A statistical mechanical equation of state is developed to predict the volumetric properties of pure and mixture liquid alkali metals at different temperatures, pressures and compositions. The temperature dependent parameters of the equation of state have been calculated using corresponding states correlation based on the normal boiling point parameters as scaling constants. It is shown that the knowledge of just normal boiling point and its liquid density are sufficient to estimate the thermodynamic properties of pure and mixture liquid alkali metals in different conditions. Besides, the performance of artificial neural network (ANN) based on back propagation training with 10 neurons in hidden layer for prediction of behavior of presented systems was investigated. A collection of 512 data points for above systems in different temperatures and pressures was used. The Tao-Mason equation of state (TM EOS) and ANN model results have good agreement with the experimental data with absolute average deviations of 0.74% and 0.299%, respectively. (C) 2014 Elsevier B.V. All rights reserved.