The aim of this study is to put forward a regression model equation to predict the membrane transport numbers in a two-compartment membrane cell for the batch/continuous electro-membrane salt splitting of lithium sulfate (Li2SO4). It was found that other than main effects (Li+ and H+ in anolyte, OH- in catholyte); the following critical factor pairs were effective in determining the significance of a specific ion concentration on the transport number of lithium; "Li+.H+ and Li+/OH-" ratios for Li+ ion, "H+/Li+ and H+/OH-" ratios for H+ ion and "OH-.Li+ and OH-/H+" ratios for OH- ion. With this finding, a multiple regression equation with an R-squared value of 0.953, adjusted by exponential terms to reflect the effect of the interactions was established. This regression equation was also helpful in predicting transport numbers of H+ and OH- ions with an approximation. The R-squared value of the equation increased even higher to 0.975 when only batch data was taken into consideration. Moreover, transport number of water was fitted in a similar regression equation having an R-squared value of 0.912. The results of test experiments, which were kept out of the model, were successfully predicted with average error values of 7% and 3% for lithium and water respectively. (C) 2016 The Electrochemical Society. All rights reserved.