한국해양대학교

Detailed Information

Metadata Downloads

DP선박의 안전한 운용을 위한 FSA의 적용

Title
DP선박의 안전한 운용을 위한 FSA의 적용
Alternative Title
An Application of FSA for Safe Operation of Dynamic Positioning Vessels
Author(s)
채종주
Publication Year
2016
Publisher
한국해양대학교 대학원
URI
http://kmou.dcollection.net/jsp/common/DcLoOrgPer.jsp?sItemId=000002302893
http://repository.kmou.ac.kr/handle/2014.oak/8219
Abstract
DP vessel operation has a lot of risks in nature. However, the number of it has been on the steady rise as the demand for marine resource development and special work increases in offshore industry. In this regard, many studies have been conducted to ensure a safe operation of the DP vessel. Regardless of this, Formal Safety Assessment(FSA) on the DP vessel has not been applied.

FSA is a structured and systematic methodology, aimed at enhancing maritime safety, including protection of life, health, marine environment and property, by using risk analysis and cost-benefit assessment. The assessment has been applied to major merchant vessels such as cruise ship, oil tanker, bulk carrier, container ship and liquified gas tanker, thereby recommending various new regulations for maritime safety and protection of the marine environment or making a comparison between existing and possibly improved regulations, with a view to achieving a balance between the various technical and operational issues. Furthermore, the need to apply FSA to human elements also rises as many studies on vessel-related accidents arising from human elements have been in progress.

This thesis is designed to identify major causes of LOP incidents directly linked to the safe operation of DP vessel and FSA on the vessel has been used in an effort to consider the countermeasures.

For this, 612 reports on LOP incidents released for the last decade (from 2001 to 2010) and reported to IMCA were analyzed to identify major causes of the incidents. The result shows that 22.1% of LOP incident occurring to DP vessel resulted from PRS error, which is the most frequent cause of it.

LOP incidents caused by human error reveals to be 11.8%, which is relatively low rate. However, the detailed analysis discovered that human error caused a total of 375 LOP incidents(61.2%) both in an indirect and a direct way. Based on this analysis, FSA on PRS and human error was applied in this study.

In applying FSA to PRS error, a frequency analysis and Bayesian Network contributed to identifying major risks. The installation of PRS as redundancy to all DP vessels and type specific training on the installed system in the vessel were recommended as RCOs in order to control identified risks. Furthermore, cost-benefit assessments of the suggested RCOs was assured. It was found that the use of a simulator is very effective as a way to verify the effect of applying RCOs.

In using FSA on human error, a frequency analysis, HFACS and Bayesian Network found out that conditional probability of drift off & drive off(74.3%), operation abort(51.9%) and time loss(64.0%) caused by skill-based human error. Unsafe supervision also made a big impact on the occurrence of the LOP incident. To control such identified risks, the following measures were suggested: make DPO training mandatory according to STCW convention, implement various onboard training by installing a DP simulator on the vessel, establish a process to ensure the understanding on a safety operation procedure of DP vessel and its update. Furthermore, cost-benefit assessments of the suggested RCOs was assured.

Bayesian Network and HFACS used to apply FSA to identify PRS and human error revealed to be a useful tool to recognize and analyze the related risks. Particularly, the graphical expression of Bayesian Network via GeNIe easily recognized prior probability and conditional probability of LOP incidents of DP vessel. Also, The classification of human error by using HFACS and the application of Bayesian Network is found to serve as a useful quantification tool of human error.
Appears in Collections:
항해시스템공학과 > Thesis
Files in This Item:
000002302893.pdf Download

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse