Abstract: Reliable estimation of the fine particle rattler fraction is crucial for understanding the structural and mechanical responses of binary granular systems with large size ratios. However, such estimation is challenged by the general inability to obtain interparticle contact force information directly from experimental images and by the lower accuracy of positional and size identification of fine particles compared with coarse particles. To address these challenges, in this study we focused on two-dimensional (2D) bidisperse granular assemblies with large size ratios (α = 7, 9, 12, and 16) and developed an approach based on Monte Carlo simulation (MCS) that relies solely on the size and positional information of coarse particles, avoiding the need for force-resolved computations. The performance of the method was evaluated against experimental measurements and discrete element method (DEM) simulations. The MCS-based predictions show close agreement with experimental results, with a slight overall overestimation. At low fines content, the approach tends to overestimate the fine particle rattler fraction relative to DEM results, whereas at higher fines content, it underestimates rattler fraction. Overall, the proposed MCS-based approach enables robust and relatively accurate estimation of the fine particle rattler fraction. This study provides a practical framework for predicting rattler fraction, contributes to advancing both experimental analysis and theoretical modeling in granular physics, and demonstrates the conceptual extendibility of the MCS framework to more complex three-dimensional (3D) packings.