Estimating fatality rate of house collapse based on human activity pattern

doi:10.16265/j.cnki.issn1003-3033.2023.11.0500

Abstract

Abstract:

Efficient estimation of casualties in earthquake scenarios involves the identification of mortality rate variations among individuals in different spatial states, which may be achieved through the simulation and replication of population spatial distributions during earthquakes. Based on the human activity pattern model set which was generated from the China Human Activity Pattern of Rural Residents Survey (HAPRRES), we simulated individuals' activities during 13 earthquakes in China from 2005 to 2018 to estimate the spatial state snapshot distributions of the earthquake affected areas at their onset moment. These estimates allowed the optimization of mortality rate models by distinguishing between sleeping and awake people in earthquake-damaged houses. The optimized mortality models reveal that the mortality rate for individuals asleep (5.16‰) is six times higher than that for individuals awake (0.85‰) in seismically collapsed houses. Notably, when the adobe houses are destroyed, the mortality rate of sleeping people (1.60%) is 11.9 times higher than that for the awake people (1.34‰). Conversely, during the destruction of non-adobe houses, the mortality rate for the sleeping people (2.35‰) is 4.3 times higher than that of the awake people (0.54‰). An uncertainty analysis proves the robustness of the mortality rate model. As this explorative research demonstrates, incorporating human activity pattern model will help to improve the rapid estimation of casualties of sudden disasters.

Key words: human activity pattern, earthquake, building damage, mortality rate, adobe house

JIANG Wen, MENG Yaobin, ZHANG Dongni, ZHANG Yuan, WU Jiaxin. Estimating fatality rate of house collapse based on human activity pattern[J]. China Safety Science Journal, 2023, 33(11): 165-173.

Figures/Tables 6

Fig.1

Tab.1

Seismic damage dataset for modeling and "6.17" Changning earthquake for model validation

研究区	时间	房屋毁损致死人数	房屋震害面积/m²			4状态分布比例				权重值
研究区	时间	房屋毁损致死人数	毁坏	严重破坏	合计	室外	室内	睡眠	工作	权重值
新疆维吾尔自治区塔什库尔干县	2017-05-11 T5:58	3	361 568	151 041	967 889	0.02	0.18	0.79	0.01	0.63
新疆维吾尔自治区皮山县^▲	2015-07-03 T9:07	3	1 368 665	474 726	5 779 090	0.04	0.38	0.02	0.56	0.65
甘肃省岷县VIII^*▲1	2013-07-22 T7:45	70	38	41	100	0.05	0.48	0.06	0.41	0.87
四川省芦山县^**	2013-04-20 T8:02	39	30	95	174	0.04	0.45	0.11	0.40	0.77
江苏省高邮市	2012-07-20 T20:11	1	438	2 679	388 303	0.05	0.61	0.02	0.32	0.86
云南省宁蒗县	2012-06-24 T15:59	3	252 990	1 407 374	1 660 364	0.12	0.47	0.00	0.41	0.86
四川省盐源县	2012-06-24 T15:59	1	252 990	1 407 374	1 660 364	0.09	0.54	0.00	0.36	0.90
云南省盈江县^▲	2011-03-10 T12:58	18	21 800	59 150	87 490	0.05	0.55	0.00	0.39	0.80
重庆市荣昌县^▲	2009-08-08 T21:26	2	148 256	593 025	3 991 079	0.08	0.65	0.20	0.07	0.90
云南省姚安县^▲	2009-07-09 T19:19	1	472 075	1 831 272	4 259 828	0.06	0.51	0.00	0.43	0.78
西藏自治区当雄县^▲	2008-10-06 T16:30	9	45 625	133 457	324 671	0.07	0.43	0.00	0.50	0.66
四川省盐边县^▲	2008-08-30 T16:30	1	172 955	909 462	2 669 345	0.07	0.55	0.00	0.38	0.72
四川省米易县^▲	2008-08-30 T16:30	1	39 456	764 577	7 306 300	0.06	0.57	0.00	0.37	0.88
四川省会理县^▲1	2008-08-30 T16:30	22	1 113 369	2 498 851	11 199 047	0.06	0.57	0.00	0.37	0.90
四川省仁和区	2008-08-30 T16:30	1	1 232 704	3 155 838	12 786 630	0.06	0.57	0.00	0.37	0.80
云南省盐津县^▲	2006-07-22 T9:10	2	98 750	574 465	2 340 845	0.07	0.46	0.04	0.42	0.94
云南省大关县^▲	2006-07-22 T9:10	2	169 393	947 677	3 933 136	0.06	0.47	0.04	0.42	0.92
江西省九江县^▲	2005-11-26 T8:49	5	338 767	1 837 879	6 060 438	0.06	0.43	0.07	0.44	0.83
江西省瑞昌市^▲	2005-11-26 T8:49	5	680 246	4 218 809	15 762 185	0.05	0.40	0.09	0.46	0.90
广西壮族自治区平果县	2005-10-27 T19:18	1	805	10 818	43 827	0.06	0.48	0.00	0.46	0.93
四川省宜宾市2	2019-06-17 T22:55	12	3 545	81 284	225 698	0.05	0.12	0.77	0.06	-

