International Journal of Engineering and Technical Research (IJETR)

ISSN: 2321-0869, Volume-2, Issue-12, December 2014

A study on the effective disaster resources assessment

and allocation for emergency response
Sumin Han, Gyusun Hwang, Sangyun Choe, Jinwoo Park

are four critical activities in the life-cycle of natural disaster

Abstract A disaster is the serious disruption of the management. The life-cycle starts with the mitigation phase,
functioning of a community. Disasters such as flood, fire, and preparedness, and response and ends in recovery activity as
earthquake affect thousands of people annually. Moreover, shown in the following Table 1. Especially, in the response
disasters occurred near the riverside structure cause human,
material, economic, and environmental losses. To deal with the
and recovery stage, it is required to address the importance of
disasters, the disaster response system takes charges of need assessment, coordination, information exchange, and
preparation, response, and recovery. This study seeks to explore logistics management.
the necessity of disaster resource assessment and allocation for
emergency response. We evaluated the assessment of disaster Table 1: Four critical activities in the life-cycle of natural
response resources and allocated them to the damaged areas. disaster management
The simulation study was conducted in order to verify the
Phase Activities
utilization of developed algorithm. This research will be helpful
to decision maker who want to allocate the disaster recovery Structural and non-structural measures
resources effectively. undertaken to limit the adverse impact of
Mitigation
natural hazards, environmental degradation
Index Terms Riverside Structure, Emergency Response, and technological hazards
Resource Assessment, Resource Allocation Activities and measures taken in advance to
ensure effective response to the impact of
I. INTRODUCTION hazards, including the issuance of timely and
In the 2000s, the number of natural disasters is increasing Preparedness
effective early warnings and the temporary
all over the world and the damages of the natural disasters is evacuation of people and properly from
gradually getting serious. According to data from the CRED threatened locations.
International Disasters Database (EM-DAT), the number of The provision of assistance or intervention
natural disasters appears to have increased worldwide. In the during or immediately after a disaster to
decade 1990-1999, natural disasters occurred 2676 times, but meet the life preservation and basic
in the period 2000-2009, the number of occurrences rose to Response
subsistence needs of those people affected. It
4015. Furthermore, the annual estimated economic losses can be of an immediate, short-term, or
from natural disasters are almost US$ 143 billion on average protracted duration.
from 2001 to 2010. The impact of disaster cuts across the Decisions and actions taken after a disaster
economy, the environment, and institutional development. with a view to restoring or improving the
In Korea, damages to a disaster are not negligible levels. pre-disaster living conditions of the stricken
Property damage caused by typhoon is about 140 trillion won Recovery
community, while encouraging and
from 1904 to 2009. In addition, 3 trillion won budget is being facilitating necessary adjustments to reduce
sent for emergency repair for the damaged areas. Like this, disaster risk.
despite the fact that damage occurs repeatedly each year, R &
D budget investment ratio at the national level for the disaster Thus, it is essential to ensure the resources required for
are staying in the 1% level. Therefore, a state of disaster disaster recovery and to efficiently allocate the resources to
response system is incomplete and the new system is indeed the damaged areas to reduce further damage to spread.
necessary. Damages caused by natural disasters are usually the cases
The purpose of the system for disaster is to reduce the that occurs simultaneously in several areas. At this time, the
damage caused by the disaster. It will be the most critical for number of resources that can be used for recovery and repair
the proper planning and execution of disaster response and in the field is limited. Therefore, for allocating each resource
recovery. According to Moe and Pathranarakul (2006), there to the recovery site to meet the needs of the resource, recovery
is required to be able to be smoothly carried out.
Manuscript received December 22, 2014.
Sumin Han, Industrial Engineering, Seoul National University, Seoul,
II. THE PURPOSE OF STUDY
Korea Republic,
Gyusun Hwang, Industrial Engineering, Seoul National University, In this paper, we focused on this appropriate response,
Seoul, Korea Republic, when emergency recovery is needed. In order to prevent
Sangyun Choe, Industrial Engineering, Seoul National University,
Seoul, Korea Republic, additional damage from spreading, appropriate emergency
Jinwoo Park, Industrial Engineering, Seoul National University, Seoul, recovery has to be conducted. We introduce an effective
Korea Republic, methodology and algorithm for resource allocation in disaster

221 www.erpublication.org
 A study on the effective disaster resources assessment and allocation for emergency response

recovery. configured, the optimal assignment proposal is not provided.

