On the Use of Network Analysis in Product Development Teams

In this paper, we consider the product development process as a network of interacting elements (e.g. development participants, physical subsystems/components, or development activities) exchanging information in order to achieve the common goal of developing a successful product or service. Drawing from network analysis (NA) techniques, we consider three network measures: single-node centrality, group-centrality, and the key-player problem. Using these measures, we determine a small subset of nodes within the network that is most important to the flow of information. That is, these nodes significantly control (i.e. receive, distribute, or process) more information than any other node in the network. Identification of this subset of nodes is essential to devise improved management strategies for information flow within the product development process. We find that when using NA techniques to analyze product development processes it is important to consider all three analysis measures because the different measures produce a different subset of top scoring nodes. We discuss some of the underlying reasons for these differences and conclude that the convenient measure(s) to use should be based on the particular development environment and the underlying managerial objectives. We demonstrate these measures and results by studying the development process of a large commercial aircraft engine.