Tab.1

Tab.2

Mortality modeling results

模型编号	模型参数				优选模型	AIC	R²
模型编号	X₁	X₂	X₃	X₄	优选模型	AIC	R²
1	毁坏房屋内睡眠人数	毁坏房屋内清醒人数	—	—	$y = 5.16 × 10 - 3 · X 1 + 8.54 × 10 - 4 · X 2 (± 9.92 × 10 - 4) (± 1.01 × 10 - 4)$	235.99	0.47
2	毁坏房屋内人数	—	—	—	$y = 1.20 × 10 - 3 · X 1 (± 8.47 × 10 - 5)$	258.26	0.39
3	毁坏的土坯房内睡眠人数	毁坏的土坯房内清醒人数	毁坏的非土坯房内睡眠人数	毁坏的非土坯房内清醒人数	$y = 1.60 × 10 - 2 · X 1 + 1.34 × 10 - 3 · X 2 + 2.35 × 10 - 3 · X 3 + 5.44 × 10 - 4 · X 4 (± 4.06 × 10 - 3) (± 2.95 × 10 - 4) (± 1.06 × 10 - 3) (± 1.09 × 10 - 4)$	155.40	0.77
4	毁坏的土坯房内睡眠人数	毁坏的土坯房内清醒人数	毁坏的非土坯房内人数	—	$y = 1.96 × 10 - 2 · X 1 + 1.17 × 10 - 3 · X 2 + 5.66 × 10 - 4 · X 3 (± 3.67 × 10 - 3) (± 2.80 × 10 - 4) (± 1.85 × 10 - 4)$	160.57	0.77

Tab.2

Tab.3

Weighted mortality model for population spatial distribution

模型编号	优选模型参数				优选模型	AIC	R²
模型编号	X₁	X₂	X₃	X₄	优选模型	AIC	R²
5	毁坏房屋内睡眠人数	毁坏房屋内清醒人数	—	—	$y = 5.12 × 10 - 3 · X 1 + 8.64 × 10 - 4 · X 2 (± 1.12 × 10 - 3) (± 1.14 × 10 - 4)$	234.96	0.47
6	毁坏的土坯房内睡眠人数	毁坏的土坯房内清醒人数	毁坏的非土坯房内睡眠人数	毁坏的非土坯房内清醒人数	$y = 1.61 × 10 - 2 · X 1 + 1.34 × 10 - 3 · X 2 + 2.20 × 10 - 3 · X 3 + 5.31 × 10 - 4 · X 4 (± 4.34 × 10 - 3) (± 3.10 × 10 - 4) (± 1.16 × 10 - 3) (± 1.16 × 10 - 4)$	154.36	0.77

