Thin-wall tubular-geometry workpieces have been widely applied in aircraft and medical industries. Though numerous research reports are available that the tool path can be planned on the basis of preset surface profile before actual milling operation is performed, it is still difficult to predict the real-time surface profile errors for peripheral milling of thin-wall tubular workpieces. This research is focused on how to real-time formulate the appropriate applied cutting torque via feedback of spindle motor current. A few suitable cutting conditions which are able to prevent potential break/crack of thin-wall workpieces and enhance productivity but almost retain the same cutting quality is proposed. To achieve this goal, estimated surface profile error on machined parts due to deflections caused by both tool and workpiece is studied. The torque estimation approach by spindle motor current feedback and the corresponding fuzzy logic controller are employed. Compared with constant cutting torque during milling operation in tradition manner, it is observed that the time consumption of milling cycle by aid of the aforesaid fuzzy logic controller is greatly shortened while the resulted cutting accuracy upon finish of workpiece can be almost retained.