This study presents a numerical approach to predict the flow resistance coefficient in a segment ball valve. If more than a few valves are installed in a pipe system, the flow resistance coefficient should be considered in selecting the valves. Thus, the design of a segment ball valve incorporating the prediction of the flow resistance coefficient is suggested in this study. A segment ball valve incorporates the advantages of a ball valve and a butterfly valve. To predict the flow resistance coefficient of a segment ball valve, two approximate methods are introduced: second-order response surface method (RSM) and kriging interpolation method. The metamodels, built by EXCEL program and determined from the two models, were compared with the true numerical model. This study proposes the most suitable approximation method for the prediction of the flow resistance coefficient in a design of valve.