The structural variables of multi-layer networks and early warning mechanism for industrial cluster risks

Compared with researches on the competitive advantages of industrial clusters, the studies which focus on the risks of industrial clusters are very limited. In this paper, we demonstrated the multi-layer network structures of industrial clusters, and identified clustering coefficient, average length of geodesic path and skewness of degree distribution as key structural variables which could effectively measure potential cluster risks. With theoretical analysis and computer simulation, we found that industrial clusters with multi-layer inter-firm networks which were characterized by high values of clustering coefficient, geodesic path length and skewness were more fragile under cluster risks. Based on our findings, we proposed an early warning mechanism for industrial cluster risks.