lifeline resilience model and its applicationcommittees.jsce.or.jp/kokusai/system/files/190522...3...
TRANSCRIPT
Session 3: Systems Resilience and Economic Impact
Lifeline Resilience Model and its Application
The first JSCE‐ASCE Symposium on Infrastructure ResilienceDate: May 22‐23, 2019
Venue: Japan Society of Civil Engineers, Auditorium
1
Nobuoto NojimaGifu University
Seismogenic Zones in and around Japan
2
1995 Hyogoken‐Nambu EQ (Kobe EQ)
http://www.j-shis.bosai.go.jp/
2011 off the Pacific Coast of Tohoku EQ
20XX Nankai Mega‐thrustHuge Earthquake (>90% in 50years)
2016 Kumamoto EQ
3
2.60
1.26 0.86
4.86
2.20
0.46 0.48 0.45 0.10
4.05
0
1
2
3
4
5
6
7
8
9
E W G E W G E W G
1995 2011 2016
Initial outage (in
million custom
ers or hou
seho
lds) Tokyo EPCO
Tohoku EPCO
1995: Hyogoken‐Nambu EQ (Kobe EQ)2011: Off the Pacific Coast of Tohoku EQ2016: Kumamoto EQ
Initial outage of lifeline services(number of households in million)
E : Electric power supplyW : Water supplyG : City gas supply
4
0%10%20%30%40%50%60%70%80%90%
100%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82
Resoratio
n ratio
Day
WaterCity gasElectric power
0%10%20%30%40%50%60%70%80%90%
100%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82
Resoratio
n ratio
Day
Water
City gas
Electric power (Tokyo EPCO)
Electric power (Tohoku EPCO)
0%10%20%30%40%50%60%70%80%90%
100%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82
Resoratio
n ratio
Day
WaterCity gasElectric power
(Day 1 = The day of occurrence of each earthquake)
Electric power supply
Water supply
City gas supply
The Great Hanshin-Awaji EQ Disaster1995 Hyogoken‐Nambu EQ (Kobe EQ)
The Great East Japan EQ Disaster2011 off the Pacific Coast of Tohoku EQ
2016 Kumamoto EQ
Normalized restoration curves(Resilience of each lifeline)
Component performance
Networkperformance
Acting load
Systemperformance
Personalperformance
Time Time
Failu
re p
rob.
Perc
ent r
esto
red
Acce
ptan
ceTimeM
ass
Acce
ptan
ce
SocietalperformanceNew criterion
Framework for assessment of indirect impact due to lifeline disruption
0.00.10.20.30.40.50.60.70.80.91.0
0 10 20 30 40 50 60 70 80 90 100
Residu
al ca
pacity
Time after earthquake (in day)
4.5(original)5.0(original)5.5(original)6.0(original)6.5(original)7.0(original)4.5(modified)5.0(modified)5.5(modified)6.0(modified)6.5(modified)7.0(modified)0.0
0.10.20.30.40.50.60.70.80.91.0
0 10 20 30 40 50 60 70 80 90 100
Residu
al ca
pacity
Time after earthquake (in day)
4.5(original)5.0(original)5.5(original)6.0(original)6.5(original)7.0(original)4.5(modified)5.0(modified)5.5(modified)6.0(modified)6.5(modified)7.0(modified)
0.00.10.20.30.40.50.60.70.80.91.0
0 50 100 150 200 250
Residu
al ca
pacity
Time after earthquake (in hour)
4.55.05.56.06.57.0
Post-earthquake lifeline serviceability model (modified, example)
Res
idua
l cap
acity
Water
Days after EQ.
( , )WP I t
Electricity
Hours after EQ.
Res
idua
l cap
acity
( , )EP I t
Res
idua
l cap
acity City Gas
Days after EQ.
