利用动态贝叶斯网络实现人群聚集风险分析

doi:10.16265/j.cnki.issn1003-3033.2017.07.028

摘要/Abstract

摘要： 为动态探究影响人群聚集风险的主要因素及定量评估人群聚集风险,依据贝叶斯估计理论改进静态贝叶斯网络模型,建立动态贝叶斯网络模型。用该模型可根据实时采集数据计算后验参数,获取动态定量风险评估结果。利用所建的动态贝叶斯网络模型动态定量评估北京市某大型商业街区人群聚集风险。结果表明:该街区初始人群聚集拥堵概率为0.8×10^-3,踩踏概率为7.6×10^-6。随着实时观测数据的引入,最终拥堵概率为2.4×10^-3,踩踏概率为1.63×10^-5,其中疏散不及时、疏散通道不畅、疏散标志不清等3个因素的相对重要度影响因子较大,是主要影响因素。实例中各基本事件的发生概率和相对影响因子动态变化,证明该模型有效。

关键词: 公共安全, 风险分析, 动态贝叶斯网络, 参数更新, 人群聚集

Abstract: In order to explore trigger factors of crowd stampede and evaluate the accident risk quantitatively, a static Bayesian network model was improved based on the Bayesian estimation theory. A dynamic Bayesian model, by which posterior parameters can be calculated according to the collected real-time data, was built for obtaining quantitative assessment results of dynamic risk. Finally, a certain commercial district with a large population in Beijing City was used as an example to verify the effectiveness of the proposed method. The results show that the risk of crowd jamming and that of stampede are 0.8×10^-3 and 7.6×10^-6 respectively, that with the introduction of real-time data, the risks are improved to 2.4×10^-3 and 1.63×10^-5 respectively, that slow evacuation, occupied evacuation routes, and the lack of safety logo are main causes of jam and stampede, that the probability of occurrence and weight of influence of each basic event in the instance are dynamic, which proves that the model is effective.

Key words: public safety, risk assessment, dynamic Bayesian network, probability updating, stampede

中图分类号:

X913.4

赵鲁炎, 马峻. 利用动态贝叶斯网络实现人群聚集风险分析[J]. 中国安全科学学报, 2017, 27(7): 157-162.

ZHAO Luyan, MA Jun. Risk analysis of stampede by dynamic Bayesian network[J]. China Safety Science Journal, 2017, 27(7): 157-162.

参考文献

[1] 刘泽照,杨帆,黄杰. 大型公共活动踩踏事故模糊风险评估应用研究:以上海"12·31"踩踏事件为例[J].南京航空航天大学学报:社会科学版, 2015, 17(2):33-38.
LIU Zezhao, YANG Fan, HUANG Jie. On fussy risk assessment for stampede accident in large public events-analysis of Shanghai "12·31" stampede incidence [J]. Journal of Nanjing University of Aeronautics and Astronautics:Social Sciencss, 2015, 17(2):33-38.
[2] 吴祺. 人流密集区防踩踏具体措施探究[J]. 黑龙江省政法管理干部学院学报,2016(5):145-147.
WU Qi. A probe into the concrete measures of anti - trampling in flowing - intensive areas [J]. Journal of Heilongjiang Administrative Cadre College of Politics and Law, 2016(5):145-147.
[3] 佟瑞鹏,李春旭,郑毛景,等. 拥挤踩踏事故风险定量评价模型及其优化分析[J]. 中国安全科学学报,2013,23(12):90-94.
TONG Ruipeng, LI Chunxu, ZHENG Maojing, et al. A quantitative risk assessment model of crowded stampede accident and its optimization analysis[J]. China Safety Science Journal, 2013, 23(12):90-94.
[4] 胡成,郭婧婷,李强,等. 开放性公共场所人群聚集风险和关键点评估方法[J].中国安全科学学报,2015,25(12):164-169.
HU Cheng, GUO Jingting, LI Qiang, et al. Method for evaluating crowd risk and identifying critical nodes in open public places[J]. China Safety Science Journal, 2015, 25(12):164-169.
[5] 张仕学,李石君,余伟. 突发事件人群异常聚集热点区域预测[J].中国安全科学学报,2015, 25(9):160-164.
ZHANG Shixue,LI Shijun,YU Wei,et al. Prediction on hot region of crowd abnormal gathering on unexpected event[J]. China Safety Science Journal,2015, 25(9):160-164.
[6] 霍宇芒,宋守信,顾一波. 基于组合权重的地铁车站拥挤踩踏事故风险评价[J] 安全与环境工程,2016,23(5):139-143.
HUO Yumang,SONG Shouxin, GU Yibo.Risk assessment of crowd crushing and trampling accidents in subway stations based on combined weights[J]. Safety and Environment Engineering, 2016, 23(5):139-143.
[7] 茆诗松.贝叶斯统计[M].北京:中国统计出版社, 1999: 6-39.
[8] 谭清磊, 陈国明,付建民.基于因果图和贝叶斯网络的高含硫井口分离器风险分析[J]. 安全与环境学报,2012,12(5):234-238.
TAN Qinglei, CHEN Guoming, FU Jianmin.Risk assessment of wellhead separator at the high-sulfur gas gathering station based on cause-effect diagram and Bayesian network [J]. Journal of Safety and Environment, 2012, 12(5):234-238.
[9] NIMA Khakzad, FAISAL Khan,PAUL Amyotte, Dynamic safety analysis of process systems by mapping bow-tie into Bayesian network[J]. Process Safety and Environmental Protection, 2013, 91:46-53.
[10] 袁娜. 基于贝叶斯网络和业务持续管理的液氨泄漏事故情景构建研究[D]. 北京:北京交通大学,2016.
YUAN Na. Research on scenario construction of liquid ammonia leakage accident based on Bayesian network and business continuity management[D]. Beijing: Beijing Jiaotong University,2016.
[11] 王春雪,吕淑然. 城市燃气管道泄漏致灾混合概率风险评估研究[J].中国安全科学报,2016,26(12):146-151.
WANG Chunxue, LYU Shuran. Mixture probabilistic model based assessment of risk of disaster induced by urban gas pipeline leakage[J]. China Safety Science Journal, 2016, 26(12):146-151.
[12] 冯耿,张楠,陈猛志,等. 事故树分析与贝叶斯网络重要度在溃坝风险分析中的应用[J].水电能源科学, 2013, 31(4):66-68.
FENG Geng, ZHANG Nan, CHEN Mengzhi, et al. Application of FTA and BN importance factor in dam break risk assessment [J].Water Resources and Power, 2013, 31(4):66-68.
[13] VILLA Valeria, PALTRINIERI Nicola, KHAN Faisal,et al. Towards dynamic risk analysis: a review of the risk assessment approach and its limitations in the chemical process industry [J]. Safety Science,2016,89:77-93.

