粒子群蚁群结合算法在车辆调度问题上的应用研究

发布时间：2019-01-22 15:13

【摘要】：随着科学技术的日益发展以及互联网在人民大众生活中的普及,电子商务是我们日常生活中一个重要的组成部分,而作为电子商务最关键的的环节——物流配送,也展现出具大的研究价值。据科学统计,在整个物流系统中,商品的配送成本占据总成本的60%左右,而车辆调度问题是商品物流配送中最核心的问题。。因此,对车辆调度问题的研究,不仅在学术理论上具有很大的研究必要,而且还有相当大的现实基础。本文是分别使用蚁群算法、粒子群算法和二者结合的算法对建立的带时间窗的车辆调度模型进行分析和求解,最终得出结合算法在优化性能上优于单纯的使用某一种算法。本文所做的主要工作包括:一、分别介绍了蚁群算法和粒子群算法,对它们的数学模型进行分析。蚁群算法采用正反馈机制,蚁群算法的实现原理是蚁群中所有个体在寻找食物时借助其分泌的化学物质信息素不断进行数据交互和传承,实现蚁群内部的相互合作,更容易得出最优解。但是蚁群算法收敛速度慢、计算时间长,而且容易过早的陷入局部最优,出现停滞现象。粒子群算法具有简单、容易实现、参数较少、收敛速度快的优点,在一个搜索空间内,所有的粒子会根据粒子自身历史的最优解和种群中粒子的全局最优解共同决定其飞行方向。所以本文将蚁群算法和粒子群算法相结合,首先使用粒子群算法求出最优解,将求出的最优解信息作为蚁群算法的初始信息素,再利用蚁群算法求出最终的最优解。二、本文在现代物流技术基础上,建立了一个带时间窗的车辆调度模型,对车辆调度模型从理论和解决方法上进行了深度的研究。然后分别用蚁群算法、粒子群算法、融合算法寻找最优解,通过matlab仿真,得到实验数据,通过对实验数据进行分析得出结合算法相比于原来的两种算法在寻找最优解时更高效。
[Abstract]:With the development of science and technology and the popularization of the Internet in the people's life, electronic commerce is an important part of our daily life, and as the most crucial link of e-commerce, logistics distribution, It also shows great research value. According to scientific statistics, in the whole logistics system, the distribution cost of commodities accounts for about 60% of the total cost, and the vehicle scheduling problem is the core problem in the distribution of commodity logistics. Therefore, the study of vehicle scheduling problem is not only necessary in academic theory, but also has a considerable practical basis. In this paper, the ant colony algorithm, particle swarm optimization algorithm and the combination of the two algorithms are used to analyze and solve the vehicle scheduling model with time window. Finally, it is concluded that the combined algorithm is superior to a single algorithm in optimization performance. The main work of this paper is as follows: first, the ant colony algorithm and particle swarm optimization algorithm are introduced, and their mathematical models are analyzed. Ant colony algorithm uses positive feedback mechanism. The principle of ant colony algorithm is that all individuals in ant colony interact and transmit data with the help of chemical pheromone they secrete while searching for food, so as to realize mutual cooperation within ant colony. It is easier to get the optimal solution. However, ant colony algorithm has the advantages of slow convergence rate, long computation time, and premature fall into local optimum, which leads to stagnation. Particle swarm optimization (PSO) has the advantages of simple, easy to implement, few parameters and fast convergence. All particles determine their flight direction according to the historical optimal solution of the particle and the global optimal solution of the particle in the population. Therefore, this paper combines ant colony algorithm and particle swarm optimization algorithm, first uses the particle swarm optimization algorithm to find the optimal solution, takes the obtained optimal solution information as the initial pheromone of the ant colony algorithm, and then uses the ant colony algorithm to find the final optimal solution. Secondly, on the basis of modern logistics technology, a vehicle scheduling model with time window is established, and the theory and solution of vehicle scheduling model are studied in depth. Then the ant colony algorithm, particle swarm optimization algorithm and fusion algorithm are used to find the optimal solution. Through matlab simulation, the experimental data are obtained. Through the analysis of the experimental data, it is concluded that the combined algorithm is more efficient than the original two algorithms in finding the optimal solution.
【学位授予单位】：湖北工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U492.22;TP18

【相似文献】