In this paper, based on the CAE technology for MIM and fractional factorial design method, the simulation for the filling process was executed in different process conditions by the Moldflow software. And then it was realized that the extent of the effects of the single factor and pairwise interaction between the process parameters on quality of green parts were forecasted. According to that, the relationships of pressure, velocity and time and their optimal combination were researched by the experiments using a FC-80 plastic-Injection-Moulding machine. The experimental results suggested that the pressure, velocity and time have a obvious pairwise interaction which can have more effect on the product than single factor. It was shown that there was a good consistency between the numerical simulation and experiment. The CAE technique and the experiments are combined to optimize the process conditions and improve the part quality automatically by using smaller number of experiments.