HUMANITARIAN LOGISTICS IN THE TOHOKU DISASTERS 2011 Eiichi - - PowerPoint PPT Presentation

▶
humanitarian logistics in the tohoku disasters 2011
SMART_READER_LITE
LIVE PREVIEW

HUMANITARIAN LOGISTICS IN THE TOHOKU DISASTERS 2011 Eiichi - - PowerPoint PPT Presentation

Aug 16, 2023 1.27k likes •1.82k views

HUMANITARIAN LOGISTICS IN THE TOHOKU DISASTERS 2011 Eiichi Taniguchi, Yuki Nakamura, Kyoto University, Japan Jos Holgun-Veras, Rensselaer Polytechnic Institute, USA Eiichi Taniguchi, Kyoto University 1. Introduction Disasters and

slide-1
SLIDE 1

Eiichi Taniguchi, Kyoto University

HUMANITARIAN LOGISTICS IN THE TOHOKU DISASTERS 2011

Eiichi Taniguchi, Yuki Nakamura, Kyoto University, Japan José Holguín-Veras, Rensselaer Polytechnic Institute, USA

slide-2
SLIDE 2

Eiichi Taniguchi, Kyoto University

  • 1. Introduction

Disasters and Catastrophes

  • Disasters
  • A limited part of region is impacted by natural events
  • Local supplies exist
  • Private sector can quickly respond to supply
  • Small demands compared with supplies
  • Catastrophes
  • Major part of region is impacted by natural events
  • Local supplies are minimal
  • Supply chain of private sector is severely impacted and

private sector cannot respond for weeks after the event

  • Huge increase in demands of surviving population and

response itself (e.g. 80,000 displaced people vs. 5,000 meals per day in Ishinomaki City just after Tohoku EQ)

slide-3
SLIDE 3

Eiichi Taniguchi, Kyoto University

Humanitarian logistics

  • Humanitarian Logistics is defined as the process of

planning, implementing and controlling the efficient, cost-effective flow and storage of goods and materials, as well as related information, from the point of origin to the point of consumption for the purpose of alleviating the suffering of vulnerable people (Thomas, 2003)

slide-4
SLIDE 4

Eiichi Taniguchi, Kyoto University

Commercial and Humanitarian Logistics

Characteristic Commercial Logistics Post-Disaster Humanitarian Logistics Objective Minimization of private (logistics) costs Minimization of social costs (logistic + deprivation) Origination of commodity flows Self-contained Impacted by material convergence Knowledge of demand Known with some certainty Unknown/dynamic, lack of information/access to site Decision making structure Structured interactions controlled by few decision makers Non-structured interactions, thousands of decision makers Periodicity/volume

  • f logistic activities

Repetitive, relatively steady flows, “large” volumes One in a lifetime events, large pulse, “small” volumes Supporting systems (e.g. transportation) Stable and functional Impacted and dynamically changing Source: Holguin-Veras et al, 2012

slide-5
SLIDE 5

Eiichi Taniguchi, Kyoto University

  • 2. Findings based on the interviews
  • Interviews to:
  • Ministry of Land, Infrastructure, Transport and

Tourism (Tohoku Regional Office)

  • Prefectures (Iwate, Miyagi, Fukushima)
  • Municipalities (Ishinomaki, Kesennuma)
  • Freight carriers (Yamato, Sagawa)
  • Retailers (Aeon)
  • Date:
  • 15 May -20 May, 29 August - 1 September 2011
slide-6
SLIDE 6

Eiichi Taniguchi, Kyoto University

Lessons learned (1)

1) The disaster response plans, particularly at the prefecture and city levels, failed to consider and prepare for worst case scenarios as they only focused on small disasters, that they could handle on their own 2) Not having disaster response plans that, in detail, considered humanitarian logistics hampered public sector response as local officials had to confront the crisis without any guidance about how to proceed

slide-7
SLIDE 7

Eiichi Taniguchi, Kyoto University

Lessons learned (2)

