Objective To achieve a high degree of process-oriented and automated decision-making on food safety risk assessment and data fusion among various business units. Methods Establish a temporal model library covering method such as dietary exposure assessment, hazard factor assessment, and spatio-temporal clustering detection, which can automatically connect to the basic data of each link of food traceability to obtain corresponding data, and calculate the risk assessment result through the selected risk assessment model construct a risk assessment matrix. Based on extract-transform-load (ETL) technology and data analysis algorithm implemented by R language, the basic data warehouse, risk assessment model base and risk decision support system were integrated. Results The establishment of a food safety risk assessment decision-making system would effectively resolve the problems of time-consuming and labor-intensive traditional data assessment and data cleaning difficulties. Based on the electronization of the original risk assessment process, the model input, calculation, and output would be integrated, and multi-year historical monitoring would be integrated to quickly customize the risk assessment research scenarios for harmful factors in common food categories. Conclusion This platform help improve the work efficiency of relevant business personnel, and promote data exchange and collaborative sharing between business units.