Abstract: Objective The aim is to study MIG process optimization problem of 5083-O aluminum alloy in the welding of outer panels for high-speed train driver’s cabs, with a focus on the control of weld reinforcement, so that robot welding system can meet the requirements of high-speed train outer panels welding. Methods Response surface methodology was used to establish a response surface model for reinforcement of MIG butt joints. It also predicted the probability of three types of defects using logistic regression. Optimization targeted ensuring standard reinforcement and minimizing the probability of defect occurrence, generating optimal process parameters (wire feeding speed and welding speed) for different weld features (groove angle and weld gap). Results The results indicate that the established reinforcement response surface predicts well, with excellent outcomes from variance and regression analyses, achieving a predictive determination coefficient of 0.938 6. Verification trials shows that error of this model in predicting reinforcement is less than 5%, and no evident macroscopic defects are observed in the welds’ metallographic photographs. Conclusion The model constructed in this study has the ability to control weld reinforcement and reduce defect probability, and weld gap is the most important factor affecting weld reinforcement of outer panels for high-speed train driver’s cabs.