【作者】 钟远晖；

【导师】 严洪森；

【作者基本信息】 东南大学 ， 控制理论与控制工程， 2004， 硕士

【摘要】 本文运用.NET Framework技术对企业生产部门管理软件进行了研发,其中涉及到管理业务流程的软件实现和生产线调度问题求解算法的研究。在软件的设计阶段,首先分析生产管理中的业务信息流程的特点并将它们归类。通过这种方法,进一步将部门的功能分割成四个子功能模块集。在此阶段,确立了软件总体的层次结构和功能结构。在功能实体挖掘阶段,基于面向对象的思想,把实际的功能业务实体映射到数据库和软件结构中,确立了数据表格和软件模块的内容,通过表格间的联系和软件模块间的调用关系描述了业务实体间的联系。在各个功能模块实现的过程中,把每个模块的业务处理需求和运行环境需求结合起来,再结合软件设计模式技术确定功能模块间的调用方式,并最终在微软公司的.NET Framework平台下实现整个软件系统。其中,通过Web Service技术实现系统中的业务层功能,从而实现了用户界面与业务逻辑在物理位置上的完全分离,保证了软件的可扩展性。在生产线调度问题的研究中,运用组合遗传算法对生产线的调度问题进行求解,实现了计划和调度的同时优化。运算实验的结果证明,将两个独立的遗传算法(计划,调度)通过交替或串行的方式组合在一起,能满足计划与调度同时优化的要求。此外,研究中通过使用存储过程完成了制造企业冲压车间、装焊和油漆线的计划和调度的生成。在结论部分,总结了本文的研究内容,并分析了研究中有待进一步解决的问题。更多还原

【Abstract】 This thesis deals with the research and development of the application software for enterprise production management by using the Microsoft’s .NET Framework technology, including the realization of the software for business process management and the algorithm for solving the problem of production line scheduling.During the software design phase, the business process of production management are analyzed and classified. In this way, the function of the department is divided into four modules. In this phase, the structure of the software is established. In the phase of finding the functional entities, the department’s functional entities and the relations between them are mapped into the function of software’s modules and the relations between these modules with the help of the theory of object-orient (OO).During the phase of the software’s modules implementation, the modules’ requirements for business process and runtime environment are both included. The technology of design pattern is used in this phase to determine the calling methods among the modules. And the whole software system is implemented on the platform of the Microsoft.NET Framework. The functions on the business level are implemented by the Web Services technology. Thus, the physical locations of user interfaces and business logics are completely separated such that the expandability of the software system can be guaranteed. Combined genetic algorithm is used to solve the problem of production line scheduling in this research. The result of numerical experiments shows that combining two genetic algorithms for production planning and scheduling in an alternative or serial way can meet the requirements for optimizing the plan and scheduling simultaneously. Store procedures are used for production planning and scheduling for punching workshops, assembly lines and painting lines in manufacturing enterprise.In the conclusion part, the contents of this thesis are summarized and some problems for future research are put forward.更多还原