Akademik Veri Yönetim Sistemi
IEEE Access, 2026 (SCI-Expanded, Scopus)
Modern manufacturing supply chains operate as complex systems-of-systems, where disruptions can propagate across organizational boundaries through ripple effects. However, existing digital twin implementations remain confined to single-facility settings and do not capture cross-functional disruption dynamics. This paper presents a Federated Supply Chain Digital Twin (FSCDT) architecture for modeling and simulating ripple-effect propagation in manufacturing supply chain networks. The framework comprises ten autonomous agents—Part, Supplier, Logistics, InventoryManagement, ProductionPlanning, Producer, QualityControl, EnergyManagement, Product, and Customer—communicating through a signal-based Event Bus supporting 61 signal types across seven categories. Behavioral specifications are formalized as ten state machines with 75 states and 125 transitions, encoding trigger–effect rules that enable automatic ripple propagation. The architecture is implemented using Model-Based Systems Engineering (MBSE) methodology with SysML in Modelio and exported via XMI for simulation execution. A Python-based simulation engine parses the XMI model and executes what-if scenarios, while a Streamlit dashboard provides interactive visualization and decision support. Experimental evaluation across 50 scenarios spanning Timing, Quality, Energy, Demand, and Multi-Factor categories reports 98% propagation coverage (49/50 scenarios producing at least one downstream state transition), with 1.64 average propagation depth and 112 total ripple steps, supporting verification of the modeled signal-trigger–effect specifications. This framework contributes to the goals of Industry 5.0 by enabling proactive risk assessment, cross-functional communication, and resilience analysis in interconnected production systems.