Prediction model for road transport accidents of hazardous chemicals based on Bayesian network

doi:10.16265/j.cnki.issn1003-3033.2022.03.024

Abstract

Abstract:

In order to accurately predict road transport accidents of hazardous chemicals, firstly, data of 1,727 such transport accidents in China from 2015 to 2020 were collected, and a Bayesian network (BN) was developed with accident influencing factors, accident types, accident emergency treatment time and the degree of casualties as main nodes. Then, a prediction model for the accidents was established in Netica, and its validity was verified according to the mean absolute error (MAE). Finally, through forward causal reasoning and reverse diagnostic reasoning, the posterior probability changes of each variable of target nodes were observed, and accident development trend and evolution process under set conditions were explored. The results show that the model can effectively predict accidents under set conditions. Through positive causal inference, it is concluded that the most likely form of accident at noon is the leakage accident caused by rear-end collision or tank leakage, while based on reverse diagnostic reasoning, it is found that carrying capacity <30 t is a significant condition for flammable liquid leakage accidents to be successfully disposed of within 0 to 3 hours.

Key words: hazardous chemicals, road transportation, Bayesian network (BN), accident prediction, Netica

LU Yi, WU Jiangle, SHAO Shuzhen, SHI Shiliang, ZHOU Rongyi, WANG Wei. Prediction model for road transport accidents of hazardous chemicals based on Bayesian network[J]. China Safety Science Journal, 2022, 32(3): 174-182.