( , )GP I t
(Nojima et al., 2012)
Deteriorated performance level after the EQ
Restored performance level
( , ) 1 ( ) ( ) ( | )P I t p I p I F t I
Emergency shutoff regulations and prompt first response compared to 1995
Effects of the vulnerability of pipelines compared to Kobe region in 1995
Assessment of Lifeline Disruption in the Nankai Trough Huge Earthquake (Mw9.0)
Right after EQ 1 day after 3 days after 1 week after 2 weeks after 1 month after 2 months after
0
500
1000
1500
2000
2500
3000
3500
4000
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102
108
114
120
126
132
138
144
150
156
162
168
供給
支障
人口
(万人
)
地震後経過時間
全国
0
100
200
300
400
500
600
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102
108
114
120
126
132
138
144
150
156
162
168
供給
支障
人口
(万人
)
地震後経過時間
新潟県 富山県
石川県 福井県
山梨県 長野県
岐阜県 静岡県
愛知県
0
5
10
15
20
25
30
35
40
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102
108
114
120
126
132
138
144
150
156
162
168
供給
支障
人口
(万人
)
地震後経過時間
豊橋市 岡崎市
豊田市 一宮市
名古屋市中川区 名古屋市緑区
安城市 豊川市
西尾市 春日井市
0200400600800
100012001400160018002000
0 7 14 21 28 35 42 49 56
供給
支障
人口
(万人
)
地震後経過日数
全国
0
50
100
150
200
250
300
350
400
0 7 14 21 28 35 42 49 56
供給
支障
人口
(万人
)
地震後経過日数
新潟県 富山県
石川県 福井県
山梨県 長野県
岐阜県 静岡県
愛知県
0
5
10
15
20
25
30
35
40
0 7 14 21 28 35 42 49 56
供給
支障
人口
(万人
)
地震後経過日数
豊橋市 岡崎市 一宮市
豊田市 安城市 西尾市
豊川市 半田市 刈谷市
稲沢市
0
50
100
150
200
250
300
350
0 7 14 21 28 35 42 49 56
供給
支障
人口
(万人
)
地震後経過日数
全国
0
20
40
60
80
100
120
0 7 14 21 28 35 42 49 56
供給
支障
人口
(万人
)
地震後経過日数
新潟県 富山県
石川県 福井県
山梨県 長野県
岐阜県 静岡県
愛知県
0
5
10
15
20
25
30
0 7 14 21 28 35 42 49 56
供給
支障
人口
(万人
)
地震後経過日数
豊橋市 西尾市
安城市 半田市
豊川市 名古屋市南区
碧南市 名古屋市港区
田原市 名古屋市中川区
Nationwide Prefectural Municipal
7
City gas
Water
Electric power
Excel/VBA tools
City gas
Water
Electric power
Socio-economic Activities
Infrastructure system of systems
Built Environment
End Users
Physical collocation/connection
Functional dependency
Operational dependency/trade-offs
System of Systems
Probability of appearance of 23=eight disruption patterns
0.00.10.20.30.40.50.60.70.80.91.0
0 20 40 60 80 100Time after earthquake (in day)
Res
idua
l cap
acity
4.55.05.56.06.57.0
City Gas0.00.10.20.30.40.50.60.70.80.91.0
0 20 40 60 80 100Time after earthquake (in day)
Res
idua
l cap
acity
4.55.05.56.06.57.0
Water0.00.10.20.30.40.50.60.70.80.91.0
0 50 100 150 200 250Time after earthquake (in hour)
Res
idua
l cap
acity
4.55.05.56.06.57.0
Electricity
( , )EP I t
0.00.10.20.30.40.50.60.70.80.91.0
0 10 20 30 40 50 60 70 80 90 100Time after earthquake (in day)
0.00.10.20.30.40.50.60.70.80.91.0
0 10 20 30 40 50 60 70 80 90 100Time after earthquake (in day)
EWG=000EWG=001EWG=010EWG=100EWG=110EWG=101EWG=011EWG=111
Seismic intensity 5.5 (severe) Seismic intensity 6.5 (very severe)
EWG=111 : All utilities are operational
EWG=000 : All utilities arenon-operational
EWG=100 : Only electricityis operational
EWG=110 : Electricity and waterare operational
Days after EQ. Days after EQ.
