SMEs Cluster Network Evolution:A Multi-Agent Simulation

The behavior of SMEs throughout the whole process of cluster evolution and determine the network’s development trends and results. Based on Social Network Theory, we first from the embedding and embedded dimensions definite the "resource-oriented" enterprise network behavior. Subsequently, quantify the "resource-oriented" behavior of each SME by the way of network power distribution, then build multi-agent simulation model of SME Cluster network evolution. Finally, analyze the characteristics and laws of the cluster evolution based on simulation results. The main conclusions are as follows:(1) Different-oriented business practices in total will result in different cluster network structure.(2) Among inner SMEs, a "power gap" will appear under the effect of competition. And over time, the gap may be enlarged. But not all clusters will have dominant firms.(3) From long-term development, the initial state has nothing to do with the evolution of SMEs cluster. During this time, enterprises will choose according to the nature of their own resources partners to establish relatively stable exchange rings. And in different growth stage, there exists varying degrees of faction phenomenon;(4) SMEs cluster network evolution is process alternating self-organization and self-disorder. And the system stability mainly depends on common action of firms and time span;(5) Firm cluster has an optimal efficiency range in pave whit the trajectory the cluster evolving under "resource-oriented".