A Unified Approach for Reliability Assessment of Critical Infrastructures using Graph Theory and Entropy

The growth in the critical infrastructure (CI) complexity, interdependency, and data flows has required enhanced reliability assessment methods. This paper proposes a reliability assessment model which considers various sources affecting the CI performance, including internal and external fault sources including natural disasters and malicious attacks. The paper proposes a hybrid data and model-based approach for a comprehensive CI reliability assessment, which considers fault source effects on CI subsystems and their interdependencies. The Shannon entropy concept is proposed for detecting any variations in the system information fault sources. The graph concept is used to determine interdependencies of CI subsystems. We have presented two case studies to illustrate the effectiveness of the proposed method, and compare the results with similar studies.