Abstract: A prediction model of oxygen decarburization efficiency is proposed based on mutual information case-based reasoning, and the oxygen blowing amounts of static and dynamic phases are calculated according to the forecasting results. A new prediction method of oxygen blowing amount is proposed firstly, which attributes the oxygen decarburization efficiency as solution properties of case-based reasoning. Then the mutual information is introduced into the process of determining weights of properties, which solves the problem that the lack of information is ignored between the problem properties and the solution properties in the traditional case retrieval method. The proposed model is applied to the actual production data of a 150 t converter, and the simulation results show that the model has high prediction accuracy. The method can achieve precise calculation of blowing oxygen volume and satisfy the requirement of production.