Face masks have become one of the most important pieces of day-to-day equipment due to the recent COVID-19 viruses. However, doing experimental studies to observe the filtration behavior was not enough due to higher time consumption and technical difficulties. There have been lots of simulation studies done to reduce this matter. Most of these studies were done using 3-dimensional structures, and they needed higher computational power. In this study, we develop a 2-dimensional computational model to predict the filtration efficiency of the electret layer of an N95 face mask. 2-dimensional fiber structures were constructed using parameters such as packing density, fiber diameter, and fiber layer thickness. MATLAB and AUTOCAD were used to generate random 2-dimensional fiber structures, and COMSOL Multiphysics was used to simulate filtration mechanisms inside the electret filter. The incompressible Navier-Stokes and continuity equations were used to solve the velocity field, and the Lagrangi-an method was used to obtain particle trajectories inside the electret filter. Interception, inertial impaction, and electrostatic attraction were used as filtration mechanisms. Particle diameter sizes vary between 10 nm and 1 μm. The behavior of the filtration mechanisms under different conditions was simulated, and the simulated results are compared with the experimental results from literature. This study gives suggestions for finding the optimum filter characteristics for electret filters. It is highlighted that sticking probability shows a more significant effect for mechanical filtration than electrostatic filtration. Finally, this study recommends that using extreme conditions is not necessary to achieve required filtration efficiency.