Simulation of Collaborative Innovation System in Cluster Environment

Abstract

Cluster innovation system (CIS) is a kind of complex adaptive system. The system performance of CIS depends critically on the interactions between its parts. Because it is hard to formalize and define the system equation, the analytical or deductive method is unfitted with the high dynamics, uncertainty, and complex structure of CIS. In the paper, swarm principles are used to deal with the complexity of CIS. The multi-agent model is established to represent the interaction, collaboration, group behavior, and the emergence of higher order system structure. Based on this multi-agent model, the emergent behaviors and the key characteristics of the system are obtained using swarm simulation. The key parameters for system evolution can be identified and the attribute features and behavior rules of an individual enterprise (agent) can be set up. The objective of the research is to find out the key factors influencing the evolution of CIS and the proper innovation strategy for CIS.