This paper presents a simulation and optimization study of the surface generation process in ultra-precision turning using a formerly proposed model by the authors in preceding work. However, the surface generation model was improved by incorporating the effect of cutting speed into the model considerations. Then, a series of simulation trials were undertaken to predict the generated surface roughness over a wide range of cutting conditions. Particularly, two different dual-phase materials, Brass 6040 and medium carbon steel AISI 1045 were examined under ranges of cutting speeds, feed rates in addition to varying geometries of cutting tools. The study was accomplished by conducting an optimization of the process aiming at identifying the optimal cutting conditions for the best possible performance of the process in terms of minimum roughness average (Ra). In particular, two optimization techniques, Sequential Quadratic Programing (SQP) and Genetic Algorithm (GA), were utilized to tackle the enhanced model. The results demonstrated that although cutting speed influenced the process, cutting edge radius was the dominate parameter and feed rate was found to be the second most effective variable. Also, optimization results confirmed the superiority of the GA technique over SQP method to obtain the optimal cutting conditions for minimum roughness