Tab.3

Tab.4

Tab.5

References 37

[1]	LI Yilong, ZHANG Zhenguo, XIN Danhua. A composite catalog of damaging earthquakes for mainland China[J]. Seismological Society of America, 2021, 92(6): 3767-3777.
[2]	刘振亮, 苑溦, 李素超. 地震及次生突发灾害下公路桥梁网络韧性评估[J]. 中国安全科学学报, 2022, 32(8): 176-184. doi: 10.16265/j.cnki.issn1003-3033.2022.08.1457
	LIU Zhenliang, YUAN Wei, LI Suchao. Resilience assessment of highway bridge networks subjected to both earthquakes and earthquake-induced secondary emergencies[J]. China Safety Science Journal, 2022, 32(8): 176-184. doi: 10.16265/j.cnki.issn1003-3033.2022.08.1457
[3]	FIEDRICH F, GEHBAUER F, RICKERS U. Optimized resource allocation for emergency response after earthquake disasters[J]. Safety Science, 2000, 35(1/2/3): 41-57. doi: 10.1016/S0925-7535(00)00021-7
[4]	陈永强, 吴志鹏, 杨湘娟, 等. 区域地震分析软件研究综述[J]. 中国安全科学学报, 2022, 32(9): 126-136. doi: 10.16265/j.cnki.issn1003-3033.2022.09.2808
	CHEN Yongqiang, WU Zhipeng, YANG Xiangjuan, et al. Review on regional seismic analysis software[J]. China Safety Science Journal, 2022, 32(9): 126-136. doi: 10.16265/j.cnki.issn1003-3033.2022.09.2808
[5]	DUGGAL R, GUPTA N, PANDYA A, et al. Building structural analysis based internet of things network assisted earthquake detection[J]. Internet of Things, 2022, 19:DOI:10.1016/j.iot.2022.100561.
[6]	QABPE Z, BJZA B, YTCA B, et al. Rapid report of seismic damage to buildings in the 2022 M 6.8 Luding earthquake, China[J]. Earthquake Research Advances, 2023, 3(1): DOI:10.1016/j.eqrea.2022.100180.
[7]	谷国梁, 安立强, 朱宏, 等. 城市地震压埋人员分布评估研究:以天津市区为例[J]. 地震工程学报, 2021, 43(6): 1352-1360.
	GU Guoliang, AN Liqiang, ZHU Hong, et al. Assessment of seismic buried personnel in urban area: a case study of Tianjin area, China[J]. China Earthquake Engineering Journal, 2021, 43(6): 1352-1360.
[8]	COBURN A W. Assessing strategies to reduce fatalities in earthquakes[C]. Proceedings of International Workshop on Earthquake Injury Epidemiology for Mitigation and Response, 1989: 107-132.
[9]	COBURN A W, SPENCE R J S, POMONIS A. Factors determining human casualty levels in earthquakes: mortality prediction in building collapse[C]. Proceedings of the Tenth World Conference on Earthquake Engineering, 1992, 10: 5989-5994.
[10]	BRAGATO P L, BARNABA C, SCAINI C, et al. Time-invariance and geographical divide of earthquake mortality in Italy since 1800[J]. Annals of Geophysics, 2020, 63(6):DOI: 10.4401/ag-8406..
[11]	MARUYAMA Y, YAMAZAKI F, MIZUNO K, et al. Fragility curves for expressway embankments based on damage datasets after recent earthquakes in Japan[J]. Soil Dynamics and Earthquake Engineering, 2010, 30(11): 1158-1167. doi: 10.1016/j.soildyn.2010.04.024
[12]	ZHANG Jian, HU Wang, CHEN Weifeng, et al. Study on the mortality of earthquake in Sichuan province based on particle swarm optimization[J]. Control Engineering of China, 2018, 25(6): 1039-1043.
