An AIS Data-Based Framework for Ship Allision Risk Assessment Near the Offshore Wind Farm

Abstract

In order to investigate the allision risk of vessels near the Offshore Wind Farm (OWF) under the complicate surrounding environment, an AIS data-based allision risk framework is presented in this paper. The Bayesian learning approach is employed in this framework to learning the AIS database. The BN model of ship navigation behaviour near the OWF is then obtained. Meanwhile, the risk model that based on the output data from the BN under different combinations of navigation conditions is then used to assess the allision risk for new scenarios. A case study of an OWF is used to illustrate the application of the proposed model and the reliability of the new approach is tested. As a result, the proposed framework presents the evidence that impacts of OWF to vessels are diverse, which depend on ship categories, size and seasons. Fishing ships and OWF maintenance ships have smaller passing distance and that for oil and gas vessel is larger. Furthermore, ten cases have been ranked by using the proposed risk model. The developed risk allision framework from the empirical database provides a sufficient approach to evaluate the ship risk near the OWF, and further support the OWF safety management and ship navigating.