Abstract As one of the most common obstacles in milling process, regenerative chatter will result in lessened productivity, poorer product surface finish and decreased cutting life of the tool. In order to suppress chatter, the spindle speed variation (SSV) has been proposed and researched for a long time. However, the previous researches mainly focus on the basic waveform variation, such as sine, cosine, square and triangular waveforms and didn't consider the effect of phase on milling stability, which weren't suitable to the complex and high spindle speed milling conditions. Based on the previous studies, this paper proposes the concept of multi harmonic spindle speed variation (MHSSV) including the phase factor for chatter suppression. The dynamic equations with MHSSV are derived to describe the milling process. Due to the existence of the milling period and the speed variation period, the least common multiple of these two periods is adopted as the Floquet period in semi-discretization method (SDM) for stability analysis. Because the speed variation function can be described by some finite parameters, the genetic algorithm is used to optimize these parameters in order to suppress chatter more effectively. As a result, the optimized milling process has higher stability limits, especially in the high speed zone, which validates the effectiveness of the multi harmonic spindle speed variation. In addition, the numerical simulation of milling process is implemented and verifies the correctness of the proposed method.