[13]	许建东, 危福泉, 张来泉, 等. 地震人员伤亡与压埋人员评估方法的初步研究:以福建省漳州市区为例[J]. 地震研究, 2008, 31(3): 382-387.
	XU Jiandong, WEI Fuquan, ZHANG Laiquan, et al. Preliminary study on evaluating the number of casualties and trapped victims by an earthquake: a case study of Zhangzhou city, Fujian province[J]. Journal of Seismological Research, 2008, 31(3): 382-387.
[14]	田丽莉. 地震灾害人员伤亡影响因素分析及人员伤亡估算公式[D]. 北京: 首都经济贸易大学, 2012.
	TIAN Lili. Analysis of factors affecting casualties in earthquake disasters and estimation formulas for casualties[D]. Beijing: Capital University of Economics and Business, 2012.
[15]	颜冬启. 服务于地震人员搜救的重点目标快速评估技术研究[D]. 哈尔滨: 中国地震局工程力学研究所, 2014.
	YAN Dongqi. Research on rapid assessment technology for key targets serving earthquake personnel search and rescue[D]. Harbin: Institute of Engineering Mechanics, China Earthquake Administration, 2014.
[16]	肖东升, 钟祖峰. 基于人类动力学的地震压埋人员先验在室率研究[J]. 灾害学, 2021, 36(1): 54-59.
	XIAO Dongsheng, ZHONG Zufeng. Research on indoor rate of the earthquake buried personnel based on human dynamics[J]. Journal of Catastrophology, 2021, 36(1): 54-59.
[17]	MAHDAVI A. Patterns and implications of user control actions in buildings[J]. Indoor and Built Environment, 2009, 18(5): 440-446. doi: 10.1177/1420326X09344277
[18]	ZHANG Xiaoyong, LU Qingshui. Estimating earthquake exposed population through mobile data and high-resolution imagery: a case study in main residential area of Zhangjiakou City, China[J]. Environmental Earth Sciences, 2021, 80(17): DOI:10.1007/s12665-021-09873-3.
[19]	程家喻. 地震发生时间对人员伤亡影响的概率[J]. 灾害学, 1993, 8(2): 13-16.
	CHENG Jiayu. Probability of the original time of earthquake affecting the casualty[J]. Journal of Catastrophology, 1993, 8(2): 13-16.
[20]	肖东升, 黄丁发, 陈维锋. 基于地震压埋率的在室率先验性研究[J]. 自然灾害学报, 2009, 18(3): 92-97.
	XIAO Dongsheng, HUANG Dingfa, CHEN Weifeng. Study on apriority of in-building rate based on buried rate[J]. Journal of Natural Disasters, 2009, 18(3): 92-97.
[21]	李良, 李路云. 地震发生时个体应急行为规律的探索性研究[J]. 管理工程学报, 2022, 36(1): 148-158.
	LI Liang, LI Luyun. Exploratory research on the patterns of individual emergency behaviors during earthquakes[J]. Journal of Industrial Engineering Management, 2022, 36(1): 148-158.
[22]	曹红斌, 孟耀斌, 赵晗萍, 等. 村镇环境居民健康风险评价与预警研究[M]. 北京: 中国环境出版集团, 2022:201-260.
[23]	中国地震局震灾应急救援司. 2006—2010年中国大陆地震灾害损失评估汇编[M]. 北京: 地震出版社, 2015:81-542.
[24]	GB/T 18208.4—2011, 地震现场工作第4部分:灾害直接损失评估[S].
	GB/T 18208.4-2011, Post-earthquake field works: part 4: assessment of direct loss[S].
[25]	王宏明, 张伟淼, 唐巍. 浅谈农村土坯房的修缮与抗震加固[J]. 工程抗震与加固改造, 2021, 43(1): 113-116.
	WANG Hongming, ZHANG Weimiao, TANG Wei. A preliminary analysis of the repair and seismic reinforcement of rural adobe house[J]. Earthquake Resistant Engineering and Retrofitting, 2021, 43(1): 113-116.
