Aiming at the problem of low surface roughness prediction accuracy and unreasonable cutting parameters selection in 6061Al milling, an optimization model based on genetic neural network and genetic algorithm was proposed to optimize the cutting parameters of 6061Al. The genetic neural network (GA-BP) was used to construct the surface roughness prediction model. Based on surface roughness prediction, the cutting parameter optimization model was constructed with the material removal rate as the optimization target. The genetic algorithm was used to optimize the 6061Al cutting parameters.The research results show that the prediction accuracy of the surface roughness of the predicted model is above 97%. At the same time, the optimization model can optimize the 6061Al cutting parameters, which can achieve better global optimization results and provide reference for optimizing the cutting parameters of aluminum alloy workpiece milling.