Figures/Tables 9

Tab.1

Fig.1

Fig.2

Tab.2

Tab.3

Tab.4

Fig.3

Fig.4

Fig.5

References 18

[1]	荣乌高速烟台莱州段“1·16”重大道路交通事故责任调查报告OL.(2015-06-11). http://www.jiaozhou.gov.cn/n1822/n2126/n2128/n2139/161005021154632842.html.
[2]	河北保定张石高速浮图峪五号隧道“5·23”重大危险化学品运输车辆燃爆事故调查报告[OL].(2018-10-29). https://yjgl.hebei.gov.cn/portal/.
[3]	河北保定张石高速浮图峪五号隧道“5·23”重大危险化学品运输车辆燃爆事故调查报告[OL].(2020-06-13) https://baike.baidu.com/.
[4]	伍江乐, 鲁义, 施式亮, 等. 危化品道路运输风险管控现状分析[J]. 能源与环保, 2021, 43(2):5-9.
	WU Jiangle, LU Yi, SHI Shiliang, et al. Analysis on status quo of risk management and control of hazardous chemicals road transportation[J]. China Energy and Environmental Protection, 2021, 43 (2): 5-9.
[5]	张江石, 冯娜娜. 基于动态贝叶斯网络情景推演的危化品事故应急处置研究[J]. 安全与环境学报, 2020, 20(4):1420-1426.
	ZHANG Jiangshi, FENG Nana. Study on emergency response of hazardous chemicals accidents based on dynamic Bayesian network scenario deduction[J]. Journal of Safety and Environment, 2020, 20 (4): 1420-1426.
[6]	NOGUCHI H, HIENUKI S, FUSE M. Network theory-based accident scenario analysis for hazardous material transport: a case study of liquefied petroleum gas transport in Japan[J]. Reliability Engineering & System Safety, 2020, 203: DOI:10.1016/j.ress.2020.107107
[7]	郭健, 胡金瑞. 多学科交叉的危化品运输安全保障技术及动态演示[J]. 安全与环境工程, 2021, 28(2):80-85.
	GUO Jian, HU Jinrui. Multi-disciplinary safety guarantee technology and dynamic demonstration of transportation of hazardous chemicals[J]. Journal of Safety and Environment, 2021, 28(2):80-85.
[8]	施式亮, 陈晓勇, 刘勇, 等. 基于AHP耦合度的危化品道路运输风险因素耦合特征[J]. 湖南科技大学学报:自然科学版, 2021, 36(1):23-29.
	SHI Shiliang, CHEN Xiaoyong, LIU Yong, et al. Research on the coupling characteristics of risk factors of dangerous chemicals in road transportation based on AHP coupling degree[J]. Journal of Hunan University of Science & Technology:Natural Science Edition, 2021, 36(1):23-29.
[9]	王军武, 王梦雨, 刘登辉, 等. 基于HFACS-BN的地铁车站施工高处坠落可能性评价[J]. 中国安全科学学报, 2021, 31(6):90-98. doi: 10.16265/j.cnki.issn 1003-3033.2021.06.012
	WANG Junwu, WANG Mengyu, LIU Denghui, et al. Evaluation of falling possibility at height in subway station construction based on HFACS-BN[J]. China Safety Science Journal, 2021, 31(6):90-98. doi: 10.16265/j.cnki.issn 1003-3033.2021.06.012
[10]	李健行, 夏登友, 武旭鹏. 基于知识元与动态贝叶斯网络的非常规突发灾害事故情景分析[J]. 安全与环境学报, 2014, 14(4):165-170.
	LI Jianxing, XIA Dengyou, WU Xupeng. Scenario analysis of unconventional sudden disasters based on knowledge element and dynamic Bayesian network[J]. Journal of Safety and Environment, 2014, 14 (4): 165-170.
[11]	王喆, 孔维磊, 方丹辉, 等. 基于贝叶斯网络的城镇洪涝应急情景推演研究[J]. 中国安全科学学报, 2021, 31(6):182-188. doi: 10.16265/j.cnki.issn 1003-3033.2021.06.024
	WANG Zhe, KONG Weilei, FANG Danhui, et al. Research on urban flood and waterlog emergency scenario deduction based on Bayesian network[J]. China Safety Science Journal, 2021, 31(6):182-188. doi: 10.16265/j.cnki.issn 1003-3033.2021.06.024
[12]	屈静, 张建彬, 李旭芳, 等. 基于贝叶斯网络的输油管道泄漏事故情景推演[J]. 中国安全科学学报, 2021, 31(1):192-198. doi: 10.16265/j.cnki.issn1003-3033.2021.01.028
	QU Jing, ZHANG Jianbin, LI Xufang, et al. Deduction of leakage accident scenarios of oil pipelines based on Bayesian network[J]. China Safety Science Journal, 2021, 31(1):192-198. doi: 10.16265/j.cnki.issn1003-3033.2021.01.028
[13]	吕学志, 胡晓峰, 吴琳, 等. 基于贝叶斯网络的任务共同体识别[J]. 计算机工程与应用, 2019, 55(5):251-257. doi: 10.3778/j.issn.1002-8331.1711-0302
	LYU Xuezhi, HU Xiaofeng, WU Lin, et al. Identification method of mission community based on Bayesian network[J]. Computer Engineering and Applications, 2019, 55 (5): 251-257.
[14]	汪子恒. 基于贝叶斯网络的道路危货运输耦合风险评估研究[D]. 南京: 南京林业大学,2020.
	WANG Ziheng. Research on coupling risk assessment of dangerous cargo transportation based on Bayesian network[D]. Nanjing: Nanjing Forestry University,2020.
[15]	化学品事故信息网. 2015年至2020年国内危险化学品道路运输事故报道[EB/OL].[2021-01-21]. http://accident.nrcc.com.cn:9090/Portalsite/Index.aspx.
[16]	曹建, 施式亮, 鲁义, 等. 2013—2018年罐车公路运输危化品事故分析[J]. 中国安全科学学报, 2020, 30(2):119-126. doi: 10.16265/j.cnki.issn1003-3033.2020.02.019
	CAO Jian, SHI Shiliang, LU Yi, et al. Analysis on tank transportation accidents of hazardous chemicals from 2013 to 2018[J]. China Safety Science Journal, 2020, 30 (2): 119-126. doi: 10.16265/j.cnki.issn1003-3033.2020.02.019
[17]	莫定源. 基于贝叶斯网络的生态环境脆弱性评估模型与应用[D]. 烟台: 中国科学院烟台海岸带研究所,2017.
	MO Dingyuan. A Bayesian network model and its application on ecological environment vulnerability assessment[D]. Yantai: Yantai Institute of Coast Zone Research Chinese Academy of Sciences,2017.
[18]	陈文瑛, 王影, 李启华. 危险化学品道路运输风险预测模型研究[J]. 安全与环境学报, 2020, 20(5):1683-1689.
	CHEN Wenying, WANG Ying, LI Qihua. Study on risk prediction model of road transportation of hazardous chemicals[J]. Journal of Safety and Environment, 2020, 20 (5): 1683-1689.

