【作者】 王亚辉；

【导师】 夏芳臣；

【作者基本信息】 南昌大学 ， 机械制造及其自动化， 2008， 硕士

【摘要】 随着市场竞争的加剧,企业生产正朝着多品种、小批量方向发展。在离散型企业尤其是机械行业,生产的产品品种多、批量小,生产组织工作复杂,特别是企业的生产作业计划安排工作难度很大,计划编制往往凭主观经验,计划的及时性、应变性差,导致产品生产周期长,在制品占用量大,机床利用效率低等,从而影响了多品种、小批量生产的经济效益。因此,研究离散型企业车间调度问题,不仅具有较大的理论意义,而且还有相当大的实用价值。本文通过对离散型企业及其车间调度问题的详细分析和对调度理论进行深入的研究,提出了具有柔性加工路径的车间作业调度模型。这里的柔性是指设备安排的柔性,即工件的设备加工路径不是固定的和预先确定的,而是柔性的。系统采用遗传算法与启发式规则相结合的方法解决此类调度问题,具体方法是把这类问题的求解分解为两个优化本原问题的求解,一个是利用遗传算法为工序选择资源,一个是利用启发式规则决定机器上工序的加工次序和开始加工时间。另外,系统充分考虑了生产中的不确定性因素,采用了周期性调度和再调度相结合的策略,扰动发生后,遗传算法能够根据动态数据库所提供的最新任务数据,快速产生新的优化方案。同时,为了提高系统的可视度和易读性,将虚拟现实技术应用其中,对调度系统中车间动态信息的实时监控做了初步的尝试。利用虚拟现实建模语言VRML实现了车间设备和车间布局简约模型的构建以及设备运行状态的表达。最终,开发了一个将静态计划和动态调度相结合,包括离线计划和动态调度的作业车间调度原型系统。更多还原

【Abstract】 As market competition intensifies, enterprises are going towards the direction of multi-varieties and small-batch production. In discrete enterprises, especially those in the machinery industry, the case with multi-varieties and small-batch production and complex production organization, will cause long production period, mass work in process, inefficient use of machine tools and so on, thus affect the economic benefits of multi-varieties and small-batch production. The complexity of production organization particularly embodies in the great difficulty of enterprises’ production planning, plan formulation based on subjective experience, and the poor timeliness and flexibilities of the plan. The research on Job-shop scheduling in the discrete manufacturing enterprises has important theoretical and practical significance.Based on the deep research of scheduling theory and a detailed analysis of the discrete manufacturing enterprises and its scheduling problems, a model for Job-shop scheduling with alternative operation routes is put forward in the paper. The alternative operation routes refer to flexibility of machine arrangement, that is, one operation arranged on which machine is not fixed and predetermined but alternative. This paper adopts the method combining genetic algorithm and heuristic rules to solve the Job-shop scheduling problem with alternative operation routes. So the solution to the problem is changed into optimizing two sub-problems, one is using genetic algorithm to select resources for process and the other is adopting heuristic rules to determine the sequence and start time of process on a machine.Moreover, by fully considering the uncertain factors in the production, a strategy by combining periodic scheduling and re-scheduling is adopted. According to the updated data in the dynamic database after disturbance occurs, the genetic algorithm can generate optimized production planning in response to the changes of the Job-shop. Meanwhile, in order to improve the visibility and legibility, a preliminary attempt has been done for the real-time monitoring of the workshop dynamic information in the scheduling system, with the application of virtual reality technology. By using of the virtual reality modeling language VRML, a simple model of the workshop equipment and workshop layout is set up and the expression of the equipment running station is done. Finally, a Job-shop scheduling prototype system, which combines the static scheduling and dynamic control and includes planning offline and dynamic scheduling, is developed.更多还原