论文标题:决策支持技术在企业ERP中的研究与应用
Research and Application of Decision Support Techniques in ERP
论文作者
论文导师 郭禾,论文学位 硕士,论文专业 计算机系统结构
论文单位 大连理工大学,点击次数 1097,论文页数 60页File Size4460K
2007-12-01论文网 http://www.lw23.com/lunwen_180092/
Data Warehouse;; OLAP;; Data Mining;; SQL Server 2005
信息技术的广泛发展，使得企业信息化建设的步伐越来越快。企业已经不满足于基于业务的简单数据处理，而进一步提出了从数据中发现对其发展有指导意义的知识，即商业智能方面的需求。决策支持的概念正是针对这一需求而提出，它将企业中庞大繁杂的数据转换为知识，进而辅助企业经营决策，甚至自动生成商业决策。因此，决策支持已经成为企业竞相追逐的目标。 决策支持技术主要包括数据仓库、OLAP(联机分析处理)、数据挖掘三项。数据仓库整合了分散在企业各个方面的数据，为提供决策准备了大量的有效数据；OLAP提供了从不同的角度分析数据的功能，能为企业经营者提供企业数据的全局视图；数据挖掘从企业数据中找出隐藏在数字背后的有价值的模式和关联，为企业经营者制定决策提供有参考意义的知识。 论文以辽宁某印染股份有限公司的实际需求为背景，合理处理历史数据，通过适当的数学模型对数据进行整合，最后使用决策支持的关键技术，构建了历史销量分析模型和新品种收益度预测模型，并将这些模型嵌入到该公司ERP系统中。 企业经营者可以使用历史销量分析模型，分析产品的淡旺季、掌握不同客户对不同品种的需求量，从而辅助经营者制定正确的营销战略，如定期促销、联系客户等等。新品种收益度预测模型可用于企业开发新品种之前，对其销售前景、收益程度进行预测，为制定是否开发该新品种的决策提供指导。 模型解决了企业经营者制定决策时，仅凭借经验判断，而缺少数据支持这一问题。实践证明，使用模型之后，经营决策的正确率得到很大提高。可见，模型为企业制定正确的发展战略提供强有力的支持。
Enterprise information construction is growing faster and faster due to rapid development of information technology. Enterprises not satisfying with the business-based simple data processing propose a business intelligence demand for development guidance by digging knowledge out of data. The concept of decision support is just for this demand. Decision support can transform the huge and complicated data into knowledge so as to assist the decision making of business management, even to make business decisions automatically. So decision support has become an aim enterprises competing for. The technology of decision support includes Data Warehouse, OLAP(On Line Analytical Processing) and Data Mining. Data Warehouse integrates the data distributing in all parts of enterprise, and prepares a large scale of effective data for decision making. OLAP can analyze the data from different points and provide global view of enterprise"s data for its managers. Data Mining digs valuable patterns and relations behind figures out of enterprise"s data, so that it can offer some reference for the managers. This paper is based on the practical demand of one printing and dyeing company in Liaoning province. It properly processes the historical data, and integrates the data with a suitable mathematical model. Finally by the key technology of decision support, it constructs an analysis model of historical sales and a prediction model of new breed profit and, embeds theses models into the company"s ERP system. The managers of the enterprise can use the analysis model to analyze dead or peak season of a breed to grasp information about different breeds" needs among different customers so as to make correct sales strategies, like periodical promoting sales, customer contact etc. The prediction model can be used before development of a new breed to predict sales prospect and profit so as to help decide weather or not to invest in this new breed. The models solve the problem that managers make decisions only by their own experience rather than data support. Results show the correctness of decision making improves greatly after using these models which proves that the models provide a powerful support for the company to make right development strategies.
|