In addition to these, there are more researches on resource

allocation for natural disasters but researches through the
evaluation of the resources were not enough.
In this paper, we not only took into account the state of the
resource and environmental factors but also evaluated the
resources with quantitative method. As the final outcome, we
presented optimized alternative of recovery resource
allocation for the recovery required for each damaged areas.

IV. RESOURCE ASSESSMENT

Figure 1: The process of methodology for resource
allocation in disaster recovery A. Characteristics of assessment
Assessment is essential to determine of qualitative or
The process of methodology is in the following Figure 1. quantitative value of the project success related to concrete
First, candidate recovery resources that can be used at the site situation. There are many requirements to project assessment
of recovery are derived from the basic data and then, the regarding some point of views. Based on purpose of
quantitative evaluation is made for each candidate recovery assessment, we classify the assessment into 2 groups:
resource considering factors which the resources have such as operational and effective, because different evaluation
distances, construction ability, etc. With results of the approach is need depending on the purpose of assessment. We
evaluation, an algorithm is conducted to appropriately assign also illustrate the characteristic of assessment based on 4
the resources at each repair site. As a result, optimized criteria.
alternative of recovery resource allocation is presented for the The reason this study try to assess the disaster recovery
recovery required for each damaged areas. resources are to allot the resources more effectively. From the
This study may be useful in the management and utilization below Table 2, our study is for operational purpose and
of resources in the recovery plan. In this paper, we proposed a middle phase assessment. Our goal is to manage the disaster
methodology and algorithm to allocate recovery resources recovery project and improve the efficiency considering the
appropriately according to the status of the recovery site, and status of the current circumstance. In this reason, we should
showed the results through experiment. The result presented evaluate the resources by operational scope and in the middle
in this paper seems to help to create the resource management of the decision status. Also, demanding on the rapid
and resource use plan. responsiveness, our assessment subject is internal due to time
limitation.
III. LITERATURE REVIEW After drawing the demand of the resources, available
There are many studies on resource allocation for natural resources are listed into the database. This study evaluates the
disaster. But, studies on the resource allocation method with resources ability by using our developed formulation. And
an evaluation of the recovery of resources are not enough. outcome of the assessment will be used in resource allotment
Most of the researches are on routing planning of vehicles and as the baseline data.
overall disaster management system.
F. Fiedrich et al. (2000) introduce dynamic optimization B. Elements of the assessment
model. The model uses detailed descriptions of the
For conducting the assessment, we generate the elements
operational areas and of the available resources to calculate
based on the disaster recovery resources characteristics. And
the resource performance and efficiency for different tasks
the extracted elements are as follows: Resource, Moving
related to the response with simulated Annealing (SA) and
distance, Usage period, Construction Capability, Holding
Tabu Search (TS) method.
volume. Outcome of assessments are used in the allotment of
Russell Kondaveti and Aura Ganz (2009) introduce a
the resources. For this reason, we have to quantify the
decision support framework and system built on rapid
elements.
information collection and resource tracking functionalities.
The decision support system integrated with real-time
Moving distance: Distance from current position
emergency response information collection system
to devastated area
developed. Based on this information collected from
Usage period: Current date Manufactured date
emergency response service agencies, operation research
Construction capability:
techniques are used to find an optimal solution for resource
deployment and dispatching. Nevertheless, it does not (Ordinarycapability- Minimum capability )
consider the state of the resource, and does not take into (Maximum capability- Minimum capability )
account a number of environmental factors. Holding volume: Company holding volume/
Lee, Seul-Bi et al. (2013) is to consider the factors for Demand volume
priority among facilities and present framework of restoration
resource allocation model. The framework only but has been

222 www.erpublication.org
 International Journal of Engineering and Technical Research (IJETR)
ISSN: 2321-0869, Volume-2, Issue-12, December 2014

Table 2: Characteristics of assessment based on criteria

Criteria Classification Characteristic

- It evaluates the project management and its stakeholders ability.

