Integrating data-limited techniques for maritime risk assessment in Small Island Developing States

Abstract

Maritime navigation is critical for the economic development of Small Island Developing States, yet resource constraints often hinder comprehensive risk assessments. This study developed a probabilistic risk assessment method using publicly available data, including historical traffic patterns and port call logs, to identify turning zones and traffic routes. The Monte Carlo approach and Poisson distribution were used to simulate traffic events, with vessels assumed to drift within a one-hour period. Overlapping safety zones were identified as potential candidates for drifting collisions, and fault tree analysis was used to calculate causation probabilities. The study results highlight port approaches, narrow waterways and turning zones as areas with the highest incident probability, followed by areas to the north of Trinidad and, separately, to the north of Tobago. This scalable model provides actionable insights, aiding policymakers and maritime professionals in prioritising resources and mitigating navigation risks.