This research is to develop an ultrasonic vibration-assisted turning system and apply the robust process design method to optimize and enhance its turning efficiency. The first part of the study uses the finite element analysis method to analyze the natural mode of the (UAT) system tool and the horn and finds the maximum resonance frequency to effectively strengthen the vibration transmitted by the horn to the tool to obtain the maximum amplitude. According to the modal analysis results, the vibration frequency 19KHz is the closest to the Y-axis vibration direction of this study. The second part of the study applies the robust process design method to turn AISI 1045 steel and optimize its surface roughness. Knowing that the optimal cutting parameters are nose radius 0.4(mm), feed rate 0.1 (mm/rev), cutting speed 250 (m/min), depth of cut 0.15 (mm), the best surface roughness 0.74(µm); and compare the surface roughness with/without ultrasonic vibration assisted turning by its cutting parameters; the experimental results show that ultrasonic vibration assisted turning has better effect than that without ultrasonic vibration assisted turning.