The Design Of Dynamic Vehicle Scheduling System Based On Ant Colony Algorithm

In logistics field, vehicle scheduling is a very important problem, which with great theoretical and realistic significance. Vehicle’s distribution route reasonable or not has great influence on the distribution speed, operating costs and economic benefits. Therefore adopt a scientific and reasonable method to determine the distribution line is a very important work in the logistics activity. Through optimizing the complex distribution system, we can improve the utilization rate of resources significantly.Due to the research of dynamic scheduling problem in logistics industry mostly focused on software research, the platform to be used for demonstration, planning, designing and simulation experiment is relatively lack, therefore, this paper based on the ant colony algorithm, Through two kinds of improved ant colony algorithm to analysis and simulate the two types of dynamic vehicle scheduling problem which were described in the paper. Besides, by using CC2430 chip as the controller of smart car, forming the ZigBee wireless communication network through z-stack, this design has set up a dynamic scheduling simulation demo system. Through this system we can intuitively felt the whole process of dynamic vehicle scheduling, for those systems similar to logistics, the system can also provide some beneficial help for their planning designing, improvement, simulation and forecast.In view of dynamic vehicle scheduling problem, this paper mainly does the following several aspects:1. At first, this paper describes the vehicle scheduling problem, and classifies the vehicle scheduling problem according to influence factors of VRP. Based on this, this paper has given two common mathematical model of vehicle scheduling problem. Then describes the dynamic vehicle scheduling problem, and introduces two methods to deal with the real-time information in the dynamic vehicle scheduling process.2. This paper introduces the basic principle of ant colony algorithm and its characteristics, and analyzes its advantages and disadvantages. Aimed at the deficiency of basic ant colony algorithm, this paper improves the two kinds of ant colony algorithm: adaptive ant colony algorithm and max-min ant colony algorithm, then introduces the principle and the solve steps of the two kinds of algorithm.3. Hardware design of the system:At first introducing the design idea and overall framework of this system, then describes the implementation process and the principle of each module of the smart car that simulating delivery vehicle in the scheduling system in detail.4. This design solves the dynamic VRP system of collecting or delivering goods with non-full cars by using adaptive ant colony algorithm and solves the dynamic VRP system with time window by using max-min ant colony algorithm, then does some simulation analysis for the two dynamic scheduling systems by using MATLAB. At last, through the hardware system, we had demonstrated the process of dynamic scheduling which had been researched in this paper.