Artificial Neural Network (ANN) is a feasible method to reflect the complicated nonlinear relationship between β transus temperature and the alloy composition. In this paper, back propagation neural network (BP neural network) was developed and trained using data from various sources of published literature. The influence of aluminum, molybdenum and zirconium on β transus temperature in titanium alloys was assessed on the base of the trained neural network. It is found that the predicted results are in good agreement with experimental values. The effect of element contents on β transus temperature simulated by ANN model presents nonlinear relationship caused by the interaction among the elements, which is different from the results of the traditional equations.