Characterization and Modeling of Product Family Evolution

Open Access
Park, Kijung
Graduate Program:
Industrial Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
March 03, 2017
Committee Members:
  • Gül E. Okudan Kremer, Dissertation Advisor
  • Gül E. Okudan Kremer, Committee Chair
  • Soundar R. T. Kumara, Committee Member
  • Conrad S. Tucker, Committee Member
  • Réka Albert, Outside Member
  • Product Family Evolution
  • Product Family Design
  • Product Family Architecture Network
  • Network Science
  • Network Characterization
  • Dynamic Network Modeling
Products continuously change over time to satisfy new customer needs and technologies. This successive emergence of new products within a product family can be considered as product family evolution, which is derived from inherent evolving properties and adaptation to dynamic product environments. Although product family design has been widely discussed, product family evolution has not been sufficiently explored in an analytical manner. To tackle this issue, this research aims to demonstrate the underlying properties of product family evolution and to model the process of product family evolution. The methodology of this research is based on a network science approach and consists of two phases: 1) characterization of product family evolution and 2) modeling of product family evolution. In the first phase, a network representation for product family architectures is devised to effectively describe relationships between components. Then, topological properties and significant network motifs in the product family network at each time period are investigated through a case study using the smartphone models of a major company. The results show that both individual product and product family networks follow the properties observed in other real networks. The network properties and significant motifs of the individual smartphone networks suggest that the individual product structure grows as a scale-free network with common building blocks. On the other hand, the smartphone product family network evolves with both scale-free and small-world properties due to commonality in a product family. Also, the product family network changes with common and specific motifs during each specific time duration. In the second phase, the complex adaptive process of product family evolution is simplified and modeled to estimate configurational changes in product family architecture design. In this model, evolution modes embedding specific variant generation mechanisms are realized through a mixed-strategy game theory and an evolving network model - fitness model. For the proposed product family evolution model, illustrative examples are introduced to describe the dynamic configuration process of a product family structure. Finally, evolutionary opportunities for product and product family design focusing on the universality of network properties are addressed from the perspectives of topological robustness and complexity. From theoretical and empirical evidences, it is identified that a design structure with scale-free network topology and commonality has topological robustness since it has less complexity in structural connectivity and variety. Given the increasing awareness of dynamic product environment, this research serves as a foundation for the analysis of product family evolution.