1
, ,
( , , ; , ) ( , ) 1 ( , ) kkE W G k k
k E W G
Q I t P I t P I t
( , )WP I t( , )EP I t ( , )GP I t
Rate of satisfaction of industrial sector(Food manufacture subsector)
Lifeline importance weights (ATC-25)
E : Electric power supplyW : Water supplyG : City gas supply
0.90
0.25
0.70
0.0
0.2
0.4
0.6
0.8
1.0
E W G
Moreimportant
Less important
Which pattern appears? When?1) seismic intensity level2) restoration process
23=Eight disruption patterns
Rate of satisfaction(Resiliency factor)
1.00
0.10
0.30
0.75
0.30
0.10 0.10 0.100.0
0.2
0.4
0.6
0.8
1.0
EW
G=1
11
EW
G=0
11
EW
G=1
01
EW
G=1
10
EW
G=1
00
EW
G=0
10
EW
G=0
01
EW
G=0
00
Intact
Completeloss
1 : Operational0 : Non-operational
E
GW
( , , )i E W GR
Time function of average rate of satisfaction
0.00.10.20.30.40.50.60.70.80.91.0
0 10 20 30 40 50 60 70 80 90 100Time after earthquake (in day)
1.00
0.10
0.30
0.75
0.30
0.10 0.10 0.100.0
0.2
0.4
0.6
0.8
1.0E
WG
=111
EW
G=0
11
EW
G=1
01
EW
G=1
10
EW
G=1
00
EW
G=0
10
EW
G=0
01
EW
G=0
00
Prob. of appearance of disruption patterns
( , , ; , )E W GQ I t ( , , )i E W GR
Seismic intensity 6.5 (very severe)
0.00.10.20.30.40.50.60.70.80.91.0
0 10 20 30 40 50 60 70 80 90 100Time after earthquake (in day)Days after EQ.Av
erag
e ra
te o
f sat
isfa
ctio
n
Weighted sumall( , , )
( , ) ( , , ) ( , , ; , )E W G
i i E W G E W GR I t R Q I t
6.5
0.00.10.20.30.40.50.60.70.80.91.0
0 10 20 30 40 50 60 70 80 90 100Time after earthquake (in day)
5.05.56.06.57.0
Days after EQ.
( , )iR I t
Various seismic intensity
5.05.56.0
7.0
Rate of satisfaction
Post-earthquake daily shipment values for various industrial subsectors
In Shizuoka prefecture for the hypothetical Tokai earthquake
( )iEC t
0
2,000
4,000
6,000
8,000
10,000
12,000
0 20 40 60 80 100 120 140Time after earthquake (in day)
FoodTextileLumber & WoodCeramicPulp & PaperChemicalPetroleum and CoalMetalGeneral MachineryPrecision Eq.Electrical Eq.Transportation Eq.Misc.
Days after EQ.
Transportation equipment
Electrical equipment
Chemical product
(in million yen)
Data compilation
MODELING POST-EARTHQUAKE SERVICEABILITY OF RAILSWAY SYSTEM BASED ON THE DATABASE OF THE GREAT EAST JAPAN EARTHQUAKE DISASTER
Nobuoto NOJIMA & Hiroki KATO Department of Civil Engineering, Gifu University , Japan, Email: [email protected]
Estimation of possibility and duration of suspension of railway service is an important issue. In this study, statistical analyses have been carriedout for evaluation of post-earthquake serviceability of railway systems in the 2011 Great East Japan Earthquake Disaster on the basis of JMAseismic intensity. By following the two-step evaluation model for serviceability of utility lifelines proposed by the authors, an empirical model hasbeen statistically-derived to predict railway service suspension in anticipate earthquake scenarios.KEYWORDS: Railway service suspension; Initial outage and duration; The Great East Japan Earthquake Disaster
Abstract
0
100
200
300
400
500
600
700
800
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Freq
uency
JMA seismic intensity
No suspensionWithin 1 day2 days or longerAffected by tsunami
Histogram of duration of suspension in terms of JMA
seismic intensity
JMA seismic intensity vs. suspension ratio
(Observed ratio and logit model)
Case 1: all data (excluding tsunami damage)
Case 2: excluding suspension within 1 day and tsunami damage
0%10%20%30%40%50%60%70%80%90%
100%
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Susp
ensi
on ra
tio
JMA seismic intensity
Obs:Case1Obs:Case2Model:Case1Model:Case2
Seismic intensities for operation control
A database was compiled with regard to GIS data on railway lines andstations, service suspension caused by ground shaking and tsunamiinundation.
Shaking intensity maps in terms Japan Meteorological Agency (JMA) seismicintensity scale were also compiled.
