| The acceleration of the process of world economic integration and the continuous advancement of science and technology makes modern logistics, which is known as "the third profit source", receive more and more concern. For businesses, the cost of logistics takes up a high proportion of the total costs. In order to survive, as well as develop in the competitive environment, reducing the cost of logistics is essential. Distribution, as an important part of the logistics, plays a key role in lowering the costs. Concerning reducing this part, the focus is to arrange the vehicle delivery path, that is, to do the optimization work.This article aims to optimize the Y’s current vehicle route. It will ensure the shortest delivery route and the highest load factor under the primise of meeting customers’needs in time, which, in turn, will minimize the company’s transportation costs. After comparative analysis of various algorithms, I have chosen the ant colony algorithm to solve the problem. Since the ant colony algorithm itself is easy to fall into local optimum, and the search time is relatively long, this article has done some improvements, combining with the actual situation of the Company Y, to remedy the defects. The improvements contain three aspects, say, the state transition rules, the pheromone update rules and the route optimization. Then, it has applied the improved one to solve the practical problems, and ultimately required the optimal route through the experimental simulation. For the Company Y, it has enhanced the vehicle utilization, while reducing the vehicle delivery cost.Firstly, this article introduces the research background and the significance of it. Then outlines the main research content and gives the framework of the paper. Finally, it concludes the method of this study.Secondly, it summarizes the research status quo at home and abroad on the VRP. It presents and comparatively analyzes the advantages and disadvantages of different algorithms, to determine which algorithm is to be used in this article. Then it introduces the research status of the ant colony algorithm. The understanding of its development process and applications points out the right direction for the application of the next improvement. In the following, it gives the general mathematical model of VRP. And then adds the vehicle load factor to the optimization goal in accordance with the actual business problems. Besides, it makes a corresponding adjustment to the general model, taking the vehicle load factor into account to get a new one. After that, it shows the basic principle of ant colony algorithm, as well as the basic model, using the example of solving the TSP problem. Moreover, it sums up the characteristics of the ant colony algorithm, laying the foundation for improvements in the next section.Fourthly, it compares the difference between ant colony algorithm for solving TSP and VRP problems. In spite of that, it also proposes three improved strategies on the ant state transition rules, the pheromone update rule and the optimal path. And then it displays the implementation steps and presents the parameters configuration for the improved ant colony algorithm.The crucial part of this article is the actual application of improved ant colony algorithm. It begins with an overview of the current situation of the research enterprise Y, and then illustrates the collected data that is required. In the following, it corresponds Company Y’s problem to the VRP model, and uses the improved algorithm to solve it. Through the experimental simulation output data, it shows that the total distance of the improved vehicle delivery path is shorter, the vehicle load factor is higher and the transportation costs is lower than before.Last but not least, it generalizes the whole article, and puts forward the inadequacies, which pointes out the direction for the next phase of the study. |
PDF Full Text Request