价值衰减下水上交通事故致因关联规则挖掘

doi:10.16265/j.cnki.issn1003-3033.2022.10.2087

摘要/Abstract

摘要：

为提升我国沿海水域交通安全水平，在考虑水上交通事故价值衰减的基础上，提出一种基于时间权重的Apriori算法，并以2015—2020年我国沿海水域水上交通事故数据为样本，分析我国沿海水域水上交通事故特征，挖掘事故属性与致因属性间潜在映射关系，形成13条强关联规则。结果表明：交通事故的挖掘价值具有随时间衰减的特性，我国沿海水域在固定季节、固定时段、特定船舶及特殊水域存在水上交通事故高发的特征，事故基础属性与致因属性间存在一定的映射关系，部分致因属性常在特定的事故基础属性条件下同时出现。

关键词: 价值衰减, 水上交通事故, 事故致因, 关联规则, 数据挖掘, 时间权重

Abstract:

In order to improve the traffic safety level of China's coastal waters， an Apriori algorithm based on time weight was proposed on the basis of considering the value attenuation of water traffic accidents. Taking the data of water traffic accidents in China's coastal waters from 2015 to 2020 as samples， the characteristics of water traffic accidents in China's coastal waters were analyzed，and the potential mapping relationship between accident attributes and causal attributes were explored， to form 13 strong association rules. The results show that the mining value of traffic accident has the characteristics of decaying with time. The coastal waters of China have the characteristics of high incidence of water traffic accidents in fixed seasons， fixed periods， specific vessels and special waters. There is a ctetain mapping relationship between the basic attributes of accidents and the causal attributes. Some causal attributes often appear at the same time under the specific basic attributes of accidents.

Key words: value attenuation, water traffic accidents, accidents causation, association rules, data mining, time weight

杨利超, 刘敬贤, 刘钊, 刘轶伟. 价值衰减下水上交通事故致因关联规则挖掘[J]. 中国安全科学学报, 2022, 32(10): 127-134.

YANG Lichao, LIU Jingxian, LIU Zhao, LIU Yiwei. Association rules mining of water traffic accidents causation under value attenuation[J]. China Safety Science Journal, 2022, 32(10): 127-134.