3) The lack of training and realistic exercises on humanitarian logistics significantly diminished the effectiveness of the public sector response to the disaster 4) The lack of technologies and systems to quickly assess the conditions of the transportation network and

  • ther critical infrastructure hampered the response

process

slide-8
SLIDE 8

Eiichi Taniguchi, Kyoto University

Lessons learned (3)

5) The lack of efficient communications with the field was a negative factor in the response that made the assessment of local needs a difficult task 6) The private sector in the construction, transportation, and retail sectors played a key role in the response as they brought to bear expertise and assets that benefited the response, though in some cases their participation was improvised and unanticipated

slide-9
SLIDE 9

Eiichi Taniguchi, Kyoto University

Lessons learned (4)

7) In catastrophic events, the bulk of the relief supplies has to be transported from the outside of the impacted area 8) The most challenging part of the entire humanitarian logistics process was the local distribution. For that reason, disaster planning must pay special attention to it so that all potential participants are prepared to deal with the challenge

slide-10
SLIDE 10

Eiichi Taniguchi, Kyoto University

Lessons learned (5)

9) Lack of fuel (for the return trips) prevented the volunteer participation of numerous trucking companies

  • n the humanitarian logistics effort during the initial

days of the crisis 10) Non/low priority donations created a lot of problems as they consumed significant amounts of resources, were not needed, and arrived at a time at which there were more important activities to undertake

slide-11
SLIDE 11

Eiichi Taniguchi, Kyoto University

  • 3. Modelling relief distribution

Background

  • Large scale earthquake
  • 11th March 2011 Great East Japan

earthquake disasters

  • M9.0 earthquake and tsunami

Damage to infrastructure 450,000 displaced people

slide-12
SLIDE 12

Eiichi Taniguchi, Kyoto University

500 1,000 1,500 2,000 （Number）

Background

  • Large scale earthquake
  • 11th March 2011 Great East Japan

earthquake disasters

  • M9.0 earthquake and tsunami

About 2,000 refuge centres at peak

Number of refuge centres

（Source: Cabinet Office）

slide-13
SLIDE 13

Eiichi Taniguchi, Kyoto University

Background

  • Large scale earthquake
  • 11th March 2011 Great East Japan

earthquake disasters

  • M9.0 earthquake and tsunami
  • Relief distribution of emergency

goods to refuge centres

Supporting the life of impacted people, reducing the secondary damage Insufficient pre-planning, a number of problems However…

slide-14
SLIDE 14

Eiichi Taniguchi, Kyoto University

Background

  • Large scale earthquake
  • 11th March 2011 Great East Japan

earthquake disasters

  • M9.0 earthquake and tsunami
  • Relief distribution of emergency

goods to refuge centres

Complicated inventory management Shortage of goods, fuel Shortage of delivery trucks and drivers

Insufficient pre- planning of relief distribution of emergency goods

slide-15
SLIDE 15

Eiichi Taniguchi, Kyoto University

Background

  • Large scale earthquake
  • 11th March 2011 Great East Japan

earthquake disasters

  • M9.0 earthquake and tsunami
  • Relief distribution of emergency

goods to refuge centres

Complicated inventory management Shortage of goods, fuel Shortage of delivery trucks and drivers

  • Need to determine the pre-planning of relief distribution in a

regional disaster prevention plan identifying the amount of goods, the number of trucks

slide-16
SLIDE 16

Eiichi Taniguchi, Kyoto University

Background

  • Large scale earthquake
  • 11th March 2011 Great East Japan

earthquake disasters

  • M9.0 earthquake and tsunami
  • Relief distribution of emergency

goods to refuge centres

 Complicated inventory management  Shortage of goods, fuel  Shortage of delivery trucks and drivers

  • Need to determine the pre-planning of relief distribution in a

regional disaster prevention plan identifying the amount of goods, the number of trucks

