The association between network centrality measures and supply chain performance: The case of distribution networks

We analyze transport data on the worldwide distribution network for 1.2 million vehicles manufactured and distributed by a large German car manufacturer in half a year. To identify central nodes in this network, we calculate various centrality measures from social network analysis. We then analyze the association of these centrality scores and the key performance measures related to stay-times and inventory for ports, distribution centers, and plants. Our main result shows that nodes with high degree centrality perform worse than less central nodes. The main theoretical contribution of our research is to confirm for the very first time that network theory applies to distribution networks, i.e. that network structure influences network node performance.