Mutual information based prediction of level of collision incident in Yangtze river

doi:10.16265/j.cnki.issn1003-3033.2018.11.027

Abstract

Abstract: In order to predict the level of ship collision dangers and evaluate the consequences of collisions, the mutual information was used to improve the traditional BN and a prediction model of ship collision risk grades was built. Firstly, the risk factors of ship collision danger grades were identified based on the historical data on ship collision accidents. Second, the qualitative part of the BN was established by the mutual information and conditional mutual information values calculated to determine the independence and independence of conditions between risk factors, and the dependencies between the factors were determined. According to data on the 516 accidents, a conditional probability table(CPT) was obtained to determine the quantitative part of BN. Finally, the feasibility and accuracy of the model were validated by using the data on the Yangtze River historical accidents. The result shows that the prediction accuracy obtained using model is up to 94%, which can well predict the ship collision risk level.

Key words: ship collision, risk factor, risk rating forecast model, mutual information, Bayesian network (BN)

CLC Number:

X928.03

CHEN Kejia, MAO Zhe, WU Bing, FAN Shen. Mutual information based prediction of level of collision incident in Yangtze river[J]. China Safety Science Journal, 2018, 28(11): 168-175.

References

[1] 杜磊, 文元桥, 肖长诗,等. 自由航行海域船舶碰撞概率计算[J]. 中国安全科学学报, 2015, 25(1): 53-59.
DU Lei, WEN Yuanqiao, XIAO Changshi,et al. Calculation of collision probability of ships in free navigation seas[J]. China Safety Science Journal, 2015, 25(1): 53-59.
[2] 齐迹. 基于突变理论的海上交通风险预测研究[D]. 大连: 大连海事大学, 2014.
QI Ji. Research on maritime traffic risk prediction based on catastrophe theory [D]. Dalian: Dalian Maritime University, 2014.
[3] 李昊, 张庆年. 内河航运安全状态监测与指数体系研究[J]. 中国安全科学学报, 2014, 24(10): 94-100.
LI Hao, ZHANG Qingnian. Research on monitoring and index system of inland navigation safety status [J]. China Safety Science Journal,2014, 24(10): 94-100.
[4] 黄道正. 水上交通事故的空间分布与后果评级研究[D]. 上海:上海交通大学, 2015.
HUANG Daozheng. Research on spatial distribution and consequence rating of water traffic accidents [D]. Shanghai: Shanghai Jiaotong University, 2015.
[5] 陈伟炯, 卢忆宁, 张善杰,等. 跨海大桥船桥碰撞模糊Bow-tie风险评估方法[J]. 中国安全科学学报, 2018, 28(1): 87-92.
CHEN Weijiong, LU Yining, ZHANG Shanjie, et al. Fuzzy Bow-tie risk assessment method for bridge bridge collisions[J]. China Safety Science Journal, 2018, 28(1): 87-92.
[6] TRUCCO P, CAGNO E, RUGGERI F, et al. A Bayesian belief network modelling of organisational factors in risk analysis: a case study in maritime transportation[J]. Reliability Engineering & System Safety, 2008, 93(6): 845-856.
[7] LEIMEISTER M, KOLIOS A. A review of reliability-based methods for risk analysis and their application in the offshore wind industry[J]. Renewable and Sustainable Energy Reviews, 2018, 91: 1 065-1 076.
[8] KOROMILA I, NIVOLIANITOU Z, GIANNAKOPOULOS T. Bayesian network to predict environmental risk of a possible ship accident[C]. Proceedings of the 7^th International Conference on Pervasive Technologies Related to Assistive Environments, 2014: 1-5.
[9] 吴贤国, 丁保军, 张立茂,等. 基于贝叶斯网络的地铁施工风险管理研究[J]. 中国安全科学学报, 2014, 24(1): 84-89.
WU Xianguo, DING Baojun, ZHANG Limao, et al. Research on subway construction risk management based onBayesian network[J]. China Safety Science Journal, 2014, 24(1): 84-89.
[10] 彭长根, 丁红发, 朱义杰,等. 隐私保护的信息熵模型及其度量方法[J]. 软件学报, 2016, 27(8): 1 891-1 903.
PENG Changgen, DING Hongfa, ZHU Yijie, et al. Information entropy model of privacy protection and its measurement method[J]. Journal of Software, 2016, 27(8): 1 891-1 903.
[11] ZHANG Di, YAN Xinping, YANG Zaili,et al. Incorporation of formal safety assessment and Bayesian network in navigational risk estimation of the Yangtze river[J]. Reliability Engineering & System Safety, 2013, 118: 93-105.
[12] 张晓东, 张浩, 陈亮,等. 基于负二项回归的水上交通事故分析与预测[J]. 上海海事大学学报, 2013, 34(2): 8-12.
ZHANG Xiaodong, ZHANG Hao, CHEN Liang, et al. Analysis and prediction of water traffic accident based on negative binomial regression[J]. Journal of Shanghai Maritime University, 2013, 34(2): 8-12.
[13] STATHEROS T, HOWELLS G, MAIER K M. Autonomous ship collision avoidance navigation concepts, technologies and techniques[J]. Journal of Navigation, 2007, 61(1): 129-142.
[14] 巢国荣. 上海港黄浦江水域船舶交通安全评价[D].大连: 大连海事大学, 2002.
CHAO Guorong. Ship traffic safety evaluation of Huangpu river in Shanghai port [D]. Dalian: Dalian Maritime University, 2002.
[15] 郑中义, 李红喜, 陈涛. 船舶吨位与碰撞事故关系[J]. 大连海事大学学报, 2007, 6(增2): 158-161.
ZHENG Zhongyi, LI Hongxi, CHEN Tao. Relationship between ship tonnage and collision accidents [J]. Dalian Maritime University Journal, 2007, 6(S2): 158-161.
[16] 李红喜, 郑中义, 李猛. 客滚船船龄与船舶事故关系研究[C]. 船舶安全管理与防污染管理学术研讨会, 2008: 23-26.
LI Hongxi, ZHENG Zhongyi, LI Meng. Research on the relationship between the age of passenger rolling ships and ship accidents [C]. Symposium on Ship Safety Management and Pollution Prevention Management, 2008: 23-26.
[17] TSOU Mingcheng, KAO Shenglong, SU Chienmin. Decision support from genetic algorithms for ship collision avoidance route planning and alerts[J]. Journal of Navigation, 2009, 63(1): 167-182.
[18] 黄常海, 肖英杰, 高德毅,等. 定线制水域航行危险度评价模型及其应用[J]. 中国安全科学学报, 2014, 24(2): 93-99.
HUANG Changhai, XIAO Yingjie, GAO Deyi, et al. Evaluation model of navigation risk in fixed water system and its application[J]. China Journal of Safety Science, 2014, 24(2): 93-99.
[19] 徐武雄, 初秀民, 刘兴龙. 船舶交通流建模与仿真研究进展[J]. 水利水运工程学报, 2014 (6): 91-99.
XU Wuxiong, CHU Xiumin, LIU Xinglong. Research progress in modeling and simulation of ship traffic flow[J]. Hydro-Science and Engineering, 2014 (6): 91-99.