An integrated risk-based modelling approach for prediction of AIS spoofing attacks on-board ship

Automatic Identification System (AIS) spoofing poses a significant threat to maritime navigation and safety, particularly in congested waterways and critical sea lanes. This study proposes an integrated risk-based modelling approach combining the Evidential Reasoning-enhanced Human Error Assessment and Reduction Technique (ER-HEART) with Fault Tree Analysis (FTA) to systematically predict and evaluate the risk of AIS spoofing attacks on-board ship. The model addresses both technical failure modes and human error probability factors contributing to the susceptibility and consequences of spoofing incidents. In the method, the ER-HEART is utilized to quantify human error probabilities under uncertainty, incorporating expert judgments and context-specific error producing condition (EPC), while FTA decomposes the spoofing risk into logical causal pathways of component and system failures. The integrated methodology allows for a comprehensive assessment of the likelihood and potential impact of spoofing events by synthesizing subjective and objective evidence sources. Results indicate that occurrence probability of AIS spoofing attack risk on-board ship is 5.502E-01 which is indicating a high level of vulnerability. The outcome of the research provides valuable insights for ship operators, maritime cybersecurity planners, and regulatory bodies aiming to strengthen on-board situational awareness and resilience against AIS spoofing attacks.