The Application Study On Multidimensional Analysis In College-Level Office Automation System

In the face of rapid growth of massive data storage, traditional various management information systems (MIS) handling the structured data already can not meet the requirements that the decision-makers want to get the valuable information from the data analysis. Then the "Data excess, lack of knowledge" phenomenon occurred. The university management team urgently needs a strong data analysis tool which can convert the data to valuable management knowledge, to meet the analysis needs of policy makers, supporting their decision-making, so that the university’s overall competitiveness can be enhanced in the increasingly competitive modern education management system. The Information island issue is expected to be resolved with the development of Data Warehouse technology, the maturity and popularity of OLAP tools. It provides a comprehensive solution for multi-dimensional data analysis.The thesis gives comprehensive analysis for the office automation application status in College of Information Science and Engineering of the Northeastern University, and also gives comprehensive analysis for the decision-maker’s decision analysis requirements. The paper gives a proposal of Multi-dimensional analysis presented in the Institute of Office Automation System. The thesis describes Multi-dimensional analysis technology’s initial exploration and application in the field of higher education and office automation.Firstly, the thesis introduces the research background, meaning, contents, and the art of state of data warehouse and online analytical processing technology. Secondly, the thesis investigates the data warehouse and ETL techniques, gives deeply research on the data model, analysis methods and the architecture of OLAP. Meanwhile, multi-dimensional modeling mechanism is proposed based on ER model. Thirdly, we focus on the construction of ETL process which is the core process of establishing the data warehouse. Through in-depth study of ETL solutions, we use open source ETL tools to complete the data loading, cleansing and transformation. We achieve the initial loading of data marts. We also discuss using hand-coding and software tools to implement the data-mart’s incremental regular update scheme.Finally, the thesis describes in detail the management of metadata, the construction of star schemas and multi-dimensional data models, the release and regular update of multi-dimensional cube. Existing OLAP solutions are analyzed and compared, finally Cognos OLAP engine is selected as the solution. Using Analysis Studio as OLAP engine, we implement several multi-dimensional analysis operations such as dimension switching, the drilling, cutting, sorting, and we give a clear visual display using chart and form. We also describe how to integrate it into OA system.Through the application of this system, it has the certain role in promoting the construction of the College management information. It provides the scientific, convenient and quick decision-making information to the high-level decision makers. At the same time, if we apply it to other industries, it has considerable practical meaning.