In most bonding processes, an adhesive is applied to a substrate in a specific pattern before the second substrate is subsequently pressed against it. During this, the adhesive flows in such a way that, ideally, it completely fills the joint. In practice, however, areas with entrapped air frequently remain in the bonded adhesive layer. Within the scope of a research project, these flows are systematically analyzed in order to identify optimal initial application patterns for the adhesive and substrate geometry to minimise such risks. For this purpose, the authors use an efficient flow model, the partially filled gaps model (PFGM), extended in this study to include the functionality of trapped air pockets. Depending on the volume fractions of air and adhesive, the flow of both phases is computed. Therefore, the model is introduced and fully described, benchmarked with respect to its plausibility and functionality, and results obtained are compared with a CFD calculation. Thereafter, the functionality of openings and closings of the pockets are analyzed. Lastly, the model is then applied to a real scenario created with a Hele-Shaw cell measurement. The benchmark as well as the comparison with the measurement results show the high potential of this technique.