Development of models Models were established in order to optimize the production process by using the Statistical Analysis System Program. Multiple regression was used to obtain the estimates of the equation parameters, the analysis of variance and regression coefficient. The dependent variables were yield (YC) and other quality characteristics namely Total solids (TS), Protein content (PC) and Fat content (FC). The independent variables or processing parameters due to the method of processing are Blanching time (BT), amount of water added during grinding per kilogram of dry seed (AW) and the temperature to which the soymilk extract is heated (HT).The models were developed using linear and quadric functions by regression (SAS, 1996) to determine the equation with the highest R2 which indicated a higher regression between dependent and independent variables. Different plots of data were made between the independent and dependent variables to observe the trend and obtain possible equations. The dependent variables were plotted against the predicted data to check the predictive ability of the models.