Developing a complex product in a concurrent engineering environment requires managing information flow among ten or even hundreds of people of different specialties organized in a large number of cross-function product development (PD) teams. This paper proposes an approach based on centrality measures in social networks for modeling and analyzing the information flows among PD teams. One advantage of this approach is that it enables PD project managers to identity the key PD teams based on a proposed new classification system that classifies PD teams under four categories: autonomous, receivers, transmitters, and transceivers.  A demonstrative example adapted from the literature is presented.