Logistics Distribution Vehicle Routing Problem And Its Intelligence Algorithms Research

With the globalization of the world economy, more and more trade cooperation have being established between regions and countries. Also as the internet economy is developing swiftly and play a great role in economic growth, the development of our contemporary economy will enter a new era, namely the logistics times. The rational planning and scientific development of logistics transportation creates a new profit space of economic growth. At the same time, with the continuous development of global logistics technology, We will create a intelligent logistics system. The first problem need to be solved is the effective application of vehicle routing problem(VRP). The reasonable optimization of the VRP problem can improve the efficiency of transportation, reduce cost and increase economic benefit. So it’s very important to the whole logistics transportation arrangement. In this paper i focuses on the research of the vehicle routing problem in logistics distribution and try to use intelligent algorithm to solve it.In this paper, firstly, research the vehicle routing problem systematically and establish the mathematical model of the standard vehicle routing problem. The practical application of vehicle routing problem is more complicated, so giving a summarize of it to provide some help for the development of intelligent model of modern logistics.Secondly, for the ant colony algorithm and particle swarm optimization algorithm to solve the vehicle routing problem put forward some improvement ideas. We all know ant colony algorithm has been proved to be a promising method for solving complex optimization problems. It must be pointed out that, as a global search algorithm, ant colony algorithm can simulate the foraging behavior of ants and find out the optimal solution of the problem gradually. But there are still some shortcomings, such as long searching time, sensitive to parameters and slow convergence speed for large scale problems. Therefore, this paper proposes two ways to improve the ant colony algorithm.Number one, an improved ant colony algorithm based on genetic operator is proposed.Combining the advantages of genetic algorithm, performance crossover and mutation operation for each generation of ant colony and accepting new individuals based on the Metropolis criterion of simulated annealing algorithm. Then solving the TSP problem, a special case of VRP problem, to verify the effectiveness of the improved algorithm. On this basis, further research show that ants can take different objectives as the path selection principle in the process of optimization and can be divided into different groups. So number two an improved hybrid behavior ant colony algorithm is proposed.Designing four specific ant behaviors, selecting different way of ant behaviors to formdifferent improved algorithms to solve the vehicle routing problem. In the second place,Particle swarm optimization algorithm is a new iterative optimization algorithm. On one hand, it has the advantage of simple rules, few parameters, fast convergence speed, but on the other hand, it is easy to premature, poor local search ability, easy to deviate from the optimal solution and so on. An improved particle swarm optimization algorithm based on interaction between particles is proposed in this paper. Introducing the neighborhood topology and defining two new concepts lepton and hadron. Lepton update speed and position according to the individual extreme and group extreme.Hadron collisions with the global optimal particle to change speed and position. When the algorithm step to stagnation, we take particle decay to increase the population diversity. Use matlab to realize the simulated experiment of vehicle routing problem. In the end, points out the deficiencies of this paper and provides the blueprint of future research in this field.