Dynamic analysis of pilot transfer accidents

Abstract

Marine pilot is the person who expert on a specific region and guide the ship masters for safe navigation. Pilot transfer (PT) operation is a process of embarkation or disembarkation of pilots from pilot boat to ship or viceverca which is very risky operation and accidents even cause death of the pilots sometimes happen. The nature of the operation require pilots transfer between two movement vessel relatively in open sea conditions expose to different weather environment along day and night. Given the dearth of international research on pilot transfer operations, this study seeks to combine fault tree analysis with bayesian network analysis using a fuzzy technique in order to analyze accidents that occur during pilot transfer operations. When first investigating the root causes and causal chain of the PT accidents, fault tree analysis (FTA) was used. The probability of crucial occurrences are often cloudy, but experts' judgment and imprecise ideas can help. A Bayesian network (BN) mapping is made to get around the problems caused by FTA's conditional dependencies and lack of flexibility. Sensitivity analysis is used to help BN simulate the analysis using the entire operating procedure. The study's findings highlighted the root nodes “poor condition of pilot ladder”, “lack of safety culture”, “adverse sea/swell” and “securing failure of pilot ladder on deck” as major contributors to PT accidents. The results show that the study's methodology is sound and practical, and that adding FTA-BN to an analysis of PT accidents will make a substantial contribution to the literature.