A model-based systems engineering approach for developing modular system architectures

Modular system architectures are increasingly important to manufacturers as they seek to decrease system development lead time and as customers have come to expect systems that allow for frequent modular upgrades. There has been considerable research into how to design modules within a system architecture and how to assess the modularity of systems. However, with these established approaches, the modularity assessment is a stand-alone, post-architecting analysis. This research demonstrates how, using model-based systems engineering (MBSE), the assessment of the modularity of a system architecture can be incorporated into the architecting process. To facilitate this, a modularity ontology is developed to guide the semantics and syntax of extensions to the Systems Modeling Language (SysML). Model metadata from behavioural, structural, and state-based system elements is directly assessed using Design Structure Matrix (DSM) based analysis. Elements in the architecture are automatically assigned to modules and the system's modularity is assessed. The approach allows for the system architect to also consider non-functional characteristics in the formation of modules. The architecting and the modularity analysis of the system are completed in a single environment, using SysML parametric diagrams as the catalyst, allowing for a more holistic and complete understanding of the system and its modularity.

Automated summary

This research demonstrates how to incorporate the assessment of system architecture modularity into the architecting process using MBSE, providing guidance and evaluation through modularity ontology and DSM analysis. This approach allows for a more comprehensive understanding of the system and its modularity by considering non-functional characteristics in the formation of modules.