基于贝叶斯网络的浙江沿海船舶通航风险分析  

Navigation Risk Analysis for Zhejiang Coastal Waters with Bayesian Network

作　　者：余静 蒋惠园[1] 胡佳颖[3] YU Jing;JIANG Huiyuan;HU Jiaying(School Institute,of Communication,Wuhan University of Technology,Wuhan 430063,China;Scientific Research Ministry of Transport of the People＇ s Republic of China,Beijing 100029;School of Navigation,Wuhan University of Technology,Wuhan 430063,China)

机构地区：[1]武汉理工大学交通学院,武汉430063 [2]交通运输部科学研究院,北京100029 [3]武汉理工大学航运学院,武汉430063

出　　处：《中国航海》2018年第2期97-101,共5页Navigation of China

摘　　要：针对浙江沿海水域通航环境复杂、商船和渔船航行矛盾突出等问题,面向该水域提出基于事故特征的船舶通航风险贝叶斯网络模型。运用风险要素相关性分析方法和贝叶斯概率模型,建立浙江沿海船舶通航风险评价模型;结合事故数据和专家的主观评价,实现对浙江沿海水域船舶通航风险的定量化计算。通过误差分析和灵敏性分析验证模型的可用性和可靠性,进一步通过模型中节点变量状态对通航风险的影响分析提取通航风险关键因素。在分区域通航风险研究的基础上,利用贝叶斯方法对某大型水域通航风险进行定量化计算,为进一步研究浙江沿海水域通航风险细粒度特征和安全保障措施提供参考。In view of the complex navigation environment and prominent navigation conflict between commercial vessels and fishing boats in the coastal waters along Zhejiang Province,a Bayesian network model of ship navigation risk based on accident characteristics is proposed. The risk evaluation model of ship navigation particularly for Zhejiang coastal waters is built based on the correlation analysis of risk factors and Bayesian probability model. Quantitative analysis is conducted through combining accident data with subjective evaluation of experts. The usability and reliability of the model are verified by error analysis and sensitivity analysis,and the key risk factors are identified through analyzing the influences of the states of the nodes in the model. For great water area divided into segments,the risk assessment of total water area is obtained by integrating the risk assessments of segments with Bayesian method,which is used in a waterway assessment for verification.

关 键 词：通航风险 相关性分析 贝叶斯网络 敏感性分析 浙江沿海水域 

分 类 号：U698.6[交通运输工程—港口、海岸及近海工程]