首页 | 本学科首页   官方微博 | 高级检索  
     

基于贝叶斯网络模型的船舶搁浅事故分析
引用本文:肖仲明,王新建,章文俊. 基于贝叶斯网络模型的船舶搁浅事故分析[J]. 安全与环境学报, 2017, 17(2): 418-421. DOI: 10.13637/j.issn.1009-6094.2017.02.004
作者姓名:肖仲明  王新建  章文俊
作者单位:大连海事大学航海学院,辽宁大连,116026;大连海事大学航海学院,辽宁大连,116026;大连海事大学航海学院,辽宁大连,116026
基金项目:辽宁省科学事业公益研究基金项目
摘    要:根据海上交通安全事故统计数据显示,船舶搁浅事故是主要的海上交通安全事故之一。为了研究船舶搁浅事故的发生机理,收集了我国东部海域部分海事局辖区的船舶搁浅事故样本。首先从人为因素、环境因素和船舶因素三方面分析导致船舶搁浅事故的原因,确定网络节点。然后找出每起搁浅事故的事故链,并建立船舶搁浅事故贝叶斯网络模型。再对建立的贝叶斯网络模型进行仿真,选取21起船舶搁浅事故对建立的模型进行验证,确定模型的有效性。最后,利用HUGIN软件找出船舶搁浅事故各影响因素的致因概率,得到导致船舶搁浅事故的致因链。在人为因素方面,瞭望不当和不熟悉航道情况占据的比例最高;在船舶因素方面,舵机设备故障对船舶搁浅事故影响最大;在环境因素方面,风/浪/流对船舶搁浅事故影响最大。

关 键 词:安全工程  贝叶斯网络  船舶搁浅  致因链

Analysis for the ship grounding accidents based on the Bayesian network model
XIAO Zhong-ming,WANG Xin-jian,ZHANG Wen-jun. Analysis for the ship grounding accidents based on the Bayesian network model[J]. Journal of Safety and Environment, 2017, 17(2): 418-421. DOI: 10.13637/j.issn.1009-6094.2017.02.004
Authors:XIAO Zhong-ming  WANG Xin-jian  ZHANG Wen-jun
Abstract:In this paper,we have collected 284 samples of ship grounding accidents that have happened in the six MSA administration zones in China coastal waters,trying to analyze such aground events.Of all such ship aground accidents,as the cases of Liaoning MSA,Hebei MSA,Shandong MSA,Shanghai MSA,Zhejiang MSA,Fujian MSA,the reasons for the aground accidents can be divided into three categories and 18 kinds through the primary analysis of the events.First of all,the factors that account for the ship grounding should be attributed to the human navigation factors,the ships own construction factors and the environment-affecting factors,when the network nodes can be determined by eliminating some less relevant factors.And,secondly,there is a great need to trace and determine the grounding accident chains of each ship step by step consequently.It is just because of all the above reasons that we have established causative chains and consulting expertise,the Bayesian network model about the ship grounding while deleting the little probability chains and the conditional probability of all the nodes through the synthetic analysis of the parameters exploitation of the samples concerned.And,next,we have simulated the ship aground accidents are by Bayesian network model and 21 ship aground accidents chosen to validate the model,which can help to ensure the validity of the model of ship aground and to demand reasoning.And,last of all,we have managed to gain the probability of the influential factors and the maximum causation chain of ship grounding accidents with the software named HUGIN.Thus,it can be concluded that the maximum causation chain can be obtained in the following descending order:the improper outlook→ the information inadequacy→the inadequate risk assessment→ failure to take measures in time→ human interference factors→ the result of aground.At the same time,it has to be pointed out that human controlling factors,the improper lookout and unfamiliar with the channel situation should account for the biggest share for the aground accidents.As to the ship equipment factors,the steering gear equipment failure can account for a quite great proportion,in case when the environmental factors,such as wind/wave/current waves,may also take the greatest share of the failure event.Thus,it can be seen that the results of our analysis can be expected to serve as the principal reasons for the ship-aground and therefore highly useful for the relevant managers and researchers to prevent from aground accidents and improve the maritime transportation and communications safety.
Keywords:safety engineering  Bayesian network  ship grounding  causation chain
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号