【作者】 王欢；

【导师】 范国祥； 赵磊；

【作者基本信息】 哈尔滨工业大学 ， 软件工程， 2015， 硕士

【摘要】 在现代的大型企业中,采用不同信息系统对于企业的生产进行管理已经不再是一个陌生的管理方式了。然而在不同信息系统之间,由于操作人员不同导致了相同数据信息在不同信息系统之间存储的数据格式不一致这一问题也越来越显著。这种情况下就会导致企业的生产部门、采购部门和财务部门的数据冲突问题,这将大大的增加企业的生产成本,在大型企业中就会造成严重的资源浪费和数据孤岛的问题出现。因此,解决相同数据在不同系统中的数据集成问题就成为了需要切实解决的一个重要任务了。本文首先介绍了主数据管理平台的相关名词和国内外研究现状,阐述了当前国内外从事主数据研究的公司团体和研究成果。其次通过需求分析,阐述系统的功能,并给出了解决不同系统中相同数据的分发与数据接收的技术方案。在系统设计章节中,以系统整体的设计为基础,详细阐述了系统各个模块的设计思路和时序设计。在系统实现章节中,对于系统的实现细节进行详细阐述,并对数据集成过程中的开源框架Mule的使用进行论述。本项目主要实现的功能有:主数据模型管理功能,主数据管理功能和主数据集成功能。其中主数据模型管理模块是系统的核心功能模块,在模型创建的同时可以创建模型上的校验信息,进而保证了模型上数据的唯一性、准确性和长期有效性;在通过审核的主数据模型上可以创建主数据,主数据信息必须符合模型的校验规则,否则主数据无法进入到系统中;在完成主数据审核操作后,便可以进行主数据的集成操作,将主数据发送到外部系统中并进行日志记录。本系统实现的软件系统已经在中国海洋石油总公司正式上线稳定运行,客户反馈使用效果良好。由于系统采用Java语言开发,因此在系统的移植和部署方面较为方便,也为大型企业解决数据问题提供了一个良好的解决方案。更多还原

【Abstract】 In modern large enterprises, the use of different information systems for the production of enterprises management is no longer a strange way of management. However, in different information systems, the problem is more and more obvious because of the different data formats which are stored in different information systems. This will lead to problems of data conflict in enterprise production department and purchasing department and finance department, which will greatly increased production costs of enterprises, in the large enterprises, it will result in serious waste of resources and data silos. Therefore, it is an important task to solve the problem of data integration among different systems.Firstly, this paper introduces the research status of the master data management platform, and describes the domestic and foreign research results of the research. Secondly, describing the function of the system, and giving the technical scheme of the distribution and data receiving of the same data in different s ystems. In the system design, the system design is based on the whole system, and the design idea and the design of the system are described in detail. In the system implementation section, the details of the implementation of the system are described in d etail, and the use of open source Mule framework in the process of data integration is discussed.The main functions of this project are: the master data model management functions, master data management functions and the master data integration functions. The master data model management module is the core module of the system, created in the model also can create a model checking information, so as to guarantee the model data uniqueness, accuracy and long-term effectiveness; in through the examination of the master data model can create a master data, master data must conform to the model validation rules, otherwise, the master data can not be entered into the system; after the completion of the operation of the master data audit, they can master data integration process, master data is transmitted to the external system and logging.The system has been implemented in the software system of China National Offshore Oil Corporation, operating stable on the official line, the feedback of using effect is good. Because the system uses Java language development, it is more convenient for the system’s transplantation and deployment, so it provides a good solution for large enterprises to solve the data problem.更多还原