Direct bonding of silicon wafers plays a crucial role in the fabrication of three-dimensional integrated circuits and MEMS devices. Surface activation bonding (SAB) using a fast atom beam (FAB) enables room-temperature wafer bonding by removing oxide layers and modifying surface properties. However, optimizing FAB irradiation parameters requires extensive experimental validation, making process development both time-consuming and inefficient. In this study, we investigated the effect of oblique FAB irradiation on the oxide removal distribution and developed a predictive model using two-dimensional plasma simulations. Oxide film removal experiments were conducted under various irradiation angles, and the results were compared with simulated total incident energy distributions. We observed that oxide removal trends followed the total incident energy distribution but exhibited shifts in peak positions due to variations in sputtering yield and experimental uncertainties. To extend the predictive capabilities, we introduced a pseudo-three-dimensional model by combining oxide removal predictions along both the x- and y-axes. This model provided a reasonable approximation. Future improvements will focus on incorporating secondary ion interactions and surface morphology effects to enhance predictive accuracy. These findings contribute to the advancement of FAB-based wafer bonding by enabling more efficient parameter optimization through simulations, thereby reducing the need for extensive experimental iterations.