A new design optimization technique is presented to improve the analytical performance of the drilling process of graphene oxide nanocomposites. A detailed study was conducted for modeling-design-optimization of the drilling process using multiple nonlinear neuroregression analyses for this goal. The data were slected from a literature study for this objective. The accuracy of the predictions of the nine potential functional structures presented for modeling the data was tested using a hybrid neuro-regression-based technique. Model selections to determine the objective functions were made by controlling the R 2 values, limit values, and statistical results, respectively. The selected models were used in the optimization studies of delamination and thrust force values with four different optimization algorithms. The results show that the R 2 training and R 2 training-adjust values give good results in the nine models as objective functions. However, R 2 testing values and statistical calculations were distinctive among all models. Furthermore, when the optimization results of the third-order polynomial and logarithmic models for both responses were compared to the reference study's results, it was observed that the current results were more closer to the test results.