Understanding Safety of Navigation Deficiencies Leading to Vessel Detention: A Data-Driven Analysis of Black Sea Memorandum of Understanding Inspections

Abstract

Preventing safety of navigation (SN) deficiencies is of critical importance in waterborne transport since the SN Code constitutes one of the most recurrent deficiency categories that serve as a basis for detention under port state controls (PSC). In pursuit of this objective, this study initially delineates and critically evaluates the interdependencies between a spectrum of influential determinants and deficiency types, as well as their respective contributions to detention occurrences attributable to the SN code. For this purpose, a data-driven Bayesian network model is developed based on 5,979 vessel detentions documented in the Black Sea Memorandum of Understanding (MoU) archives between January 2003 and December 2024. The novelty of this study lies in the integration of a probabilistic graphical modeling approach (tree augmented naïve (TAN)-based Bayesian networks) with association rule mining to jointly capture both causal dependency structures and frequent co-occurrence patterns in PSC detention data, which has not been previously explored in the context of SN-focused deficiency analysis. Findings indicate that deficiencies related to labor conditions have the highest impact to explain ship detentions under the SN Code, increasing the likelihood by 16.1%, highlighting a strong interrelationship between crew welfare conditions and navigational safety performance. The study contributes methodologically by integrating Bayesian network learning (TAN) with association rule mining in a unified analytical framework for PSC detention analysis, and practically by providing a data-driven basis for targeted inspection prioritization and risk mitigation strategies for port authorities and maritime operators.