Cyber-to-physical framework assesses water system security
Original title: Closing the Applied Cybersecurity Gap in Critical Infrastructure: A Cyber-to-Physical Security Assessment Framework for Water Systems
Machine learning model predicts physical impacts from cyber compromises in water networks using hydraulic simulations.
The study evaluates an expanded Applied Cybersecurity Gap framework for water infrastructure using 2160 synthetic trials across three benchmarks. A source-trained machine learning model predicts physical-impact outcomes from declared cyber-attack rules with higher ROC-AUC than a scenario-only baseline. Additional sensitivity analysis preserves most physical-impact labels and demonstrates the ACGI-CPPR association is partly induced by simulation design.
- ML model achieves ROC-AUC 0.772 on Net3 trials
- 717 of 720 physical-impact labels preserved in sensitivity analysis
- ACGI-CPPR association partly induced by simulation design
- Reproducible simulation-based assessment workflow
From the original article
1 min read at arxiv.orgThe integration of information technology and operational technology in water infrastructure creates pathways through which cyber compromise may affect physical service delivery.
By Halil Dursunoglu · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.