A Fuzzy Rule-Based Ship Risk Profile Prediction Model for Port State Control Inspections

Abstract

Maritime transport is the backbone of the international trade, more than 80% of global freight transport is carried by ships on the seas. However, in a complex and high-risk environment at sea, substandard ships in maritime transport causes serious accidents and hence bring out many threats to the maritime industry. As a result, with the aim of detection and elimination the substandard ships, port state controls (PSC) have been developed to inspect the ships respect to their Ship Risk Profile (SRP). SRP helps to determine the ship’s priority for PSC inspections through categorizing the ships in high risk, standard risk or low risk using various generic and historic parameters. In this study a new model is proposed to predict SRP by using fuzzy clustering analysis (FCA) and fuzzy rule-based classification system (FRBS) different from the standard calculation of the SRP in the PSC regimes. The proposed model is structured on five different parameters which are ship type, ship flag, ship age, deficiency number and detention. In the proposed model, to predict the SRP, 53788 ship inspection data belonging to the parameters has been analyzed gathered from the Paris MoU online database between the years of 2017 and 2020. The results obtained with the proposed approach help to identify the risk profile for the ship targeted at almost certain risk level. As a result of the study, it is aimed to provide decision supports for port state control (PSC) officers to detect the most appropriate/risky ship for the inspection.