<p>Although resistive force during intruder penetration into granular layers plays a crucial role in various applications, its underlying mechanisms remain insufficiently understood. In this study, we investigate penetration resistive force using discrete element simulations, systematically varying the angle of repose, interparticle cohesion stress, intruder shape (tip angle and horizontal cross-sectional geometry), and the interface friction between the intruder and particles. The simulation results are then compared with estimations from the extended modified Archimedes’ law. As a result, the current model cannot fully capture the effects of these factors, except for intruder shape. Through the detailed strain field analysis of granular layer during intruder penetration, we identify that the discrepancy between the model and simulation results arises from differences in the failure modes of the granular layer. To address this, we modify the model parameters based on the failure modes. Furthermore, we introduce a formula that incorporates the effect of the interface friction, which is not accounted for in the current model. With these modifications, the model can quantitatively estimate penetration resistive forces in dry and cohesive granular layers across various simulation conditions. The analysis of variance indicates that the interface friction and angle of repose have a significant impact on prediction accuracy of the model, supporting the effectiveness of the modification. This study offers a comprehensive understanding of the key factors influencing penetration resistive forces and contributes to the development of more accurate predictive models.</p>