Abstract: Firstly, the meteorological decision-making services paradigm and its two driving mechanisms of knowledge and data were defined, and it was pointed out that paradigm shift is inevitable, necessary, and feasible; Then, a comparative study was conducted on the differences and advantages of the meteorological decision-making services paradigm under the two driving forces of knowledge and data in terms of information sources, decision-making participants, judgment logic, and decisionmaking processes. The mathematical expressions of control rules were given separately, and the Architecture of Complexity theory was introduced to demonstrate the practical path of paradigm shift. It was pointed out that at the technical level, dynamic decomposition principles and logical judgment criteria based on causal rules and supplemented by association rules should be met. At the application level, strong coupling within hierarchy and weak coupling between hierarchy should be met, and a data governance community oriented towards scenario based and refined decision-making needs should be constructed. At the strategic level, a dynamic collaborative driving process with knowledge as the main data supplement, knowledge and data combination, and data as the main knowledge supplement can be deployed. The research results provide a theoretical basis and practical path for the development of intelligent meteorological services for emergency disaster reduction.