of study for us to use a scientific and rational approach for optimizing the vehicle routing. So, vehicle routing problem with Time Windows is current focus on study.Modern intelligent optimization algorithm including taboo search algorithm, simulated annealing algorithm, ant colony algorithm and artificial fish swarm algorithm etc., the algorithms was for the delivery vehicle routing problem solving and optimizing provides new tools. Vehicle Routing Problem with Time Windows is a NP-hard problem. In this paper, the model of VRPTW is built, Then a hybrid optimization algorithm to resolve optimization of Logistics Distribution Routing problem is proposed in this paper. Based on the defects of AS(ant system) in this paper, firstly it improves through the inspiration of the initial solution, updated the pheromone on the choice of strategy and probability of the transfer; A swarm degree in the artificial fish swarm algorithm is used in ant colony algorithm. During the initial process of the optimization, the swarm degree plays the main role to guide the ants to search the new path randomly, which makes the algorithm have the stronger periodicity searching ability. To solving long stagnant, easy lead to transfer coefficient of Ant algorithm, the concept of transfer coefficient is proposed, which can improved ant colony optimization calculation speed, thus enhancement algorithm to traverse optimization ability. Experimental results show that the optimal or nearly optimal solutions to the logistic distribution routing can be quickly obtained by the hybrid optimization algorithm.Then it bases on the examples of problems in the Solomon's R101 data, using object-oriented C++ language to prepare a program to Hybrid swarm intelligence algorithm that is calculated verified, confirming the feasibility and effectiveness of it; at the same time, some of Solomon data is verified, compared with other heuristics the results showed that the Hybrid swarm intelligence algorithm is superiority; finally, the parameters of the Hybrid swarm intelligence algorithm is comparatively analysis, and then it Discusses the optimal combination on the parameters.