Two-phase flows are present in all the value chain of the oil industry, being of significant interest in pipeline transportation. They are essential for the calculation of production rates or separation process design; therefore, multiphase flows have been studied for several years, and numerous models, data banks and computational software have been developed to design more efficient processes. For this reason, the purpose of this study is to develop a CFD numerical model capable of predicting air-oil and air-water annular flow for up-and downward flows in vertical pipes with the objective of present a methodology to develop a reliable numerical model and present CFD tools as an alternative to the empirical models, or other commercial computational codes such as the dynamic multiphase flow simulators, for the study of multiphase flow. To achieve this objective, 36 simulations using CFD were compared against 150 simulations using OLGA and 66 different empirical correlations to predict void fraction and compare the obtained results against experimental data. Different liquid viscosities (0.00089, 0.127, 0.213, 0.408 and 0.586 Pa s) and three different pipelines were used: a 22.72 m long and 0.0508 m ID pipe, a 15.24 m long and 0.0508 m ID pipe, and a 15.24 m long and 0.1016 m ID pipe. The obtained results showed that the CFD model accurately predicts the void fraction for both down-and upward cases, while the obtained results using OLGA and the empirical correlations showed a lower accuracy.