Sensor-Based Cyber Risk Management in Railway Infrastructure Under the NIS2 Directive

Area of Science:

• Cybersecurity Engineering
• Railway Systems Engineering
• Risk Management

Background:

• Traditional railway reliability analysis often overlooks cyber-induced failures impacting data.
• Regulatory frameworks like the NIS2 Directive mandate robust cybersecurity measures for critical infrastructure.
• Static risk assessments fail to adapt to evolving cyber threats in real-time.

Purpose of the Study:

• To introduce a sensor-centric cybersecurity framework for railway infrastructure.
• To dynamically extend Failure Mode and Effects Analysis (FMEA) for cyber-physical systems.
• To bridge regulatory compliance (NIS2 Directive) with field-layer monitoring.

Main Methods:

• Developed a framework integrating FMEA with sensor data streams.
• Utilized a dataset from ~250 sensors across a 25 km railway segment.
• Simulated cyber anomalies (spoofing, replay, injection) and evaluated risk using RPN scores.
• Assessed anomaly detection via latency, RPN distribution, and alert responsiveness.

Main Results:

• Demonstrated dynamic recalibration of Occurrence (O) and Detectability (D) factors using real sensor data.
• Showcased reduction in undetected exposure windows through near-real-time risk evaluation.
• Provided evidence of measurable compliance documentation aligned with NIS2 Article 21.
• Validated the framework's ability to adapt risk indicators beyond static documentation.

Conclusions:

• Coupling FMEA with streaming telemetry establishes a verifiable risk-evaluation loop for railway cybersecurity.
• The framework supports a transition towards continuous, evidence-driven cybersecurity governance.
• Sensor-centric, dynamic risk assessment is crucial for modern railway infrastructure resilience.