节点变量	分类	频数	占比/%	节点变量	分类	频数	占比/%
时段	凌晨(0-5时)	198	20.39	公路等级	高速公路	395	40.68
	早晨(5-8时)	175	18.02		城市道路	191	19.67
	上午(8-11时)	148	15.24		一级公路	146	15.04
	中午(11-13时)	91	9.37		二级公路	84	8.65
	下午(13-16时)	137	14.11		三级公路	70	7.21
	傍晚(16-19时)	90	9.27		四级公路	39	4.01
	晚上(19-24时)	132	13.60		其他	46	4.74
季节	春(3、4、5月)	257	26.47	危化品事故类型	泄漏事故	774	79.71
	夏(6、7、8月)	282	29.04		火灾事故	141	14.52
	秋(9、10、11月)	224	23.07		爆炸事故	33	3.40
	冬(12、1、2月)	208	21.42		无危化品事故	23	2.37
危化品类别	易燃固体	24	2.47	初发事故类型	追尾	235	24.20
	易燃液体	556	57.26		相撞	157	16.17
	易燃气体	37	3.81		单方翻车	267	27.50
	腐蚀品	141	14.52		避让翻车	66	6.80
	爆炸品	130	13.39		罐体泄漏	132	13.59
	有毒品	63	6.49		车辆故障	114	11.74
	非易燃无毒气体	11	1.13	运载量	<30 t	552	56.85
	其他	9	0.93	运载量	≥30 t	419	43.15
事故应急处理时间	0~3 h(包括3 h)	302	31.10	事故伤亡程度	轻微事故	731	75.28
	3~6 h(包括6 h)	295	30.38		一般事故	158	16.27
	6~9 h(包括9 h)	167	17.20		重大事故	76	7.83
	9~12 h(包括12 h)	116	11.95		特别重大事故	6	0.62
	12~15 h(包括15 h)	31	3.19
	15 h以上	60	6.18

节点	互信息值	占比/%	方差
初发事故类型	0.025 65	1.36	0.003 688 5
危化品类别	0.021 32	1.13	0.003 170 3
时段	0.002 23	0.118	0.000 289 6
季节	0.000 58	0.030 6	0.000 093 9

类型	2015—2017年				2018—2020年				Netica
类型	泄漏	火灾	爆炸	无	泄漏	火灾	爆炸	无	泄漏	火灾	爆炸	无
单方翻车	30.62	13.48	15.15	26.09	28.15	13.46	12.50	32.00	25.80	16.50	17.90	17.50
相撞	15.50	15.60	30.30	21.74	20.13	13.46	18.75	16.00	15.20	15.90	17.10	16.10
追尾	25.71	17.73	15.15	26.09	26.52	20.19	25.00	28.00	24.60	20.60	19.80	20.50
罐体泄漏	16.28	2.83	6.06	0.00	17.35	4.81	12.50	0.00	15.60	13.80	15.80	16.00
车辆故障	4.52	45.39	15.15	21.74	4.58	46.15	12.50	20.00	9.60	21.70	16.50	16.70
避让翻车	7.37	4.96	3.03	4.34	3.27	1.92	0.00	4.00	9.17	11.50	12.90	13.10

危化品事故类型	泄漏	火灾	爆炸	无危化品
模型预测概率与2015—2017年实际概率对比(MAE)	0.022 9	0.078 9	0.069 2	0.082 7
模型预测概率与2018—2020年实际概率对比(MAE)	0.036 5	0.081 5	0.054 1	0.084 2