Operational
Purpose - It focuses on the project suitability and efficiency.

Effective - It diagnosis the gap between project goal and project outcome.

Prior - It analyses the project plans validity.

- It evaluates on-going project system improvement based on the operational

Phase Middle
assessment.

Posterior - It assesses the project outcome and effectiveness.

- It serves the recheck step for project goals and plans so as to take some
project quality improvements.
Internal
- After project completion, it is hard to take an objectivity and reliability, due
Subject to absence of external view point.

- It is able to take dependency, objectivity, and specialty of project

External
assessment.

Economical - It is able to find a project economical effectiveness.

- It assesses the project political effectiveness, and it has tendency to be

Contents Administrative
non-qualitative assessment.

Technical - It evaluates the project technical problem and it is used in prior assessment.

C. Weighted value setting

Table 3: Example of the weighted value using Rank sum
Elements of the assessment are assigned to weighted
value so as to consider their priority. Weighted values are Moving Usage Capabil Holding
calculated based on the questionnaires to field experts. distance period ity volume
There are many methods to set weighted values: Rank sum, Priority 2 4 3 1
Rank inverse sum, Rank multiplication, etc. This study
Weighted
applies the Rank sum method basically. 0.13 0.27 0.20 0.07
value
Rank sum method is based on the relative importance
idea. The higher rank factor is given to higher weighted D. Conducting resource assessment
value, and the lower rank factor is assigned to lower For evaluating the resources, we generate formulas with
weighted value. The formulation of the rank sum method is weighted value and desired values. First, each elements are
as follows. multiple by weighted value. Then, Assessments are deduced
n : Number of factors by summing up the result. The formulas are as follows:
ri : Rank of factor i among factors n

w i : Final weighted score of factor i

i
* MD j * UP j * CaP i * HV i
1

n - ri 1 Since the disaster may affect damage to a number of sites,

w i= n we will evaluate the resources according to each of the
(n ri 1)
i1
affected area. The outcome of assessment is used in allotment
algorithms. And following table shows the example of the
assessment results.
Table 4: Example of the assessment result
Moving Usage Holding
Capability Assessment
Distance Period Volume
A B C Common Common Common A B C
Fork crane 1 25 14 40 4 50 0.3 14.42 12.95 16.42
Fork crane 2 30 12 5 2 20 0.3 8.55 6.15 5.22
Fork crane 3 19 4 32 8 80 0.1 20.67 18.67 22.41
Fork crane 4 19 4 32 8 80 0.7 20.71 18.71 22.45

223 www.erpublication.org
 A study on the effective disaster resources assessment and allocation for emergency response

V. RESOURCE ALLOCATION 3) Objective function

- Maximize the summation of assessment.
A. Characteristics of allocation problem

After resource assessment, disaster response resources has

to be deployed to damaged structure. Since there are a lot of
factors which are related to disaster scene, it is very complex 4) Decision Variable
problem. - Using amount of resource r in disaster scene s at ith plan:

In the allocation problem, this study tries to build the good

resource allocation plan. The allocation plan contains the - Whether disaster scene s is allocated in ith plan:
destination, amount, and allocation timing of every disaster
response resources. The allocation problem uses the result of
resource assessment and disaster scene assessment. The
disaster scene assessment contains the resource demand and
priority of each disaster scenes. And that is conducted by 5) Constraints
construction and administration specialist. There is the - Current amount constraint:
detailed input and outputs in Table 5.

Table 5: Input and Output of Resource Allocation

Category - Number of allocation constraint:

resource assessment
disaster scene assessment
Input
amount of available resource
resource demand of disaster scene

Destination of resource
Output Amount of usage - Resource usage constraint:
Allocation timing.