  • Mathematical programming models

Relief distribution plan Mathematical programming models Quantitative analysis Analysis

slide-17
SLIDE 17

Eiichi Taniguchi, Kyoto University

Objectives

Establishing multi-objective optimisation models for improving pre-planning of relief distribution of emergency goods after disasters and applying them in Ishinomaki city cases

(1) Develop multi-objective optimisation models (2) Evaluate performance of algorithms (3) Apply in Ishinomaki city cases (4) Compare calculation results with the reality (5) Analyse for improving pre-planning

slide-18
SLIDE 18

Eiichi Taniguchi, Kyoto University

Modelling-1

Starting term

Push type supply (smaller than needs) Self defence force

Active term

Pull type supply (various needs) Private freight carriers and Self defence force

Transition term

Reducing supply Transition to normal situation

（～2,3 days） （～2,3 months） （～6 months）

Target of modelling

slide-19
SLIDE 19

Eiichi Taniguchi, Kyoto University

Target of modelling

Modelling-1

Companies, individuals, municipalities outside the impacted areas Refuge centres Second ary depot Primary depot Senders’ depot Municipal ity Prefecture Nation

Flow of emergency goods Flow of information

transport request Order of sending

Active terms (after 3 days – 2, 3 months) Various needs Various delivery systems

Less difficult difficult

Focusing on last mile delivery by trucks

transport transport transport request request request Order of sending Order of sending

slide-20
SLIDE 20

Eiichi Taniguchi, Kyoto University

Objectives of relief distribution of emergency goods

  • Allocate appropriate amount of

goods to refuge centres

Modelling-2

Target of modelling

  • Active term (2,3 days – 2,3

months)

  • Last mile delivery by trucks
  • Foods (rice ball and bread)

Objective function f1

Minimize

( )

∑

∈

−

N i i i i

q d p

Minimising penalty of total shortage of supply

Allocate appropriate amount of goods which are small than demands to refuge centres

：delivered goods

i

q

i

p

：penalty to shortage of supply (priority )

i

d

：demand At refuge centre

i

( )

i i

q d −

：shortage of supply Decision variable

slide-21
SLIDE 21

Eiichi Taniguchi, Kyoto University

Modelling-2

Objective function f1

Minimize

( )

∑

∈

−

N i i i i

q d p

k K k A j i ijk ij

E x c /

) , (

∑ ∑

∈ ∈

Objective function f2

Minimize

Minimising total fuel consumption

Number of trucks and visiting

  • rder to reduce fuel consumption

Allocate appropriate amount of goods which are smaller than demands to refuge centres

Minimising penalty of total shortage of supply

∑

∈A j i ijk ijx

c

) , (

：distance travelled

k

E

：fuel efficiency Decision variable

Target of modelling

  • Active term (2,3 days – 2,3

months)

  • Last mile delivery by trucks
  • Foods (rice ball and bread)
  • Last mile delivery by trucks
  • Foods (rice ball and bread)

Objectives of relief distribution of emergency goods

  • Allocate appropriate amount of

goods to refuge centres

slide-22
SLIDE 22

Eiichi Taniguchi, Kyoto University

Formulation

( )

          − =

∑ ∑ ∑

∈ ∈ ∈ K k A j i k ijk ij N i i i i

  • pt

E x c q d p Minimize Z

) , (

/ ,

1 =

∑∑

∈ ∈ K k V j ijk

x

, N i ∈ ∀

1 =

∑

∈V j jk

x

, K k ∈ ∀

= −∑

∑

∈ ∈ V j hjk V i ihk

x x

, , K k N h ∈ ∀ ∈ ∀

1

, 1 ,

=

∑

∈ + V i k n i

x

, K k ∈ ∀

1

\

≥

∑ ∑

∈ ∈ S i S V j ij

x

), , ( , V S S V S ≠ ≠ ⊂ ∀ φ

Q x q

V j ijk N i i

≤

∑ ∑

∈ ∈

, K k ∈ ∀

i i

d q q ≤ ≤

min

, N i ∈ ∀

{ }

1 , =

ijk

x

, , , K k V j i ∈ ∀ ∈ ∀

Subject to

Total fuel consumption Penalty for total shortage of supply

T q

N i i ≤

∑

∈

Constraint of visiting by trucks Constraint of truck capacity Integer Minimum delivery to refuge centre Total supply in impacted area

