基于混合蚁群算法的动态JSP研究与仿真

发布时间：2019-04-11 10:47

【摘要】：在新的世纪,制造企业之间的竞争日益激烈,现代生产正朝着“品种多样、批量变小注重交期、减少库存”的方向发展。车间是制造企业的物化中心和信息集散地,车间生产调度是企业快速响应市场需求的关键。车间调度问题既是实际生产中的一个重要问题,也是一个典型的NP-hard问题,而蚁群算法非常适合该类问题的处理。采用高效的优化调度技术能够显著提高设备利用率和准时交单率,降低库存及成本。本文应用融入人工免疫原理的混合蚁群算法求解车间调度问题,其主要内容如下： 1)介绍课题的来源、背景、研究目的和研究意义。叙述了调度和排序的相关概念,重点介绍了制造企业车间调度的重要性、目前存在的问题以及解决方法。 2)设计了将人工免疫原理融入基本蚁群算法的融合策略。首先详细介绍了基本蚁群算法的思想来源、基本模型、实现步骤和程序流程,阐述了算法的自身优势和目前的应用领域、算法可以改进的方面和改进途径。然后介绍了人工免疫基本原理、实现步骤和自身特点。最后,根据基本蚁群算法和人工免疫原理各自特点将人工免疫原理融入基本蚁群算法形成混合蚁群算法。在混合算法的前过程应用人工免疫原理生成初始信息索分布,初始信息素启发蚁群算法继续寻优,在提出了这样的融合策略之后,详细设计了混合算法的实现流程,以求实现更强的优化能力。 3)利用混合蚁群算法研究静态作业车间调度问题。首先详细介绍了车间调度的基本概念、特点和分类。其次明确了静态作业车间调度的研究属性,在此基础上建立了静态作业车间调度问题的数学模型和析取图模型,将它们与混合蚁群算法模型准确对应,使车间调度问题形成了适介混合蚁群算法的自然表达。应用混合蚁群算法对模型求解,设计出了节点访问规则,给出了详细的求解步骤和算法流程。最后,以文献中的实例验证了混合蚁群算法求解静态作业车间调度问题的可行性和有效性。 4)利用混合蚁群算法研究动态作业车间调度问题。阐述了动态作业车间调度的基本概念和属性,明确了静态和动态车间调度的本质区别,根据动态作业车间调度的关键问题分析设计出动态问题的调度策略,建立了基于事件驱动的仿真模型。最后,对文献中的实例求解,将求解结果和文献中的结果对比,得出混合蚁群算法在求解动态作业车间调度问题时具有较强的优化能力。 5)对全文进行总结并展望进一步的研究工作。
[Abstract]:In the new century, the competition among manufacturing enterprises is increasingly fierce, and modern production is developing towards the direction of "diversity of varieties, decreasing batch size, paying attention to delivery period and reducing inventory". Workshop is the physical and chemical center and information distribution center of manufacturing enterprises, and workshop scheduling is the key for enterprises to respond rapidly to market demand. Job-shop scheduling problem is not only an important problem in practical production, but also a typical NP-hard problem. Ant colony algorithm is very suitable for dealing with this kind of problem. Using efficient optimal scheduling technology can significantly improve the equipment utilization and on-time delivery rate, reduce inventory and cost. The main contents of this paper are as follows: 1) the source, background, purpose and significance of the research are introduced in this paper. The main contents of this paper are as follows: 1) the source, background, purpose and significance of the research are introduced in this paper. The related concepts of scheduling and scheduling are described, and the importance, existing problems and solutions of workshop scheduling in manufacturing enterprises are emphatically introduced. 2) A fusion strategy is designed to integrate the artificial immune principle into the basic ant colony algorithm (ACA). Firstly, the thought source, basic model, realization steps and program flow of the basic ant colony algorithm are introduced in detail. The advantages of the algorithm and the current application fields, the aspects of the algorithm can be improved and the way to improve the algorithm are described in detail. Then introduces the basic principle of artificial immunity, the realization steps and its own characteristics. Finally, according to the characteristics of basic ant colony algorithm and artificial immune principle, the artificial immune principle is incorporated into basic ant colony algorithm to form a hybrid ant colony algorithm. In the process of the hybrid algorithm, the artificial immune principle is applied to generate the initial information distribution, and the initial pheromone inspires the ant colony algorithm to continue the optimization. After this fusion strategy is put forward, the implementation flow of the hybrid algorithm is designed in detail. In order to achieve stronger optimization ability. 3) the hybrid ant colony algorithm is used to study the static job shop scheduling problem. Firstly, the basic concept, characteristics and classification of job shop scheduling are introduced in detail. Secondly, the research attribute of static job shop scheduling is defined. On this basis, the mathematical model and disjunctive graph model of static job shop scheduling problem are established, and they are exactly corresponding to the hybrid ant colony algorithm model. So that the job shop scheduling problem forms a natural expression of the suitable-mediating hybrid ant colony algorithm (HACA). The hybrid ant colony algorithm (HACA) is used to solve the model, and the node access rules are designed. The detailed solution steps and algorithm flow are given. Finally, the feasibility and effectiveness of hybrid ant colony algorithm (HACA) for static job shop scheduling problems are verified by an example in the literature. 4) the hybrid ant colony algorithm is used to study the dynamic job shop scheduling problem. This paper expounds the basic concepts and attributes of dynamic job shop scheduling, clarifies the essential difference between static and dynamic job shop scheduling, and designs the scheduling strategy of dynamic job shop scheduling according to the analysis of the key problems of dynamic job shop scheduling. An event-driven simulation model is established. Finally, compared with the results in the literature, the hybrid ant colony algorithm has a strong optimization ability in solving the dynamic job shop scheduling problem. 5) summarize the full text and look forward to the further research work.
【学位授予单位】：西安工业大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH166;TP18

【相似文献】