图/表 12

图1

表1

表2

图2

图3

表3

图4

图5

图6

表4

表5

表6

参考文献 14

[1]	交通运输部. 2020年交通运输行业发展统计公报[EB/OL].[2021-05-19]. http://www.gov.cn/xinwen/2021-05/19/content_5608523.htm.
[2]	张欣欣, 胡甚平, 陈渊, 等. 船舶港口引航风险致因人-组织因素影响分析[J]. 中国安全科学学报, 2019, 29(12): 78-84. doi: 10.16265/j.cnki.issn1003-3033.2019.12.013
	ZHANG Xinxin, HU Shenping, CHEN Yuan, et al. Impact analysis of human-organizational factors on pilotage risks at harbor[J]. China Safety Science Journal, 2019, 29(12): 78-84. doi: 10.16265/j.cnki.issn1003-3033.2019.12.013
[3]	付姗姗, 刘燕平, 席永涛, 等. 北极海冰环境下船舶事故严重程度影响因素分析[J]. 中国安全科学学报, 2019, 29(10): 84-90. doi: 10.16265/j.cnki.issn1003-3033.2019.10.013
	FU Shanshan, LIU Yanping, XI Yongtao, et al. Analysis of influencing factors on severity of ship accidents in ice-covered Arctic waters[J]. China Safety Science Journal, 2019, 29(10): 84-90. doi: 10.16265/j.cnki.issn1003-3033.2019.10.013
[4]	XI Jianfeng, ZHAO Zhonghao, LI Wei, et al. A traffic accident causation analysis method based on AHP-apriori[J]. Procedia Engineering, 2016, 137: 680-687. doi: 10.1016/j.proeng.2016.01.305
[5]	SAKAR C, TOZ A C, BUBER M, et al. Risk analysis of grounding accidents by mapping a fault tree into a bayesian network[J]. Applied Ocean Research, 2021, 113(1):1-12.
[6]	HASUGIAN S, RAHMAWATI M, WAHYUNI A I S, et al. Analysis the risk of the ship accident in indonesia with bayesian network model approach[J]. Annals of the Romanian Society for Cell Biology, 2021, 25(2): 3341-3356.
[7]	HASANSPAHIĆ N, VUJIČIĆ S, FRANČIĆ V, et al. The role of the human factor in marine accidents[J]. Journal of Marine Science and Engineering, 2021, 9(3): 1-16. doi: 10.3390/jmse9010001
[8]	ZHANG Xiaohui, LIU Zhengjiang. Analysis of multi-dimensional association rule in marine casualties[C]. ICTIS 2011:Multimodal Approach to Sustained Transportation System Development, 2011:2697-2705.
[9]	胡瑾秋, 张曦月, 吴志强. 结合TF-IDF的企业生产隐患关联预警及可视化研究[J]. 中国安全科学学报, 2019, 29(7): 170-176. doi: 10.16265/j.cnki.issn1003-3033.2019.07.027
	HU Jinqiu, ZHANG Xiyue, WU Zhiqiang, et al. Research on associated early-warning and visualization of hidden danger in enterprise production based on TF-IDF[J]. China Safety Science Journal, 2019, 29(7): 170-176. doi: 10.16265/j.cnki.issn1003-3033.2019.07.027
[10]	付姗姗, 刘燕平, 席永涛, 等. 北极水域船舶事故特征及关联规则分析[J]. 极地研究, 2020, 32(1):102-111.
	FU Shanshan, LIU Yanping, XI Yongtao, et al. Feature analysis and association rule mining of ship accidents in arctic waters[J]. Chinese Journal of Polar Research, 2020, 32(1):102-111. doi: 10.13679/j.jdyj.20190010
[11]	WENG Jinxian, LI Guorong. Exploring shipping accident contributory factors using association rules[J]. Journal of Transportation Safety & Security, 2019, 11(1): 36-57.
[12]	ÇAKıR E, FıŞKıN R, SEVGILI C. Investigation of tugboat accidents severity: an application of association rule mining algorithms[J]. Reliability Engineering & System Safety, 2021, 209:DOI:10.1016/j.ress.20201.107470. doi: 10.1016/j.ress.20201.107470
[13]	殷春武. 基于时间权重的区间型组合预测权重确定方法[J]. 统计与决策, 2013(21):67-70.
[14]	王凯, 王腾飞, 吴兵, 等. 船舶导航与海洋运输安全的动态发展与展望: TansNav2017国际会议综述[J]. 交通信息与安全, 2017, 35(4): 1-11.
	WANG Kai, WANG Tengfei, WU Bing, et al. Development trend and prospect of ship navigation and marine transportation safety: summary of TansNav 2017 international conference[J]. Journal of Transport Information and Safety, 2017, 35(4):1-11.

基本属性			时间属性		空间属性	船舶属性
事故类型A	事故等级B	人员伤亡C	事发季度D	事发时间E	事发水域F	船舶类型G	船舶吨位H
碰撞A₁ 搁浅A₂ 触礁A₃ 浪损A₄ 火灾/爆炸A₅ 风灾A₆ 自沉A₇ 操作性污染A₈ 其他事故A₉	一般B₁ 较大B₂ 重大B₃ 特别重大B₄	存在C₁ 不存在C₂	第1季度D₁ 第2季度D₂ 第3季度D₃ 第4季度D₄	0000~0400E₁ 0400~0800E₂ 0800~1 200E₃ 1 200~1 600E₄ 1 600~2 000E₅ 2 000~2 400E₆	港口F₁ 锚地F₂ 航道F₃ 其他F₄	散货船G₁、油化船G₂、杂货船G₃、渔船G₄、砂石船G₅、集装箱船G₆、客滚船G₇、干货船G₈、工程船G₉、其他船舶G₁₀	3000 GT以下H₁ 3 000~10 000 GTH₂ 10 000~30 000 GTH₃ 30 000 GT以上H₄

