Research on dynamic influence mechanism of port state control and ship risk level

As an accident prevention mechanism, Port State Control (PSC) is essential for the development of maritime transportation and security. To evaluate the effectiveness of PSC and provide recommendations for its enhancement, we develop Dynamic Bayesian Network (DBN) models using the latest five inspection records from the Tokyo and Indian Memorandum of Understandings (MoUs). These models incorporate a hidden variable called "ship condition" to represent the overall influence of four ship characteristic factors. Additionally, a "risk level" hidden variable is determined by two inspection results during each inspection: detention and the number of deficiencies. Two analyses are conducted in this paper. One aims to identify changes in risk levels, thereby investigating the effectiveness of dynamic inspections. The other involves comparing inspected and uninspected ships, with the goal of assessing the impact of PSC and ship condition on accident type, consequences, and severity. The results indicate that the inspection effects of the two MoUs demonstrate the same changing pattern, with the ship risk levels initially rising and then falling. However, the higher detention rate of the Indian MoU justifies its lower level of risk. Ships with inspection records are less likely to encounter capsizing accidents and have a lower probability of being taken out of service, while collision accidents and casualties are more likely to occur. The findings provide valuable insights and practical suggestions on how to enhance the effectiveness of PSC.