Temperature adjustment is one of the critical tasks affecting the quality of manufactured steel. This is controlled by the Basic Oxygen Furnace’s (BOF) blowing procedures. As many factors influence variations in temperature, it is often difficult to predict the blowing quantity necessary to achieve a required temperature. In this study, we assume the framework used by the intelligent blowing control system uses the Case Based Reasoning (CBR) and Neural Network (NN) to predict the appropriate blowing quantity in the BOF. Our proposed framework consists of three steps. First, we retrieve the similar cases for a new order requirement using CBR. Next, we train the NN engine with the selected case set. Finally, we predict the appropriate blowing quantity using a trained neural network. Experimental results show that the proposed framework performs more effectively than the framework without using CBR process.