Abstract: The spot-scanning carbon ion radiotherapy relies on the ripple filter (RiFi) to improve treatment efficiency. The stacking method of the multi-layered metal mesh ripple filter (mRiFi) is random, resulting in lower dose distribution consistency. This study designed a 3D matrix of mRiFi with wire diameters of 0.1 mm, wire spacing of 0.5 mm and layer numbers of 10, 20, 30, and 40, based on Python, and optimized the stacking parameters using a genetic algorithm. The Monte Carlo dose simulation program TOPAS was used to evaluate the Bragg peak broadening ability of carbon ion beams passing through mRiFi and the dose distribution consistency before and after optimization. The results show that the stainless steel mRiFi has stronger broadening capabilities than the aluminum mRiFi. When a 100.5 MeV/u carbon ion passes through 40 layers of stainless steel mRiFi, the Bragg peak width broadens from 0.23 mm to 4.08 mm, while the aluminum mRiFi broadens to 1.73 mm. After optimizing the stacking parameters with the genetic algorithm, the dose distribution consistency significantly improved. For the 10, 20, 30, and 40 layers of aluminum and stainless steel mRiFi, and the average \gamma passing rate increased when comparing the dose distributions at different irradiation positions with that at the central irradiation position, and there are statistically significant differences in most cases. The study demonstrates that the mRiFi optimized by the genetic algorithm can significantly improve the Bragg peak dose distribution consistency, providing a better solution for the construction of mRiFi in carbon ion radiotherapy.