[26]	YÖN B. Identification of failure mechanisms in existing unreinforced masonry buildings in rural areas after April 4, 2019 earthquake in Turkey[J]. Journal of Building Engineering, 2021, 43: DOI: 10.1016/j.jobe.2021.102586.
[27]	徐舜华, 孙军杰, 王兰民, 等. 甘肃省土坯房空间分布特征与多因素分类方法研究[J]. 震灾防御技术, 2010, 5(1): 125-136.
	XU Shunhua, SUN Junjie, WANG Lanmin, et al. Study on regional distribution characteristic and multi-factor classification about adobe house of Gansu province[J]. Technology for Earthquake Disaster Prevention, 2010, 5(1): 125-136.
[28]	朱雁茹, 谢杰, 钟宪明, 等. 既有村镇房屋现状调查及抗震对策研究展望[J]. 工程抗震与加固改造, 2016, 38(1): 150-158.
	ZHU Yanru, XIE Jie, ZHONG Xianming, et al. Prospect of research on seismic countermeasures based on village building situation[J]. Earthquake Resistant Engineering and Retrofitting, 2016, 38(1): 150-158.
[29]	林伟庆. 农村危旧土坯房改造工作存在的问题及对策[J]. 审计与理财, 2013 (8): 52-53.
	LIN Weiqing. The problems and countermeasures in the renovation of dangerous old adobe houses in rural areas[J]. Audit and Wealth Management, 2013(8): 52-53.
[30]	中国地震局工程力学研究所. 建(构)筑物地震破坏等级划分[Z]. 2009-12-01.
[31]	AKAIKE H. A new look at the statistical model identification[J]. IEEE Transactions on Automatic Control, 1974, 19(6): 716-723. doi: 10.1109/TAC.1974.1100705
[32]	MOLLAHALILOGLU S, YARDıM M, TELATAR T G, et al. Change in the geographic distribution of human resources for health in Turkey, 2002-2016[J]. Rural and Remote Health, 2021, 21(1): 1-7.
[33]	ZHENG Xinqi, XIA Tian, YANG Xin, et al. The land Gini coefficient and its application for land use structure analysis in China[J]. PLoS One, 2013, 8(10): DOI:10.1371/journal.pone.0076165.
[34]	CAVANAUGH J E, NEATH A A. The akaike information criterion: background, derivation, properties, application, interpretation, and refinements[J]. Wiley Interdisciplinary Reviews-Computational Statistics, 2019, 11(3): DOI: 10.1002/wics.1460.
[35]	宜宾市年鉴(2020)[Z]. 2021-03-10.
[36]	籍裴希, 孟耀斌, 叶琬. 村镇居民生活模式模拟在地震死亡估计中的应用[J]. 中国安全科学学报, 2017, 27(1): 163-168. doi: 10.16265/j.cnki.issn1003-3033.2017.01.029
	JI Peixi, MENG Yaobin, YE Wan. Modeling living pattern of China village and township residents and its application in earthquake fatality estimation[J]. China Safety Science Journal, 2017, 27(1): 163-168. doi: 10.16265/j.cnki.issn1003-3033.2017.01.029
[37]	夏朝旭, 聂高众, 李华玥, 等. 基于历史地震数据的地震烈度与人员死亡率相关性研究[J]. 中国地震, 2022, 38(1): 153-165.
	XIA Chaoxu, NIE Gaozhong, LI Huayue, et al. Research on correlation between earthquake intensity and mortality rate based on historical earthquake data[J]. Earthquake Research in China, 2022, 38(1): 153-165.

不确定性类型	房屋结构参数	是否采用权重修正	仿真选中优选模型次数	仿真选中次优模型次数
土坯房占比	土坯房与否	否	108	63
土坯房占比	土坯房与否	是	107	75
死亡人数	不考虑	否	200	0
	不考虑	是	200	0
	土坯房与否	否	200	0
	土坯房与否	是	200	0