Incorporation of deficiency data into the analysis of the dependency and interdependency among the risk factors influencing port state control inspection

Abstract

Port State Control (PSC) inspection aids to control substandard ships and ensure safety at sea. Current risk-based PSC research and practice fail to incorporate ship deficiency records into detention probability analysis, because of the difficulty introduced by the involved big deficiency data. In this paper, a new Bayesian Network (BN) based PSC risk probabilistic model is developed to analyze the dependency and interdependency among the risk factors influencing PSC inspections based on big data derived from the inspection database of Tokyo MoU for the period between 2014 and 2017. The results reveal that ship's safety condition related deficiencies as well as technical features of the inspected vessel itself are among the most influential factors concerning PSC inspections and ship detention. New Bayesian learning methods are used to improve the model efficiency in ship detention prediction. As a result, the newly developed model has shown a reliable performance on dynamic prediction and cause-effect diagnosis of ship detention probabilities by pioneering the incorporation of ship deficiency records in the analysis. The findings provide important insights on how to facilitate risk-based PSC inspections for both ship owners and port states. They provide support for port state authorities to implement rational inspection policies.