Commercial Bank Data Warehouse Modeling Analysis

With the development of the computer technique, data warehouse technique has already become mature. Meanwhile, with the increasingly improvement of commercial bank's various business systems, particularly the drastic competition between the banks, management decision based on information becomes very vital. It is the only way for commercial bank to construct its data warehouse system.In this paper, we begin with the introduction of the fundamental, system framework and "business dimension lifecycle method" of the data warehouse. Further more, standing on perspective of the whole banking business and management, and using the dimension modeling method, we analyze the commercial bank's data warehouse model in detail, Mainly include the following four aspects:1) Subject model. After the native commercial bank's business system and request of stat and analysis have been analyzed in detail,we conclude the six subject about client, product,account,business,channel and bank organization.2) Logic modeling. Using the dimension modeling method,We analye and constuct the logical model of commercial bank's data warehouse,mainly include the following four step:one is to create the data warehouse general thread structure matrix,the second is to design ervery fact table according the data mart; The third is to design the consistent dimension table, and the last is to design mass fact table according the request of stat and analysis.3) Physical design. According the result of logical modeling, we design the database tables including dimension tables,basic fact tables and mass fact tables4) Data integration. According the database table structure of the data warehouse, we analyze the source of data and confirm the corresponding relation between the source table and the data warehouse table,then we design the algorithm of data extract,transform and checkout, So we can integrate the data source into the data warehouse.As a result, we put forward a new method on how to construct the commercialbank's data warehouse. At the same time, we provide the preferable analyses model and successful case of commercial bank's data warehouse. The model and method would be effectively referenced to other commercial banks.