Why good processes sometimes produce bad results: A formal model of self-reinforcing dynamics in product development




Nelson P. Repenning

Department of Operations Management/System Dynamics Group
Sloan School of Management
Massachusetts Institute of Technology
E53-339, 30 Wadsworth St.
Cambridge, MA USA 02142
Phone 617-258-6889; Fax: 617-258-7579; E-Mail:<nelsonr@mit.edu>


July 1999


version 2.0




Work reported here was supported by the MIT Center for Innovation in Product Development under NSF Cooperative Agreement Number EEC-9529140, the Harley-Davidson Motor Company and the Ford Motor Company. Extremely valuable comments have been provided by Andrew Jones, John Sterman, Scott Rockart , Rogelio Oliva and Steven Eppinger. Minsoo Cho developed the web site that accompanies this paper. Special thanks to Don Kieffer of Harley-Davidson for providing the catalyst for this study.






Abstract



The design of product development processes has received considerable attention from both scholars and practitioners. Unfortunately, however, practice does not always follow theory: many organizations continue to experience difficulty in following the development processes prescribed in the literature. One source of such difficulties is the existence of self-reinforcing processes that drive the organization to allocate an increasing fraction of its scarce resources towards fixing problems in existing projects and away from preventing problems in future ones. While the existence of such dynamics has been widely acknowledged (popular labels include vicious circles, death spirals, and doom loops), they remain poorly understood. In this paper I develop a simple model of a self-reinforcing process in a multi-project product development environment. The analysis of this model leads to a number of new insights. First, it provides boundary conditions for the phenomenon. Second, it shows that such dynamics can work as either virtuous or vicious cycles and provides the conditions under which each of these will occur. Third, the analysis highlights the role that resource utilization plays in determining the robustness of the system to undesirable behavior modes. I also consider one extension to the basic model, the introduction of new tools and processes, and discuss policies for system improvement.






Download the paper and the model built for the paper.

Start the tour of the model's technical documentation.






Abstract

Overview | Advanced Work | Current Work | Current Defects | Market Defects | Capacity Allocation

Download | Simulation