Model development (Step 1: Initial probability of suspension) The relationship between the incidence of railway service suspension
and shaking intensity was evaluated. A logit model was fitted to predict the probability of occurrence of service
suspension as a function of shaking intensity.
0 1
0 1
exp[ ]1 ex ]
(p[
) b b Ib
p Ib I
JMA seismic intensity along railways
JMA seismic intensity
1) Based on the relationship between occurrence of railway servicesuspension and JMA seismic intensity, functional fragility functionswere derived for two cases: Case 1: “whether there wassuspension or not,” and Case 2: “whether there was suspensionwith 2 days or longer.” Case 1 showed high goodness-of-fit,although suspension ratio is too high at low seismic intensity. Onthe contrary, Case 2 showed low goodness-of-fit.
2) The relationship between the duration of suspension and JMAseismic intensity. Case 1 showed clear tendency of increasingduration of suspension with increasing intensity in the range from5.0 to 6.0. Case 2 showed such increasing tendency for wider
Conclusions
Takada et al (2001)
Moving average and moving standard deviation of duration of suspension
Statistical model of duration of suspension represented
by gamma distribution
Prototype of post-earthquake serviceability curves of railways
for various JMA intensities.
range of intensity. The coefficients of variation were as large asalmost 100%. Gamma distribution was fitted to predict theduration of suspension under the condition that servicesuspension occurs.
3) By combining two sub-models, a prototype of post-earthquakeserviceability curve for railway systems was derived in term ofJMA seismic intensity. Because of the statistical fluctuation ofparameters, some irregular tendency can be seen, which shouldbe eliminated in the model for practical use. For improving themodel, further analysis is needed to consider additional factorsand incorporate appropriate explanatory variables.
0
10
20
30
40
50
60
70
80
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Dur
atio
n of
sus
pens
ion(
in d
ay)
JMA seismic intensity
Within 1 day2 days or longer
0
10
20
30
40
50
60
70
80
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Dur
atio
n of su
spen
sion(
in d
ay)
JMA seismic intensity
Moving average(obs)
Moving std.dev(obs)
0
10
20
30
40
50
60
70
80
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Dur
atio
n of su
spen
sion(
in d
ay)
JMA seismic intensity
Moving average(obs)
Moving std.dev(obs)
0
10
20
30
40
50
60
70
80
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Dur
atio
n o
f su
spens
ion(in
day)
JMA seismic intensity
Non exceedance 90%
Non exceedance 70%
Non exceedance 50%
Non exceedance 30%
Non exceedance 10%
0
10
20
30
40
50
60
70
80
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Dur
atio
n o
f su
spen
sion(in
day
)JMA seismic intensity
Non exceedance 90%
Non exceedance 70%
Non exceedance 50%
Non exceedance 30%
Non exceedance 10%
0.00.10.20.30.40.50.60.70.80.91.0
0 10 20 30 40 50 60 70 80 90 100
Prob
abilit
y of
ser
vice
ava
ilabi
lity
Time after earthquake (in day)
3.03.54.04.55.05.56.06.5
0.00.10.20.30.40.50.60.70.80.91.0
0 10 20 30 40 50 60 70 80 90 100
Prob
abilit
y of
ser
vice
ava
ilabi
lity
Time after earthquake (in day)
3.03.54.04.55.05.56.06.5
JMA seismic intensity vs. duration of suspension
Model development (Step 2: Duration of suspension given that suspension occurred) Model development (Step 3) The relationship between the duration of suspension and shaking intensity was evaluated. Gamma distribution was fitted to predict the duration of suspension under the condition that service
suspension occurs.
By combining two sub-models in steps 1 and 2, aprototype of an integrated of post-earthquakeserviceability curve for railway systems was derived.
Case 1: all data (excluding tsunami damage)
Case 2: excluding suspension within 1 day and tsunami damage ( , ) 1 p IP I t F t Ip I
( ) 1
( )
exp( )
( ) ( (( | )
))
I
I
ttI
I If t I
)()()(,
)()()(
22
III
III
0(( | )) |
tfF dIt I
Case 1: all data (excluding tsunami damage)
Case 2: excluding suspension within 1 day and tsunami damage
Pre-EQ TimeEarthquake
Physical spread Functional spread Impeding recovery works
Inter-dependent
total system
Lifeline System InteractionsSupply : Water/Electricity/Gas Disposal : Sewage/Garbage Transportation : Road/Railway Communication : Telephone
Physical/Functional/Operationalinterconnection
System interactions
15
Lifeline System Interactions due to Geographical Proximity of Network Facilities
• Water Road : - Leak of water can wash out a road.
