Effective control of the endpoint steel temperature and contents of carbon, sulphur, etc. is one of the main tasks of BOF steelmaking process. This paper established a multivariable predictive model for BOF steelmaking using RBF neural network. The input data is pretreated and standardized. Receding horizon control method is used to increase the accuracy of the model. Simulation and experiment comparisons show that the model is validated and has high hit rate.