B. Formulation
C. Heuristic Approach
Before the solution, detailed formulation is needed for its
precise problem definition. In this study, this problem is If the size of the problem increases, it may take long time to
formulated by mixed integer problem. compute optimal solution with mathematical programming.
So heuristic approach is suggested in this study. The heuristic
1) Notation rule follows these steps.
- Set of resource category: G / g = 1 ~ |G| 1. Initiate
- Set of resource: R / r = 1 ~ |R| A. Set time 1
- Set of damaged area: S / s = 1 ~ |S| 2. Select the disaster scene
- Time horizon: I / i = 1 ~ |S| 3. Select resources for result in step 2.
A. If we have enough resources, allocate them.
2) Data
B. If not, check the disaster scene is not available.
- Category information of each resource:
4. Do all disaster scenes are allocated?
A. If yes, finish.
5. Is there available and not allocated disaster scene?
- Demand information: A. If yes, go back to step 2.
B. If no, increase time and go back to step 2.
- Current amount of resource:
Suggested heuristic algorithms consist of disaster scene
selection and resource selection. In disaster scene selection,
- Result of resource assessment:
there are four heuristic rules are applied.
1. Select the disaster scene with maximum assessment
- Result of disaster scene assessment: score.
2. Pairing all disaster scene and resource with sum of its
assessment score. Select the disaster scene with
maximum number of the pair.

224 www.erpublication.org
 International Journal of Engineering and Technical Research (IJETR)
ISSN: 2321-0869, Volume-2, Issue-12, December 2014

3. Pairing all disaster scene and resource with sum of its

assessment score. Select the disaster scene with Table 9: Result with 12 disaster scenes
maximum summation of the assessment score. Disaster
4. Select the disaster scene with maximum gap between scene
1 2 3 4 total
most assessment and 2nd most assessment score.
Resource
In resource selection, there are two rules are applied.
1. Select the resource with maximum assessment score. 1 3075.97 3030.46 3071.57 3131.33 3077.33
2 2867.42 2826.59 2861.91 2909.89 2866.45
2. Select the resource with maximum gap between most
total 2971.70 2928.52 2966.74 3020.61 2971.89
assessment and 2nd most assessment score.

Table 10: Result with 16 disaster scenes

D. Simulation Experiment Disaster
scene
1 2 3 4 total
Eight heuristic rules, which are combination of rules in two
Resource
stage were applied to simulation experiment to compare their
results. 1 4004.21 3936.05 3985.92 4068.85 3998.76
2 3848.93 3774.40 3828.94 3915.14 3841.85
total 3926.57 3855.23 3907.43 3992.00 3920.30
1) Simulation Setting
Table 6 is the setting of simulation experiment.

Table 6: Simulation Setting

Item Contents
Number of experiment 100
# of resource category 4
# of damaged disaster scene 4,8,12,16
Resource demand Unif[0,10]
Resource assessment score Unif[0,100]
Disaster scene assessment score Unif[0,100]
# of resources per category Unif[1,5]
Stocks of each resources Unif[5,10]

2) Result
Table 7, 8, 9, 10 is the result of simulation experiment
according to the number of disaster scene. Figure 2, 3, 4 is the
result of simulation result which summarize the results of all
disaster scene settings.
The combination of rule 4 and rule 1 has best performance.
Since the gaps between disaster scene selection rules are not Figure 2: Performance of all combination of rules
significant, other rules may do better than rule 4 in certain
situations. But the gap between resource selection rules is
significant.