• Sewer Water : - Leak of wastewater from pipe breaks can contaminate drinking water.- Use of water may be restricted until damaged sewer system is restored.
• Water Gas Sewer Electric power- Conflicts of repair works may degrade recovery efficiency.- Organizational coordination may be required so as to avoid conflicts.
System 1System 2 System 1
System 2
System 1
Surface Ground
System 2
Collocation Geographical interdependency
16
How many coincident damages occur to the multiple lifeline systems in an earthquake?
Water delivery network(Extended length: 23,600km)
Water Distribution and Sewer Lines
下水道管路無し
Sewer network(Extended length: 14,400km)
Pipe failure:8135 breaks Pipe failure: 17160 breaksDamage length of 25m = 1 pipe break
Data source: Chiba Prefectural Government Offices and Chiba University
Penetration ratio=93% Penetration ratio=64%
17
The Hypothetical Northern Tokyo Bay Earthquake (M=7.3)
18
1-D modell =100 m
Total: 988
Coincident Damage to Water Delivery and Sewer Systems
Total: 1688
2-D model
l l
l
Reason of evacuation in 2016 Kumamoto EQ
19
The reasons to return home from shelters (Single answer)
The reasons to evacuate to shelters (Multiple answer, top 3)
Source: Mainichi Daily News
Disruption of electricityand/or water supply
Fear of aftershocks
Fear of danger ofdamaged houses
OthersClosure of shelters
Found the other placeRecovery of city gas
Recovery of electricityRecovery of water
Decrease of aftershocks
Questionnaire survey conducted by Kumamoto City Office
% %
Lifelines
16,050
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220
避難者数(人)
経過日数(日)
益城町 西原村 南阿蘇村 御船町
108,266
0
20000
40000
60000
80000
100000
120000
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220
避難者数(人)
益城町 熊本市
Number of evacuees in EQ disasters
316,678
103,178
468,600
183,746
0
50000
100000
150000
200000
250000
300000
350000
400000
450000
500000避難者数
経過日数(日)
兵庫県南部地震
新潟県中越地震
東北地方太平洋沖地震※1
東北地方太平洋沖地震※2
熊本地震※3
(最大避難者数)2011 Tohoku EQ: 468,600
1995 Kobe EQ: 316,678
2016 Kumamoto EQ: 183,746
2014 Niigata-ken Chuetsu EQ: 103,178
Elapsed days
Elapsed days
Kumamoto City: 108,266
Mifune TownMinamiaso Village
Nishihara Village: 2,951 (42% of population)
Mashiki Town: 16,050 (46% of population)
21
0
5000
10000
15000
20000
25000
30000
35000
40000
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220
Num
ber o
f peo
ple
Elapsed days(day)
CollapsedHalf‐collapsed or collapsedEvacuees(observed)Without waterWithout powerEvacuees(predicted,proposed)Evacuees(predicted,conventional)
0
5000
10000
15000
20000
25000
30000
35000
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Day 0=April 14
Observed number of evacuees
Estimatednumber of evacuees
Population without water
Population without power
Population occupying major damage houses
Population occupying major + minor damage houses
Observed and estimated number of evacuees
Mashiki Town
𝑀 𝑡 : Population without electric power supply
𝑒 : Evacuation ratio under disruption of electric power and/or water supply
𝑅 𝑡 : Resilience time function describing the long‐term effect
𝑁 𝑡 𝑀 𝑒 𝑀 𝑒 𝑅 𝑡𝑀 𝑀 𝑒 𝑀
𝑀 max 𝑀 𝑡 ,𝑀 𝑡 𝑒
Building damage Disruption and recovery of water and electricity
Multi-disciplinary mitigation options
Component performance
Networkperformance
Acting load
Systemperformance
Personalperformance
Time Time
Failu
re p
rob.
Perc
ent r
esto
red
Acce
ptan
ceTimeM
ass
Acce
ptan
ce
SocietalperformanceMinimization!