Draw-bead is applied to control the material flow in sheet metal forming process and improve the product quality by providing sufficient draw-bead restraining force (DBRF). However, actual die design depends mostly on the trial-and-error method without the accurate prediction of DBRF and the normal force. Die design with the predicted value of them can be utilized effectively to reduce the cost and time. In order to construct a prediction model, process parameters, which affects significantly to DBRF and the normal force, are selected as design variables. At first, the prediction model without considering the effect of BBF is constructed using the Box-Behnken design which is the most representative DOE method. Results obtained from finite element analysis are approximated with the second order regression model. It provides the maximum value of DBRF and the normal force because the male bead fully presses down the sheet. After that the variation of DBRF with respect to the change of BHF is approximated with the first order regression model by adopting the fractional factorial design. It provides the equation for the scale factor with respect to the change of process parameters. Using two equations, DBRF can be obtained in general sheet metal forming process. To verify the reliability of the prediction model, results obtained from the prediction model are compared with them obtained from finite element analysis.