lifeline resilience model and its applicationcommittees.jsce.or.jp/kokusai/system/files/190522...3...

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


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

)


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

)


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

)


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

)


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

)


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

)


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

)


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


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


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


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


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


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)


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)


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)


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)


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


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


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 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 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!

lifeline resilience model and its applicationcommittees.jsce.or.jp/kokusai/system/files/190522...3...

Documents