A single screw extruder with inner restrictions that is self-heated by friction and viscous dissipation was investigated. The effect of the processing parameters: feed moisture, screw speed, and feed rate on the melt viscosity and die temperature rise were researched using a dual orifice die. The melt viscosity was also measured using an offline capillary rheometer. The offline data was compared to and used to predict the inline viscosity of the melt. Various methods such as multilinear regression and superpositioning were used to model viscosity as a function of the processing parameters. Next, the current analytical model was evaluated using Finite Element Methods (FEM) paired with the Newton-Rhapson Method. FEM was then used to further the current analytical model to consider 2-D flow as well as nonlinear temperature rise along the screw channel. Overall, as the screw speed increased while maintaining a similar degree of screw fill, the melt viscosity decreased and the temperature rise increased. Comparisons to the offline data at higher moistures were made, but not at lower moistures due multiple dependent variables producing nonlinear behavior along with scatter in the data.