Modeling Complexity in New Product Development: Decisions and Dependencies in Team-Based Design Projects

New product development (NPD) is a key activity for generating and sustaining a competitive advantage for firms. Many business processes aim for standardization and repeatability (e.g. manufacturing) and can be modeled using “paths” or sequences of events. However, in NPD, the analogy of a “path” is less applicable because there are many interdependent and multi-disciplinary tasks, many of which are novel. This dissertation explores the issue of how to model the NPD process. Particular attention is devoted to examination of the NK model—a popular model of interdependent nodes in a network—and its applicability as a potential framework for modeling NPD activity. First, the assumptions that underpin the NK model are reviewed. Several of these assumptions are not congruent with real world NPD. The NK model is then adapted to account for experimentation costs and exploration of the design space. It is shown how the search for a better product configuration is moderated by these two factors. Next, the NK model is extended for NPD by incorporating two realities of NPD management: 1) generally, knowledge exists regarding whether a dependency between components is complementary or conflicting in nature and 2) the outcome of design changes are uncertain, but not entirely random. It is shown that the nature of dependencies within a system is more important than the sheer number dependencies or interactions in a system. It is also shown that product improvements are less rapid than the original NK model would suggest. Lastly, the organizational structure of an NPD project is analyzed to assess its impact on the modeling of such a project. It is shown that the co-evolutionary nature of multiple, interdependent teams are more appropriately modeled using the NKC model, a variant of the original NK model. It is shown that the degree to which teams are coupled has a significant influence on project outcomes—a result that is not evident with the original NK model. A heuristic that helps to overcome the unstable nature of tightly coupled design teams is proposed. A discussion of the theoretical and managerial implications of the research concludes the dissertation.