Security engineering.
From systems to adversaries.

Textile and clothing manufacturing environments where production orders move through multiple operations and physical items must remain traceable throughout production.

Can existing object-level Digital Twins provide enough information to infer higher-level twins representing manufacturing orders and their work-in-progress state?

A layered Digital Twin platform where information from lower-level industrial objects is aggregated and contextualized into richer representations of manufacturing activity.

Demonstrated the feasibility of deriving higher-level manufacturing Digital Twins from existing lower-level twins, improving visibility and work-in-progress traceability.

NetworkX provides established shortest-path algorithms, including A*, but several specialized A* extensions are useful for dynamic, memory-constrained and real-time pathfinding scenarios.

Implemented and evaluated IDA*, D* Lite, SMA*, Bidirectional A* and Real-Time Adaptive A* within the NetworkX ecosystem.

Implementations were integrated with the NetworkX codebase, validated through its testing infrastructure and benchmarked using path cost, execution time, memory usage and dynamic recalculation metrics.

The work produced validated implementations and demonstrated how different A* extensions can provide advantages over traditional A* depending on graph structure, memory constraints and environmental changes.