An optimized power management strategy for Belt-Starter Generator (BSG) mild hybrid system is proposed and used to study its benefit of fuel economy on a 2.2L turbo engine. First, a cost function is defined as fuel and battery power consumptions. In order to obtain an optimal fuel economy of BSG, Dynamic Programming (DP) optimization approach is employed to minimize the defined cost function over New European Driving Cycle (NEDC). A nonlinear vehicle dynamics simulation model is established using Matlab®/Simulink. Experimental data are utilized to verify the nonlinear model. Since the detailed simulation model is not suitable for the DP due to its large number of states, a simplified dynamic model is developed in this paper. An optimized power management strategy is extracted by analyzing the results of DP and is evaluated using NEDC. Preliminary simulation results show that the proposed strategy presents lower fuel consumption than that of the conventional strategy by about 11.4%.