Ultra-precision machining (UPM) technology is extensively applied to manufacture top quality products with high precision level and complicated geometry. As complicated machining factors affect the surface quality of machined components in UPM, large numbers of experiments for understanding the influences from particular machining factors are needed, leading overestimate or underestimate of significance of machining factors at certain machining conditions and raising of experimental cost. For these reasons, a crucial approach is urged to adapt for providing a fast track to an optimal machining condition. In this study, social network analysis (SNA) is introduced firstly to develop UPM network, which the network shows the relationship between dominant machining factors in UPM. A complicated UPM network containing interdependencies between each machining factor is generated by SNA. The determinations of network metrics in the UPM network support the selection of optimal machining factors under various machining conditions. Furthermore, the constructed UPM network using SNA provides the complete framework of dependencies in UPM for well predicting the machining outcomes when particular machining factors are adjusted in practical situations. The study contributes to offering a detail guideline for constructing machining strategies or experimental plans to efficiently achieve desired machining outcomes.