【作者】 王景华；

【导师】 韩江洪；

【作者基本信息】 合肥工业大学 ， 计算机应用技术， 2014， 博士

【摘要】 制造执行系统是现代制造企业信息化的一个重要阶段,而生产调度问题是制造执行系统的研究核心,对于该问题的解决程度直接关系到企业的生产效率和活力。早在上个世纪初,人们就开始对该问题展开研究。但是由于生产调度问题的复杂性,目前该问题仍然存在着很多值得研究的要点。本论文从生产调度问题中的离散型生产方式和混合型生产方式两种不同的分类入手,分别从建模方法和求解方法两个侧重点上进行研究,并建立了基于M-Petri网的离散生产调度系统和基于遗传算法的混和生产调度系统。论文的主要工作如下：(1)探讨了生产调度问题中的关键研究内容,分析了离散型生产调度系统和混合型生产调度系统的构造及常见的调度目标,概括总结了常见的研究生产调度的建模方法和求解方法,研究各种方法的理论及应用；(2)针对于离散型生产调度系统,采用基于层次化和面向对象思想的M-Petri网对问题进行建模,然后根据其不同特点采用调度规则组合的优化方案,针对动态生产调度问题,提出建立调度规则知识库。根据实际生产的需要,通过选择不同的调度规则组成目标函数,实现调度动态化、多目标化。最后对一个具体实例进行分析,证明了该模型的动态性和有效性；(3)针对于混合型生产调度系统,选用遗传算法作为问题的求解工具,并设计了一种自适应遗传算法来弥补标准遗传算法中存在的缺陷,给出了自适应遗传算法在该调度问题中的应用步骤。然后,对订单型企业中常见的插单现象,给出了企业插单问题的处理流程,建立了插单问题的数学模型。最后,通过实例来进一步分析所设计的方法在混合生产型企业生产调度中的应用。更多还原

【Abstract】 Manufacturing Execution System is an important stage of modern manufacturing enterprise informationization, and production scheduling problem is the research core of manufacturing execution system, for solving this problem is directly related to the enterprise production efficiency and vitality. Early in the last century, people began to focus on this problem. But because of the complexity of the production scheduling problems, the problems still have many points worthy of study. In this paper, based on two different classifications of the production scheduling problem-the discrete production mode and the mixed mode-we study how to build the mathematics models and how to solve the related questions, and set up a discrete production scheduling system based on M-Petri net and hybrid production scheduling system based on genetic algorithm. The main work is as follows:(1) Discuss the key research content in production scheduling problem, analysis the structure of discrete production scheduling system and hybrid production scheduling system and common scheduling objectives, summarize the modeling method and the solving production scheduling way in common, and study the theory and application of various methods;(2) To the discrete production scheduling system, based on the optimization scheme, adopt hierarchical and object oriented M-Petri net modeling the problem. Then according to the different characteristics of the combination of scheduling rule, propose the establishment of scheduling rule knowledge base. According to the actual production needs, select different scheduling rules of objective function, and realize the dynamic, multi-objective scheduling model. Finally, a concrete example is analyzed, which proved the proposed model’s feasibility and effectiveness.(3) To the hybrid production scheduling system, select the genetic algorithm as the tool of solving problem, and design an adaptive genetic algorithm to make up the defects existing in the standard genetic algorithm, give the adaptive genetic algorithm in the scheduling problem. Then, on the orders of enterprises in the common insertion of a single phenomenon, give the insertion of a single problem, and establish the mathematic model of insertion for a single problem. Finally, further analyze the designed method in the production scheduling of hybrid production enterprises based on a example.更多还原