<![CDATA[This research develops and evaluates an Automatic Identification System (AIS)-based probabilistic collision risk assessment model as both an operational decision-support tool and an innovative educational resource for maritime safety and traffic management courses. The study processes large-scale AIS data from a designated Traffic Separation Scheme to derive traffic density patterns, encounter scenarios, and near-miss incidents based on ship domain theory and relative motion criteria. A probabilistic framework combining encounter frequency and consequence severity estimates collision risk levels across route segments and temporal variations. The model is integrated into classroom instruction and simulator-based exercises, in which maritime cadets, VTS operators-in-training, and safety professionals analyze high-risk zones and evaluate traffic management strategies. Through thematic analysis of interviews with 15 maritime experts, 22 maritime lecturers, and 18 maritime graduates, complemented by pre- and post-intervention assessments, the research demonstrates significant improvements in learners' risk-based decision-making capabilities and spatial awareness of collision hazards. Findings reveal that mean competency scores increased from 65.4% to 87.2% post-intervention, while stakeholder feedback confirms the model's practical relevance for operational safety planning, effectively bridging theoretical maritime education with evidence-based traffic management requirements.]]>
