Compared to conventional machine tools, industrial robots offer great advantages such as multitasking, larger workspace, and lower price. However, these advantages of robots are undermined by their high structural flexibility leading to excessive deflections, severe vibrations, and ultimately violating dimensional tolerances and poor surface finish. Modeling the dynamics of robots under machining (e.g. milling and drilling) forces is essential for reducing deflections and vibrations during the process. Although modeling the dynamics of traditional machining systems is a well-studied subject, the existing modeling approaches are not applicable to robotic manipulators because of the posture-dependent dynamics of industrial robots. Within this context, the presented thesis aims to predict the stability of vibrations during robotic machining operations through prediction of posture dependent dynamic behavior of robots. A rigid-body modeling approach is used to identify the dynamic parameters of the robotic manipulator based on least squares estimation method. Next, by adopting a rigid link flexible joint model and employing experimental modal analysis to identify the joint stiffness and damping parameters, posture dependent dynamic response prediction of the robot is achieved. Finally, the posture-dependent milling stability is presented as a function of the predicted tool center point transfer function, spindle speed, and axial depth of cut. A Staubli TX200 robot and a Kuka KR90 robot are used as experimental case studies. / Graduate