Simulating I-Space (SimISpace): An Agent-based Approach to Modeling Knowledge Flows

Knowledge management often generates theories that are too general or abstract to be easily testable. In some cases, simulation modeling can help. In this paper we develop an agent-based simulation model derived from a conceptual framework, the Information Space or I-Space and use it to explore the differences between a neoclassical and a Schumpeterian information environment. After introducing the knowledge management issues involved, we first briefly present the conceptual framework. This is followed by a presentation of the agent-based model and a development of a number of hypotheses designed to help validate the model. We then present a number of model runs designed to test these hypotheses in a preliminary way. We find broad support for the hypotheses and conclude that the model exhibits enough consistency to warrant further development.