Non-Dominated Sorting Genetic Algorithms（NSGA-II）for multi-

  • bjective optimisation
slide-23
SLIDE 23

Eiichi Taniguchi, Kyoto University

Case studies

  • Target area：Ishinomaki city

Obtain lessons learned from Great Tohoku Disasters

  • Analyse number of trucks and location of depot
  • Compare calculation results and reality

Obtain knowledge useful for the future disaster prevention plan

  • Assume some impacts
  • Present effective relief distribution under the shortage
  • f goods and fuels
slide-24
SLIDE 24

Eiichi Taniguchi, Kyoto University

Setting of case 1

Road network

  • Shortest path between refuge centres on

Ishinomaki road net work

  • Almost same as the roads where trucks can pass on

31st March

slide-25
SLIDE 25

Eiichi Taniguchi, Kyoto University

Road network

  • Shortest path between refuge centres on

Ishinomaki road net work

  • Almost same as the roads where trucks can pass on

31st March

Setting of case 1

Refuge centres

  • 21st March: 110, 41,922 displaced people
  • 11th April:152, 30,930 displaced people

Demand was assumed to be 3 meals for each person

slide-26
SLIDE 26

Eiichi Taniguchi, Kyoto University

Setting of case 1

道路ネットワーク

  • 石巻基本道路のうち避難所間最短経路（距離）
  • 3月31日時点での通行可能道路とほぼ一致

Location of depot

  • 3 candidates
  • Sport park, former vegetable market, Ushio house

ground

避難所（配送先）

  • 3月21日：110箇所，41,922人，需要量125,766個
  • 4月11日：152箇所，30,930人，需要量92,790個

JSDF and private freight carrier used this depot Private freight carrier used this depot after the sport park was closed Hypothetical depot Sport park Former vegetable market Ushio house

slide-27
SLIDE 27

Eiichi Taniguchi, Kyoto University

Setting of case 1

Road network

  • Shortest path between refuge centres on

Ishinomaki road net work

  • Almost same as the roads where trucks can pass on

31st March

Refuge centres

  • 21st March: 110, 41,922 displaced people
  • 11th April:152, 30,930 displaced people

Location of deport

  • 3 candidates
  • Sport park, former vegetable market, Ushio house

ground

slide-28
SLIDE 28

Eiichi Taniguchi, Kyoto University

Planning of delivery trucks（case1）

How many trucks of what capacity…

slide-29
SLIDE 29

Eiichi Taniguchi, Kyoto University

Planning of location of depots（case1）

Where and how many depots……

slide-30
SLIDE 30

Eiichi Taniguchi, Kyoto University

Results-1（case1）

5 10 15 20 25 30 35 40 45 50 30,000 35,000 40,000 45,000 50,000 55,000 60,000 65,000

1トントラック 4トントラック 10トントラック

Number of trucks Total fuel consumption(mL) Penalty of total shortage of supply 総供給不足ペナルティ 総消費燃料(mL) トラック台数（台）

Analysis of capacity of trucks

Number of trucks Total fuel consumption Penalty of total shortage of supply 10 ton truck 1 ton truck 4 ton truck

slide-31
SLIDE 31

Eiichi Taniguchi, Kyoto University

Results-2（case1）

Analysis on the location of depots

5 6 7 8 9 10 11 12 13 14 20,000 30,000 40,000 50,000 1配送拠点 2配送拠点 3配送拠点

Number of trucks

Total fuel consumption(mL) Penalty of total shortage of supply 総供給不足ペナルティ 総消費燃料(mL) トラック台数（台） Penalty of total shortage of supply Total fuel consumption Number of trucks 1 depot 3 depots 2 depots

slide-32
SLIDE 32

Eiichi Taniguchi, Kyoto University

Comparison of calculation results with reality on the number of trucks

  • 1 depot at sport park

Calculation results

  • 12 4-ton trucks

Reality

  • 20 trucks (JSDF and private freight carrier)

