Using network science to disentangle supply networks: a case study in aerospace industry

Supply chains in the aerospace sector are becoming more complex than ever before, frequently causing delays on the production process. Complexity gave rise to the term “supply networks”, changing the way we view supply chains from a structural point of view. Structural properties are important to investigate as they help define robustness and efficiency of systems. Although complexity in structure is suspected by previous researchers who studied these networks, empirical data to characterise what complexity means, and how it effects properties of networks has been largely absent from literature. If empirical data is available, network science can be used to understand structural properties of such complex supply networks. Network science is a suitable Mathematical tool for analysing the complex relationships and collaborations in the network and summarizing the properties of network from a fundamental, structural perspective. In this report, the author will apply network science to analyse the structure of the Airbus supply network. Due to the lack of aerospace supply chain data, firstly an empirical database is built. Analysis then focuses on the real structure of Airbus supply network and identification of key firms or communities under two scenarios: a non-weighted network in which the value of link is either 1 or 0, and a weighted network in which the value of link presents the strength of relationships among firms. While the weighted network indicates more informed features of the supply network structure by considering the weight of relationships, the non-weighted network can help us understand fundamental patterns that determine the structure of the connections in the network. The analysis indicates the Airbus supply network carries a power law distribution, which means most resources are dominated by few firms, and the network is robust to random firm failure but vulnerable to hub failure. The network contains communities with strong relationships between them.These communities do not only belong to the same industry and same region but have emerged as the result of an interaction between the two effects. Some key firms in the network own significant power of control the supply chain and fiancial resources, occupying key positions that bridge communities in the network.The study presents key structural features of a large scale network using empirical data and act as a case example for using network science based analysis in supply chains.