自动化立体仓库拣选作业调度优化研究

发布时间：2018-03-02 09:07

本文关键词： 自动化立体仓库拣选作业调度优化 Petri网改进遗传算法 Flexsim仿真　出处：《重庆大学》2015年硕士论文　论文类型：学位论文

【摘要】：企业将降低人工和材料作为第一利润源泉,扩大市场销售作为为第二利润源泉,而把降低物流费用作为第三利润源泉。随着生产制造系统的迅速发展,其系统的高效率与物流系统的低效率越来越不匹配。在企业生产制造过程中,自动化立体仓库在物料及成品的物流过程中扮演非常重要的角色,立体仓库是一个作业任务随机的复杂系统,若对自动化立体仓库的生产调度不合理,将极大地降低仓库的运行效率,从而影响整个企业的运行效率。当自动化立体仓库面临高强度与大批量快速作业时,作业若仅依照作业到达的顺序执行,忽略存在的冲突及共享资源竞争等情况,导致系统有效作业运行时间、堆垛机的运行效率无法达到最优,将很大程度上降低系统整体运作效率,所以对作业调度进行优化特别重要,所以本文针对作业调度优化问题主要完成了以下工作:①针对自动化立体仓库拣选作业调度的复杂性,本文通过对自动化立体仓库的组成及作业过程进行了深入分析,随着拣选作业任务的增加,调度组合呈指数增长,从物理的层面阐述了拣选作业调度优化问题,并提出拣选作业调度过程中可变时间因素概念。②针对自动化立体仓库拣选作业调度优化问题,论文采用赋量Petri网方法对其建立综合模型及拣选作业调度Petri网模型,提出作业调度优化的可变部分时间消耗及固定时间部分消耗的数学表达,从数学的层面研究拣选作业调度优化问题。③针对自动化立体仓库拣选作业调度面对调度组合呈指数增长难以快速优质寻优的问题,本文在拣选作业调度Petri网模型基础上,设计了基于Petri网与改进遗传算法结合算法对其进行优化调度,并采用拣选作业调度算例比较所设计的算法与标准遗传算法优化效果,验证了所设计的算法对于自动化立体仓库拣选作业调度优化效果更优。④针对拣选作业调度过程建立Flexsim仿真模型,设计3组实验进行对比分析,3组实验中巷道堆垛机采用不同的调度策略,通过对统计数据的分析验证所设计的基于Petri网与改进遗传算法结合算法对拣选作业调度优化过程的有效性,并对比标准遗传算法优化的优势。
[Abstract]:Enterprises will reduce labor and materials as the first source of profit, expand market sales as the second source of profit, and reduce logistics costs as the third source of profit. The high efficiency of the system and the low efficiency of the logistics system are more and more unmatched. In the manufacturing process of the enterprise, the automated warehouse plays a very important role in the logistics process of materials and finished products. Stereo warehouse is a complex system with random job tasks. If the production scheduling of automated warehouse is not reasonable, the efficiency of warehouse operation will be greatly reduced. This affects the efficiency of the whole enterprise. When the automated warehouse is faced with high intensity and large quantity of fast work, if the jobs are executed only according to the order of the jobs' arrival, the conflicts and the competition of shared resources are ignored, and so on. As a result of the effective operation time of the system, the efficiency of the stacker can not reach the optimum, which will greatly reduce the overall operational efficiency of the system, so it is very important to optimize the job scheduling. So this paper mainly completes the following work: 1 aiming at the complexity of sorting job scheduling in automated stereoscopic warehouse. This paper deeply analyzes the composition and operation process of automated three-dimensional warehouse. With the increase of job picking tasks, scheduling combination increases exponentially. The optimization problem of picking job scheduling is expounded from the physical aspect. The concept of variable time factor in the process of sorting job scheduling is proposed. 2. Aiming at the optimization problem of sorting job scheduling in automated stereoscopic warehouse, a comprehensive model and a Petri net model of picking job scheduling are established by the method of Petri net. The mathematical expressions of variable partial time consumption and fixed time partial consumption for job scheduling optimization are presented. This paper studies the optimization problem of picking job scheduling from the mathematical level. 3 in view of the problem that the scheduling of picking jobs in automated stereoscopic warehouses is difficult to find optimization quickly and high quality in the face of exponential growth of scheduling combinations, this paper is based on the Petri net model of picking job scheduling. The optimal scheduling algorithm based on Petri net and improved genetic algorithm is designed, and the optimization effect of the algorithm is compared with that of standard genetic algorithm by picking job scheduling example. It is verified that the designed algorithm is more effective in optimizing the scheduling of picking jobs in automated warehouse. 4. The Flexsim simulation model is established for the scheduling process of picking jobs. Three groups of experiments were designed to compare and analyze the different scheduling strategies used in the tunnel stacker in three groups of experiments. The effectiveness of the proposed algorithm based on Petri net and improved genetic algorithm is verified by analyzing the statistical data, and the superiority of the standard genetic algorithm is compared with that of the standard genetic algorithm.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F274

【相似文献】