4 ton truck

slide-33
SLIDE 33

Eiichi Taniguchi, Kyoto University

Outline of case studies

  • Target area：Ishinomaki city

Obtain knowledge useful for the future disaster prevention plan

  • Assume some impacts
  • Present effective relief distribution under the shortage
  • f goods and fuels
slide-34
SLIDE 34

Eiichi Taniguchi, Kyoto University

Setting for case 2

Road network

  • Same as case 1

Depots

  • 1 depot at sport park

Assume that all residents are impacted

  • 50% of residents go to refuge centres(30% in

Tohoku disasters）

  • 44 refuge centres including elementary schools, high

schools, hospitals and old people’s home

slide-35
SLIDE 35

Eiichi Taniguchi, Kyoto University

Sufficiency rate

Sufficiency rate (%)：rate of satisfaction of demand at refuge centre

 ：sufficiency rate at refuge centre  ：supply at refuge centre  ：demand at refuge centre

100 × =

i i i

d q s

i

s

i

s

i

q

i

d

i i i i

slide-36
SLIDE 36

Eiichi Taniguchi, Kyoto University

Case 2

Case 2-0

Amount of goods proportional to the number of displaced people

Case 2-1

Amount of goods based

  • n priority

Case 2-2

Analysis for multi- terms (4 days)

slide-37
SLIDE 37

Eiichi Taniguchi, Kyoto University

Case 2-0

  • Deliver emergency goods proportional to the number of

displaced people

Sufficiency rate is 44.2% for all refuge centres.

sufficiency rate

slide-38
SLIDE 38

Eiichi Taniguchi, Kyoto University

Case 2-1

  • Determine the amount of delivery based on the priority rate
  • The priority rate is …..

i

p

  • Hospital, old people’s home

：1.2

  • Ordinary refuge centres in the

affected area by tsunami ：1.1

  • Other ordinary refuge centres

：1.0

i

q

Legend Penalty of total shortage of supply

slide-39
SLIDE 39

Eiichi Taniguchi, Kyoto University

10 20 30 40 50 97,000 98,000 99,000 100,000 101,000

総 消 費 燃 料 ( L ) 総供給不足ペナルティ ケース2-1 ケース2-0

Case 2-1

Both objective function decreased

  • Determine amount of delivery based on the priority
  • Set the priority rate:
  • Objective function

i

p

i

q

Penalty of total shortage of supply Total fuel consumption(L) case case

slide-40
SLIDE 40

Eiichi Taniguchi, Kyoto University

Case 2-1

There are refuge centre whose sufficiency rate is less than 20%

  • Determine the amount of goods delivered based on priority
  • Set the priority
  • The objective function value decreased

Sufficiency rate (%) i

q

i

p

slide-41
SLIDE 41

Eiichi Taniguchi, Kyoto University

Case 2-2

  • Set the sufficiency rate for the next day based on

sufficiency at the current day Rule for determining the priority

Sufficiency rate for the current day (%) 0.0~20.0 20.0~40.0 40.0~60.0 60.0~80.0 80.0~100.0 Priority for the next day Hospital, Old peoples’ home Ordinary refuge centres 1.2 1.2 1.2 1.1 1.2 1.0 1.2 0.9 1.2 0.8

slide-42
SLIDE 42

Eiichi Taniguchi, Kyoto University

Case 2-2（day 1）

Sufficiency rate (Day1)

Sufficiency rate (%)

slide-43
SLIDE 43

Eiichi Taniguchi, Kyoto University

Case 2-2（day 2）

Sufficiency rate (Day2)