人为因素属性I	船舶因素属性J	环境因素属性K	管理因素属性L
瞭望疏忽I₁	超载J₁	水文环境异常K₁	船员不适任L₁
身体状态不良I₂	结构缺陷J₂	交通秩序混乱K₂	非法运输L₂
违规作业I₃	货物缺陷J₃	航道弯曲度异常K₃	维护保养不当L₃
未使用安全航速I₄	货物装载不当J₄	大风天气K₄	配员不足L₄
操纵行为不当I₅	设备缺陷J₅	助航设施异常K₅	工作态度不端L₅
值班/值守不当I₆	船舶稳性不足J₆	渔区密集K₆	船员培训不足L₆
避让行为不当I₇	货物具有危险性J₇	能见度不良K₇	监管不足L₇
仪器操作不当I₈	其他船舶因素J₈	可航水域不足K₈	其他管理因素L₈
号灯号型显示不当I₉	—	图书资料不完备K₉	—
其他人为因素I₁₀	—	其他环境因素K₁₀	—

年份	季度	权重	年份	季度	权重
2015	1	0.500 0	2018	1	0.909 7
	2	0.625 7		2	0.915 6
	3	0.704 8		3	0.920 8
	4	0.757 8		4	0.925 4
2016	1	0.795 2	2019	1	0.929 6
	2	0.822 8		2	0.933 2
	3	0.844 0		3	0.936 5
	4	0.860 8		4	0.939 6
2017	1	0.874 3	2020	1	0.942 3
	2	0.885 5		2	0.944 8
	3	0.894 9		3	0.947 1
	4	0.902 8		4	0.949 2

序号	后项	前项	支持度	置信度	提升度
1	事故类型= “碰撞”	事故致因=“操纵行为不当”&事故致因= “值班值守不当”	0.566 75	0.833 34	1.1280 2
2	事故类型= “碰撞”	事故致因=“瞭望疏忽”&事故致因= “操纵行为不当”&船舶类型=“渔船”	0.466 78	0.833 34	1.035 71
3	事故类型= “碰撞”	事故致因=“避让行为不当”&事故致因= “操纵行为不当”	0.433 33	0.766 67	1.002 54
4	事故类型= “碰撞”	事故致因=“避让行为不当”&事故致因=“瞭望疏忽”& 事故致因=“操纵行为不当”&船舶类型=“渔船”	0.433 33	0.766 67	1.000 00
5	事故类型= “碰撞”	事故致因=“避让行为不当”&事故致因=“瞭望疏忽”& 事故致因=“操纵行为不当”&事故水域=“航道水域”	0.410 67	0.712 25	1.000 00

序号	后项	前项	支持度	置信度	提升度
1	事故类型= “碰撞”	事故致因=“操纵行为不当”&事故致因= “值班值守不当”	0.477 29	0.755 16	1.150 25
2	事故类型= “碰撞”	事故致因=“瞭望疏忽”&事故致因= “操纵行为不当”&船舶类型=“渔船”	0.434 94	0.755 16	1.048 17
3	事故类型= “碰撞”	事故致因=“避让行为不当”&事故致因= “操纵行为不当”	0.414 95	0.744 93	1.013 73
4	事故类型= “碰撞”	事故致因=“避让行为不当”&事故致因=“瞭望疏忽” &事故致因=“操纵行为不当”&船舶类型=“渔船”	0.414 95	0.728 02	1.037 28
5	事故类型= “碰撞”	事故致因=“避让行为不当”&事故致因=“瞭望疏忽” &事故致因=“操纵行为不当”&事故水域=“航道水域”	0.400 04	0.728 02	1.000 00
6	事故类型= “碰撞”	事故致因=“避让行为不当”&事故致因=“瞭望疏忽” &船舶类型=“渔船”&事故水域=“航道水域”	0.400 01	0.735 40	1.236 31
序号	后项	前项	支持度	置信度	提升度
7	事故类型= “自沉”	事故致因=“超载”&事故致因= “大风天气”&船舶吨位=“小于3 000 GT”	0.434 94	0.791 54	1.000 00
8	事故类型= “触碰”	事故致因=“瞭望疏忽”& “未使用安全航速”	0.421 45	0.718 40	1.016 27