Multi-objective optimization of EDM machining process of hardened 90CrSi steel will be presented in this study. The input parameters including Concentration of powder, Pulse on time, Pulse off time, Current, and Voltage are chosen to minimize the electrode wear rate and maximize material removal rate. Grey Relational Analysis (GRA) is combined with Taguchi method to find the optimal set of machining parameters which can satisfy the responses. It is found that Current has the strongest influence (24.89%), but Volgage has the smallest effects (1.63%) on the responses. The proposed model corfirmed by experiment and ANOVA exhibits the reability approch to predict electrode wear rate and material removal rate.