The purpose of this paper is to investigate the particle removal performance of pickup head for a street vacuum sweeper numerically. An integrated 3D numerical model was constructed based on particle suction process in computational fluid dynamics (CFD) software. The airflow through the pickup head was treated as a continuum, while particles were modeled as dispersed phase. The Reynolds stress model (RSM) and discrete particle model (DPM) were chosen in order to predict the air and particles flow accurately. The numerical simulation results show that the sweeper-traveling speed and the pressure drop across the pickup head have great effects on the particle removal performance. The removal efficiency of particles increases with the lower sweeper-traveling speed or the higher pressure drop, and small size particles have higher grade efficiency than that of large size particles under the same operating conditions. Moreover, the removal mass flow rate of particles increases with the higher sweeper-traveling speed. Therefore, a trade-off should be considered among high removal efficiency, low energy consumption, and high removal mass flow rate. Through the numerical simulation, the effectiveness of street vacuum sweeper for removing particles from road surface is evaluated, and an optimal operating condition is obtained. Besides, more information is generated to better understand the particle suction process of the pickup head. (C) 2010 Elsevier B.V. All rights reserved.