[1]	刘继川, 桂蕾. 城市公共安全风险评估与控制对策研究:以武汉市为例^*[J]. 中国安全科学学报, 2022, 32(1): 164-171.
[2]	胡成, 李俊伟, 李强, 陈晋. 考虑人群聚集时空分布的安全容量计算方法:以北京什刹海景区为例[J]. 中国安全科学学报, 2021, 31(6): 136-143.
[3]	华俊杰, 申满对, 宋贤生. 化工装置泄漏频率数据源比较研究[J]. 中国安全科学学报, 2021, 31(3): 35-40.
[4]	高涵, 多依丽, 孙铁, 王志荣, 郭品坤. 基于DBT-DBN模型的气化炉超温动态风险分析[J]. 中国安全科学学报, 2021, 31(3): 73-81.
[5]	石娟, 郑鹏, 常丁懿. 大数据环境下的城市公共安全治理:区块链技术赋能[J]. 中国安全科学学报, 2021, 31(2): 24-32.
[6]	龙丹冰, 魏君豪, 杨成. 知识图谱改进的施工行为安全风险与危险位置识别算法^*[J]. 中国安全科学学报, 2021, 31(12): 10-16.
[7]	李浩然, 王子恒, 杨起帆, 张亚君, 欧阳作林. 复杂网络下地铁灾害链演化模型与风险分析^*[J]. 中国安全科学学报, 2021, 31(11): 141-147.
[8]	王起全, 刘志刚, 杨鑫刚, 刘建, 王睿智. 化工园区重大危险源风险情景构建分析[J]. 中国安全科学学报, 2020, 30(8): 63-69.
[9]	谢小良, 褚琦, 张淑君, 卫国, 成佳祺. 基于DBN的疫苗运输质量安全风险监测方法[J]. 中国安全科学学报, 2020, 30(7): 19-26.
[10]	陈志煌, 刘国恒, 王莹莹, 朱春丽, 单荐, 翟小东. 基于动态贝叶斯网络的水下连接器故障诊断[J]. 中国安全科学学报, 2020, 30(5): 81-87.
[11]	李华, 李琳倩, 益朋. 景区密集人群踩踏事故情景分析[J]. 中国安全科学学报, 2020, 30(4): 108-113.
[12]	罗军华, 林孝松, 牟凤云, 李宏伟, 张莉. 山区暴雨-农业灾害链复杂网络静态风险分析[J]. 中国安全科学学报, 2020, 30(3): 163-170.
[13]	王金江, 王舒辉, 张兴, 张来斌. 基于RBI技术的储气库分离器风险分析[J]. 中国安全科学学报, 2020, 30(2): 21-27.
[14]	吕伟, 蒋翠玲, 姜雅娟, 赵彩虹, 汪京辉, 李加伟. 考虑多因素影响的毒气扩散风险潜势图构建[J]. 中国安全科学学报, 2020, 30(11): 175-181.
[15]	宋英华, 张哲, 方丹辉. 城市洪涝下承灾体暴露性及行人失稳风险分析[J]. 中国安全科学学报, 2020, 30(10): 105-111.