1. Managing Information in PLM: The size and complexity of products, coupled with the scope of PLM (engineering, manufacturing, requirements, MBSE, quality management, etc.), mean that a PLM system must handle enormous quantities of information. This includes both business objects and the relationships between them, varying based on the granularity of the information.
  2. Graph Theory Application: Graph theory shoul be considered for this management but must be carefully studied and adapted. The PLM business objects are complex, polymorphic and subject to revisions, while the relationships are strongly typed, carry applicability information (like dates or batch numbers), and require interpretation for optimal navigation among objects.
  3. Digital Thread Management and Impact Analysis: Managing digital continuity and analyzing the impact of modifications necessitates a reasoned navigation through the graph of relationships. For graphs with tens of thousands of objects and hundreds of thousands of relationships, this task is nearly impossible for humans without powerful filters.
    • If filters are static, we revert to « classical » views (requirements, MBSE, engineering, manufacturing, quality), which diminishes the benefits of graph theory.
    • Dynamic filters, defined according to the context or type of change, enhance search efficiency and navigation. In such cases, AI should be strongly considered, as it can provide a holistic and comprehensive view of digital continuity and the impacts of modifications.

Conclusion:This is where artificial intelligence (AI) plays a crucial role. AI can provide a holistic and exhaustive perspective on digital continuity and the implications of modifications, something indispensable in complex PLM environments. The integration of AI in PLM, particularly in conjunction with graph theory, transforms what would otherwise be an overwhelming influx of data into a structured, navigable, and insightful resource. Graph theory alone, albeit useful, achieves its full potential only when complemented with advanced AI tools, making it an indispensable asset in modern PLM strategies and impact analyses.