Enhancing semantic belief function to handle decision conflicts in SoS using k-means clustering

Area of Science:

• Computer Science
• Systems Engineering

Background:

• System of Systems (SoS) integrate individual systems for complex functions, aiming for simplified operations and cost reduction.
• Co-integrating systems can lead to conflicts, undermining the core benefits of SoS.
• Existing SoS approaches require enhancement in conflict detection and resolution times.

Purpose of the Study:

• To improve the efficiency of detecting and resolving conflicts arising from co-integrating systems within a System of Systems (SoS).
• To propose an enhanced SoS model that facilitates faster conflict management and accommodates future system additions.

Main Methods:

• Utilized k-means clustering to partition the SoS into smaller, manageable clusters (Sub SoS or S-SoS).
• Developed the Smart Semantic Belief Function Clustered System of Systems (SSBFCSoS), an advancement over the Ontology Belief Function System of Systems (OBFSoS).
• Applied the SSBFCSoS to partition SoS entities, enabling localized conflict detection and resolution.

Main Results:

• The SSBFCSoS demonstrated rapid conflict detection and resolution capabilities.
• Achieved an average of 89% conflict resolution, outperforming other approaches (77%).
• Showcased a 16% acceleration in conflict resolution and a 23% reduction in recurring conflicts compared to previous methods.

Conclusions:

• The proposed SSBFCSoS effectively enhances conflict management in SoS environments.
• The clustering approach improves scalability and maintains SoS objectives.
• SSBFCSoS offers a significant improvement in conflict resolution rates and efficiency.