基于多Agent的物流配送车辆智能调度研究

发布时间：2018-03-13 01:14

本文选题：物流配送　切入点：车辆调度　出处：《集美大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着市场经济的不断发展,物流作为“第三利润源”对企业各项经济活动的影响和作用日益凸显,成为企业竞争的重要领域,也越来越引起企业的重视。在物流研究领域内,配送车辆的调度问题也一直是各学者和企业研究的焦点。车辆调度是一个涉及多个目标、多个约束条件的复杂组合优化问题,配送车辆的安排和线路规划的合理与否对配送时间、配送成本以及整个配送的质量影响很大。如何能利用科学合理的方法来规划配送线路,关系到物流系统优化,是车辆优化调度的核心部分,同时也是一个难题。近年来,各种信息技术的快速发展,在一定程度上为物流配送车辆的智能调度奠定了基础。Agent作为人工智能发展领域的典型代表,具有智能性、分布式、动态性等特点,通过感知自身状态和外界环境的变化,做出相应的动作。而多Agent组成的系统中,各Agent皆有一定的独立功能,且Agent之间的结构关系是动态可调整的,不同功能的Agent组成的系统可以通过通信和协作完成复杂的运作,具有良好的适应性、自组织性以及协调性,多Agent系统的这些特点使其在解决物流配送中的车辆调度问题上非常适用。本文中,首先阐述了选题的背景、意义和相关国内外研究现状,在大量理论研究的基础上,设计了基于多Agent的车辆调度系统的结构模型,包括系统层、结构层和功能层,通过分析系统中Agent之间的通信和协作方式,阐明了实现车辆调度的过程。然后,针对结构模型中车辆调度Agent中所需封装的模型和算法进行研究,建立了车辆调度问题的数学模型,并根据模型的特点结合模拟退火算法进行求解,通过实例验证了模型和算法的有效性。最后,本文通过Anylogic仿真软件,基于设计的多Agent的车辆调度系统的结构模型进行仿真设计,并结合车辆调度问题的数学模型和算法进行了实现,通过实例以仿真结果展现出来,验证了Agent技术应用于车辆调度领域的可行性,具有一定的理论和现实意义。
[Abstract]:With the continuous development of market economy, logistics, as a "third profit source", is playing an increasingly important role in the economic activities of enterprises. It has become an important field of enterprise competition and attracted more and more attention of enterprises. Distribution vehicle scheduling problem has been the focus of scholars and enterprises. Vehicle scheduling is a complex combinatorial optimization problem involving multiple objectives and constraints. Whether the arrangement of distribution vehicles and route planning is reasonable or not has a great influence on the distribution time, distribution cost and the quality of the whole distribution. How to use scientific and reasonable methods to plan distribution routes is related to the optimization of logistics system. It is not only the core part of vehicle optimal scheduling, but also a difficult problem. In recent years, with the rapid development of all kinds of information technology, To a certain extent, it lays the foundation for the intelligent dispatching of logistics distribution vehicles. As a typical representative of artificial intelligence development field, Agent has the characteristics of intelligence, distribution, dynamics and so on. In the system composed of multiple Agent, each Agent has certain independent function, and the structure relationship between Agent is dynamically adjustable. The system composed of Agent with different functions can accomplish complex operation through communication and cooperation. It has good adaptability, self-organization and coordination, and the characteristics of multi-#en0# system make it very suitable to solve the vehicle scheduling problem in logistics distribution. On the basis of a large number of theoretical studies, the structure model of vehicle scheduling system based on multiple Agent is designed, which includes system layer, structure layer and function layer. By analyzing the communication and cooperation between Agent in the system, the process of vehicle scheduling is clarified. Then, the encapsulation model and algorithm of vehicle scheduling Agent in the structural model are studied. The mathematical model of vehicle scheduling problem is established, and the model is solved with simulated annealing algorithm according to the characteristics of the model. The validity of the model and algorithm is verified by an example. Finally, the simulation software of Anylogic is used in this paper. The structural model of the vehicle scheduling system based on the design of multiple Agent is simulated and designed, and the mathematical model and algorithm of the vehicle scheduling problem are implemented, and the simulation results are presented by an example. The feasibility of applying Agent technology to vehicle scheduling is verified, which has theoretical and practical significance.
【学位授予单位】：集美大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F252;U116

【参考文献】