Bearing acts as a pad that supports a shaft to rotate without excessive friction, hold the radial load and maintain the motion towards the left and right shafts (Thrust Load) when turning together. Due to frequent taper bearing damage to the wheels, predictive maintenance is therefore necessary. One of the predictive maintenance methods widely used today is vibration analysis. The principles of vibration-based bearing damage detection using the Frequency Response Function (FRF) method will be shown in this paper. Harmonics signal is given to the bearing surface in a vertical or perpendicular direction to the taper bearing surface. The vibration response measurement was carried out on three axes (x, y, and z). The results of this study indicated the ability of FRFs to predict any damage on the bearing taper.