Design And Implementation Of Business Intelligence Systems For The Insurance Industry

The rapid development of Socio-economic and information technology provides a solid foundation and broad space for China’s rapid developing insurance industry, the competition between the companies become more and more intense. Extracting valuable information from the distributed mass data and rapid decision-making is the key to determine the competitiveness of the insurance companies. Business Intelligence provided a good solution. The paper described and analyzed the implementation of the business intelligence systems in insurance industry combining with the business intelligence projects in a life insurance company.This paper carried out a detailed analysis for the life insurance company’s business intelligence system requirements, On this basis, using the data warehouse model design, source data extraction transformation and loading (ETL) process, and online analytical processing technology to achieve the construction of a business intelligence system. In data warehouse modeling aspect, using a modeling method based on multi-dimensional model implemented the design of it’s concept model, logical and physical models. used a method of combining DataStage tools and stored procedures to achieve the ETL process and task scheduling for source data to the target data warehouse. In the online analytical processing aspect, made full use of the MOLAP server’s multidimensional data structures which was stored by data cube and powerful pre-processing function, proposed optimization algorithm for multi-way array aggregation algorithm, which can easily find the optimal scanning order and reduced the time spent for the pre-computation of the data cube. Finally, using Cognos powerful report shows function developed a multi-dimensional analysis results showing platform, completed the design of business intelligence system. The users can view data from a variety of dimensions, more than one side and multiple degree of data integration through the analysis operating on dimensions, such as drill up and down. Thus they can more in-depth understand the law implied by data. At the same time, for the consideration of system security, the system was given permission management approach. After a period of testing, the system is running well, and achieved the desired results.