Correlation of Collision Damage Statistical Properties by Gaussian Copula Functions

Oil spills from tankers decreased significantly in recent years, but they still occur. After a brief review about the methods and procedures implemented to prevent oil spills from tankers, the paper focuses on the statistical properties of oil spillage occurrences following collision events. Starting from the probability density functions provided by the IMO guidelines for longitudinal, transverse and vertical damage extents, the paper focuses on the employment of the Gaussian copula function to investigate the joint probability between the above-mentioned random variables, by Monte Carlo simulation. As a result of the correlation analysis, it was found that copula function are a reliable method to generate statistically correlated random variables, starting from the relevant unimodal distribution and avoid unreliable collision scenarios.