Sufficiency rate (%)

slide-44
SLIDE 44

Eiichi Taniguchi, Kyoto University

Case 2-2（day 3）

Sufficiency rate (Day3)

Sufficiency rate (%)

slide-45
SLIDE 45

Eiichi Taniguchi, Kyoto University

Case 2-2（day 4）

Sufficiency rate (Day4)

Sufficiency rate (%)

slide-46
SLIDE 46

Eiichi Taniguchi, Kyoto University

Case 2-2（Total of 4 days）

Sufficiency rate for 4 days

No refuge centres whose sufficiency rate is under 20 %.

Sufficiency rate (%)

slide-47
SLIDE 47

Eiichi Taniguchi, Kyoto University

Case 2-2（Change of priority and amount of goods delivered)

Saito hospital Okaido elementary school

Delivering goods reflecting the priority has been achieved

0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 50 100 150 200 250 300 350 1日目 2日目 3日目 4日目 priority Amount of goods delivered 配送量 需要量 優先度 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1,000 2,000 3,000 4,000 5,000 1日目 2日目 3日目 4日目 priority Amount of goods delivered 配送量 需要量 優先度

supply supply demand demand priority priority Day 1 Day 2 Day 3 Day 4 Day 4 Day 3 Day 2 Day 1

slide-48
SLIDE 48

Eiichi Taniguchi, Kyoto University

Case 2-2（Sufficiency rate for 4 days）

The sufficiency rate for 4 days is:  High in hospitals and old people’s home  Average in ordinary refuge centres

30 35 40 45 50 55 60 病院老人ホーム 一般避難所 Sufficiency rate (%)

ケース2-2 ケース2-0×4日 Hospital and old people’s home Ordinary refuge centres Case 2-2 Case 2-0 * 4 days

slide-49
SLIDE 49

Eiichi Taniguchi, Kyoto University

Case 2-2（total fuel consumption during 4 days）

Total fuel consumption can be reduced while keeping good sufficiency rate.

50 70 90 110 130 150 170 190 ケース2-0×4日 ケース2-2 Total fuel consumption (L) 総消費燃料

Reduction by 23%

Case 2-0 *4 days Case 2-2 Total fuel consumption

slide-50
SLIDE 50

Eiichi Taniguchi, Kyoto University

  • 4. Conclusion
  • It is important to highlight:
  • the need to bring large amounts of outside resources to the

impacted area to satisfy the needs of survivors and the response process

  • the importance of prepositioning critical supplies
  • the magnitude and complexity of the humanitarian logistics

challenge and particularly of the local distribution, undoubtedly the most difficult part

slide-51
SLIDE 51

Eiichi Taniguchi, Kyoto University

Conclusion

  • the necessity to integrate all segments of society—the military,

public sector, the various strata of the private sector—to ensure the most efficient response permitted by the circumstances

  • the need to proactively manage and control the flows of

non/low/high priority supplies that accompany large disasters

  • The failure to prepare for the worst case scenario

hampered the ability of all layers of government to respond to the challenge

slide-52
SLIDE 52

Eiichi Taniguchi, Kyoto University

Conclusion

  • Catastrophic events require humanitarian logistics that

are orders of magnitude larger and more complex than the ones required in, the more frequent, smaller disasters

  • The importance of proper and effective donation

management

slide-53
SLIDE 53

Eiichi Taniguchi, Kyoto University

Conclusion

  • Regarding relief distribution of emergency goods after

disasters

  • We developed multi-objective optimisation models
  • Applied to Ishinomaki city which was impacted by the Great

Tohoku disasters

  • Comparison between calculation results and reality
  • Applied to problems considering the state of destruction

Appropriate allocation of delivering goods to refuge centres in the

condition of shortage of supply

Can decrease fuel consumption keeping the sufficiency of refuge centres Possible to apply these models in the impacted areas with the fuel

consumption constraint