Parameters related to the drift-flux (DF) 2-phase flow model are optimized to improve predictive performance. To facilitate this work, an extensive data review was performed resulting in the construction of a generalized, consistent and uniform data bank (suitable for all 2- and 3-phase data) that currently holds over 3500 data points. A simple 'stratified equivalence' (SE) model, based on the classic Taitel & Dukler model, is also presented for predictions of horizontal flow, which the existing DF model does not cater for. Optimized parameters for this SE model are presented and evaluated against data. A method for a seamless transition between the two models is also presented. The result is a unified model, valid for inclinations from -5° (downward) to +90° (vertical upward), suitable for the strict requirements of complex coupled system optimization where smoothness and robustness are essential.