Observations in the field suggest that the existing guidelines for bed protection design may lead to overly conservative protections at berthing structures of inland vessels. In this study, physical scale model tests have been performed using particle image velocimetry to assess the near-bed velocity and the velocity decay of jets generated by transverse bow thrusters. The effect of the bed roughness, under-keel clearance (UKC) and quay wall clearance on the resulting near-bed velocities are discussed here. The results show that the maximum near-bed velocity occurs in the vicinity of the quay wall and results from an interplay between jet diffusion and the blockage effect of the nearby boundaries (e.g., quay wall, bed, and vessel), where the roughness of the confining boundary plays a role. The existing design guidelines predict fairly well the maximum near-bed velocity for smooth beds and for relatively large UKC's. However, the guidelines tend to overestimate the maximum near-bed velocities for rough beds and for small UKC's. This study proposes a new method based on measurement data to estimate the maximum near-bed velocity for confined bow thruster jets.