Abstract: Due to the adoption of operational data storage method, the coal mine safety monitoring system can't use massive data effectively and the data analysis capability is poor. In order to solve the above problems, this paper proposes the key technologies of data warehouse for coal mine safety monitoring. According to the business requirements of coal mine safety monitoring, the functional structure of coal mine safety monitoring data warehouse is proposed. Moreover, the five business subjects are designed, including overrun analysis, calibration analysis, abnormal data analysis, measuring point network interruption analysis and personnel management analysis. The logical model of coal mine safety monitoring data warehouse is established by using the fact constellation model. The fact table and dimension table are designed by subject. The physical model of data warehouse is established by using SQL Server. According to the characteristics of coal mine safety monitoring data warehouse, data extraction, conversion and loading strategies are proposed. The different data extraction rules are used to extract data by subject. The data from different sources are processed through format conversion, cleaning and sorting. In the process of data loading, pre-loading, loading and post-processing operations are carried out.