Perforation technology is used to establish flow channels to provide pathways for proppant to enter the oil and gas reservoir, while the proppant is applied to support the fracture walls. Proppant transport behaviors within the perforation intervals, such as, proppant settling and clogging, could affect the fracture conductivity so that the fracturing effectiveness and hydrocarbon production would be reduced. Therefore, understanding the proppant transport behavior characteristics and optimizing proppant distribution uniformity within the perforation intervals are crucial. In this paper, a solid-liquid model was established using the coupled approach of Computational Fluid Dynamics (CFD) and Discrete Element Method (DEM) in a horizontal wellbore with perforations. Firstly, the model was validated by comparing its accuracy and capability with experimental data. The results investigated that a recycle flow existed here, which brought the proppant back in a swirling motion. Meanwhile, the recycle flow drove the collisions between proppant so that the proppant velocity would be decreased, and due to the effect of gravitational forces, the proppant tended to settle down at the bottom, led to increasing a higher proppant bed so that the proppant were easily plugged at the blind section. Additionally, a new multivariate regression equation was introduced to predict proppant distribution uniformity. These findings provide solutions for achieving uniform proppant distribution, which has significant practical implications for oil field operations.