The conventional prediction of milling stability has been extensively studied based on the assumptions that the milling process dynamics is time invariant.However,nominal cutting parameters cannot guarantee the stability of milling process at the shop floor level since there exists many uncertain factors in a practical manufacturing environment.This paper proposes a novel numerical method to estimate the upper and lower bounds of Lobe diagram,which is used to predict the milling stability in a robust way by taking into account the uncertain parameters of milling system.Time finite element method,a milling stability theory is adopted as the conventional deterministic model.The uncertain dynamics parameters are dealt with by the non-probabilistic model in which the parameters with uncertainties are assumed to be bounded and there is no need for probabilistic distribution densities functions.By doing so,interval instead of deterministic stability Lobe is obtained,which guarantees the stability of milling process in an uncertain milling environment.In the simulations,the upper and lower bounds of Lobe diagram obtained by the changes of modal parameters of spindle-tool system and cutting coefficients are given,respectively.The simulation results show that the proposed method is effective and can obtain satisfying bounds of Lobe diagrams.The proposed method is helpful for researchers at shop floor to making decision on machining parameters selection.