A Model for Set-Based Design at the System-of-Systems Scale with Approaches for Emergent Properties

Abstract

Large projects with long time horizons and substantial capital investments challenge designers and engineers. Inevitably, the project’s large scale leads to competing requirements for the design team to balance. Further, the prolonged lifetimes of these projects create uncertainty regarding the project’s value retention from delivery until decommissioning or disposal.

Set-based design (SBD) offers a solution to the first issue. The method allows designers to intelligently canvass a more extensive solution space using sets and ranges of characteristics to define the design instead of specific instances. The practice of SBD continues to grow, but few examples exist of projects at scale outside of theoretical studies.

Flexibility offers a solution to the latter issue. Designing for flexibility accounts for this uncertainty by intentionally integrating options in the architecture of a project. This practice gives future managers the right, but not the obligation, to execute these options. Unfortunately, however, flexibility is often treated as an emergent property.

This research reveals a case study of an instantiation intersecting these principles in the design of a naval vessel. It uses an action research approach to develop and execute an SBD process at a large scale that integrates specific emergent properties as sets of alternatives. It shares the philosophical basis of the process, the team structure formed, the steps of the method, and the documentation created to capture the knowledge. It contributes generalizable knowledge for consideration when a team may consider establishing their SBD methods, including starting small to test the process before growing, establishing the first sets, and managing communications.

This research contributes to the development of integration by intersection by sharing how we integrated our sets and what conditions allowed their intersection. It offers a special case of practical point design within SBD called benchmarks, which acted as virtual prototypes and created reusable design knowledge for future efforts. The research could not conclude that treating emergent properties as sets suited the SBD method, but it provides insight towards answering that question in future work, especially regarding how to form those sets so that they intersect appropriately with others.