Discrete Element Method (DEM) has become a valuable tool to predict the behaviour of granular assemblies. Furthermore, it has proved that can provide more accurate loads when it comes to estimate forces in applications where the interaction between bulk material and machines can be challenging. Some studies have attempted to predict DEM loads experimentally with different degrees of success[1-2], however, all of these were focused only on the DEM side and usually calibration of DEM parameters was made by trial and error or using bulk calibration approaches such as the angle of repose [3]. On the other hand, there are just a few studies that cover the one-way DEM-FEA coupling [4], however, mostly lacking experimental validation and explanation about how loads are transferred into FEA to validate strains. This highlights the importance of a better understanding of this coupling which can lead to accurate load estimation and therefore machine design improvements. The present investigation looks at the one-way DEM-FEA experimental and numerical validation using a cohesionless bulk material. The experimental set-up consists of a rectangular steel plate that moves along a material bed made of loose gravel. The plate is instrumented with a load cell and a strain gauge. It is assumed that the loads are within the elastic behaviour of the plate and with small deformations, therefore, only one-way coupling is applied. For the DEM simulation, a multi-sphere particle shape was used where sliding and rolling friction parameters were calibrated using the draw down test. Coefficient of restitution, sliding friction between boundaries and particles, and Young’s modulus of particles were also estimated. It was found that DEM forces are in agreement with the experimental forces, capturing the trend throughout the trajectory of the plate. DEM forces over the plate were imported into FEA obtaining a good correlation with experimental results. Sensitivity analysis was also carried out where sliding friction coefficient property of the particles was found to affect the loads proportionally. This property is proposed as the most important parameter that should be calibrated beforehand in order to capture loads accurately using DEM.