Using network science to support design research: From counting to connecting

Abstract

A network-based perspective on designing permits research on the complexity of product, process, and people interactions. Strengthened by the latest advances in information technologies and accessibility of data, a networkbased perspective and use of appropriate network analysis metrics, theories, and tools allow us to explore new data-driven research approaches in design. These approaches allow us to move from counting to connecting, meaning to explicitly link disconnected pieces of data, information, and knowledge, and thus to answer far-reaching research questions with strong industrial and societal impact. This chapter contributes to the use of network science in empirical studies of design organisations. It focuses on introducing a network-based perspective on the design process and in particular on making use of network science to support design research and practice. The main contribution of this chapter is an overview of the methodological challenges and core decision points when embarking on networkbased design research, namely defining the overall research purpose and selecting network features. We furthermore highlight the potential for using archival data, the opportunities for navigating different levels of the design process that network analysis permits, what we here call zooming in and out, and the use of network visualisations. We illustrate the main points with a case from our own research on engineering communication networks. In this case, we have used more than three years of archival data, including design activity logs and work-related email exchanges from a recently completed large-scale engineering systems project of designing and developing a renewable power plant.