To evaluate nuclear power plant reactor building containment, on-site pressurization tests up to 4 bars relative pressure are performed to measure leakage rates, influencing the French Nuclear Safety Authority's decision to authorize plant operation. Given the concrete's aging nature in the context of lifespan expansion, there is a need for a tool to estimate leakage rates over time, considering potential accidental loadings like coolant loss or seismic events. Experimental studies such as MAEVA (Granger et al.,2001) or VERCORS (Charpin et al., 2021) have assessed the structural scale's leakage ratio through concrete porosity and cracks. These results are crucial for developing simulation tools within the quasi-brittle material mechanics framework (concrete) and porous/fractured medium transport (diffusion and flow). Numerous investigations have demonstrated that various aspects of crack geometry (e.g., opening, roughness, tortuosity) influence leakage rates within the specimen scale (Akhavan et al., 2012). However, precisely characterizing the three-dimensional geometry of concrete cracks is a significant challenge. A new method has been created that merges Finite Element Analysis, Beam-Particle Modeling, and Computational Fluid Dynamics (CFD) to predict concrete crack geometries and air leakage rates accurately. A first finite element model calculation is performed in the model's first phase, implying macroscopic continuous finite element analysis, not discussed in this article, to handle the high processing needs of discrete simulations. Then, as described in this work, a refined discrete beam-particle model computation will be utilized as input for a CFD model for future work. By predicting the reinforced concrete crack patterns in terms of tortuosity and crack opening variations, the model greatly improves our understanding of crack modeling in nuclear reactor containments with the ability to estimate leakage rates that strengthen safety assessments and regulatory processes.