Innovation is a topic that has received much attention in the literature in recent years. For the most part, these articles have not solved an important problem facing the managers in today’s large organizations -- how to manage a portfolio of interactive product- and process- innovations, addressing the interrelated forces, including monetary constraints, manpower planning & technology capability, to a dynamic environment. By systems thinking of these problems, the author first set up a generic S.D. model as a Microcosm for portfolio analysis of technological innovations. Based on this Microcosm, an experiment aimed at pattern selection of product-& process- innovations was conducted, drawing the conclusion different from the famous Abernathy/Utterback’s. Finally, the mechanism of group decision on project selection of innovation portfolio using the Microcosm was explained, and the group decision support system was constructed.