On Data Warehouse Technology In Hospital Decision Support System

Accompanied with increasing competition among hospitals, rapid development of database technology, hospitals rely on information day by day. In order to provide support for managers of hospitals and to improve management level of hospitals, one necessary requirement is to distill useful knowledge from a mass of data in hospital information system. A new tendency in hospital informationization construction is to combine data warehouse, online analytical processing and data mining technology in exploitation of hospital decision support system.This thesis focuses on application of data warehouse and online analytical processing in hospital decision support system. It uses the design and operation of data warehouse system applied in The Fourth Hospital of Shijiazhuang as the example. In that system, data analysis is developed on visiting time for outpatients theme and geographic distribution for inpatients. A multi-dimensional cube theme view of hospital data is constructed based on a star model on visiting time for outpatients them and a snowflake model geographic distribution for inpatients. Data from various dimension will be obtained by slicing, carving, drilling, and pivoting methods, which can be analyzed by online analytical processing. As analytic outcome, multi-dimensional data on practical operation of that hospital, will be outputed as report forms. At last, those report forms can be used for inquiry, multi-dimensional anylasis and reason analysis, predicting the future, and supporting decision-making. Furthermore, in order to decrease data error and to ensure scientific conclusion, data quality and data cleaning should be focused in the preceding process.Application of warehouse can collect scattered data in different sub-systems of certain hospital and manage them efficiently to provide analytical data for decision support, so that managers can omnidirectionally grasp whole of business operation. Meanwhile, it can relieve inquiry pressure in business process and increase instant operating effciency and data service efficiency for inquiry and statistics.