A digital controller for the regulation of glucose in the metabolic system was achieved through an open-loop scheme in which the continuous measurement of the actual glucose concentration is not required, assuming the model is sufficiently accurate and the controller is robust. The analysis of individual human subjects has shown great differences in their responses to glucose infusion. Consequently there were also distinct differences in the optimized controller parameters. In principle, such differences lead to significant differences in the closed-loop dynamics. A fourth-order digital model was obtained from the state differential equations which represent the system metabolism. The sample time was chosen sufficiently large to suit such a slow system. The zeros of the controller were selected to match the undesired complex poles of the metabolic system and the controller gain was determined using the root locus portrait. A set of well-behaved performances together with their corresponding internal states were obtained with variable overshoot. The controller was shown to be robust and the results obtained suggest the application of the developed open-loop scheme for glucoregulation in vivo.