Design And Implementation Of Decision Support System Based On Data Warehouse

Government and branches of the government informationization has accumulated a great deal of grassroots data from all industries after many years of building. But the data scattered and the centralized concordance now, could not meet the government revenue decision-making and the data analysis need. With the government informationization improved further and the rapid development of data warehouse technology, data storage technology used in the work to government has become reality, because it can support decision making and the data analysis, improve the efficiency and the quality of the work. This topic is produced in this kind of situation.First, this article introduced the research background of the topic, the current level of government informationization and the situation of the government informationization on the integration. Then it made a detailed study in the basic concept of the data warehouse, the structure of the data organization as well as the design method and steps of the data warehouse. This article based on current popular Oracle 10g database as a platform and the practical situation, put forward a data warehouse application plan on all industries for the government management applications subjects, and focused on the study of the government system in data warehouse methodologies and the steps, as follow:Firstly, design the models of the government data warehouse based on the analyses and special hierarchy of government data. Secondly, with the OWB tool, described the process of the data extracted, transformed and loaded including time-tag schema and trigger schema, for integrating original data to the data warehouse based on models. Additionally, came up with a design incremental ETL plan to deal with the deleted data. Thirdly, used Oracle OLAP Analytic Workspace (AW) analyses the data multidimensionally in the data warehouse, and presented them by reports and graphs. Finally, for the requirement of the comprehensive data analysis in the government DSS and the realization of mutil_subject ad_hoc, an approach extracting data from several subjects in a data warehouse to create integration analysis report is proposed to solve the problem of Oracle report tool to some extent.