Table 7: Result with 4 disaster scenes

Disaster
scene
1 2 3 4 total
Resource

1 1107.92 1106.18 1108.76 1121.92 1111.20

2 828.09 828.31 826.78 830.88 828.51
total 968.02 967.24 967.77 976.40 969.86

Table 8: Result with 8 disaster scenes

Disaster
scene
1 2 3 4 total
Resource
1 2107.60 2084.34 2102.64 2150.27 2111.22
2 1861.87 1849.58 1859.51 1894.66 1866.40
total 1984.74 1966.96 1981.07 2022.46 1988.81
Figure 3: Performance of disaster scene selection rules

225 www.erpublication.org
 A study on the effective disaster resources assessment and allocation for emergency response

[6] S.B. Lee, M.S. Park and H.S. Lee, Research on the Priority of
Resource Allocation for Disaster Restoration Planning of Facilities,
Korea Institute of Construction Engineering and Management, vol. 33
no. 1, Apr. 2013, pp. 555-556.
[7] H.D. Sherali, J. Desai and T.S. Glickman, Allocating emergency
response resources to minimize risk under equity considerations,
American Journal of Mathematical and Management Sciences, vol.
24, 2004, pp.367-410.
[8] W. Orabi, A. Senouci, K. El-Rayes, and H. Al-Derham, Optimizing
Resource Utilization during the Recovery of Civil Infrastructure
Systems, J. Manage. Eng., vol. 26 no 4, Oct. 2010, 237246.

Sumin Han was born in Seoul, Republic of Korea on January, 1 st, 1985.
He got bachelors degree in Industrial Engineering from Seoul National
University, Seoul on 2010. At 2010, He entered the graduated school of
Seoul National University. His research interest is in manufacturing system,
software development process, and emergency logistics.
. Now, He is in Ph. d. course of Seoul National University. He participates
in the various research projects related to manufacturing system and
emergency response systems.
Figure 4: Performance of resource selection rules
Gyusun Hwang was born in Kwangju, Republic of Korea on 29th
October 1986. He received his bachelors degree in Management
Information System from HanYang University, Seoul on 2011. At 2011, He
VI. CONCLUSION entered the graduated school of Seoul National University. And He received
Masters degree in Industrial Engineering from Seoul National University in
This study suggested the disaster resource assessment and 2013. Now, He is in Ph. D. course of Seoul National University. His research
allocation algorithm. The suggested assessment algorithm interest is in manufacturing system, Management Information System, and
considered various attributes related to resource itself and its performance measurement.
environment to build the assessment score. Sangyun Choe was born in Changwon, Republic of Korea on May, 22 th,
With the result of assessment, the allocation algorithm 1989. He received his bachelors degree in Industrial and Information
builds the resource allocation plan which contains the systems engineering from Seoul National University of science and
destination, using amount, and timing. To build resource technology, Seoul on 2013. At 2013, he entered the graduated school of
Seoul National University. He is currently a masters candidate and he has
allocation plan, this study suggested heuristic algorithm. And studied for his doctors degree in Industrial Engineering at Seoul National
simulation experiment was conducted to verify the their University. His current research interests are in manufacturing system,
performance simulation, and photovoltaic, renewable energy.
Since the algorithms of this study can build the allocation
Jinwoo Park was born in Seoul, Korea on December, 18 th, 1952. In
plan with relatively small computation resources, the disaster 1985, he received a Ph.D. degree in Industrial Engineering at University of
response system can apply them in disaster response. California, Berkeley and now he is a professor of Industrial Engineering at
Seoul National University. His current research interests are in
manufacturing systems engineering, simulation, scheduling, Enterprise
Resource Planning/Supply Chain Management and Flexible Manufacturing
ACKNOWLEDGMENT System/CIM. He is a member of Institute of Industrial Engineers (IIE),
Society of Manufacturing Engineers (SME), Association of Computing
We would like to acknowledge the support by the ASRI Machinery (ACM), Institute of Electrical & Electronic Engineers (IEEE),
(Automation and Systems Research Institute) in Seoul and American Production & Inventory Control Society (APICS).
National University in the form of resources and
administrative support.
We also would like to acknowledge the support by Ministry
of Land, Infrastructure, and Transport and Korea Agency for
Infrastructure Technology Advancement
(13ConstructionTechnologyS01).

REFERENCES
[1] D. Guha-Sapir, Ph. Hoyois and R. Below, Annual Disaster Statistical
Review 2013, Centre for Research on the Epidemiology of Disasters
(CRED), 2014.
[2] B. Kusumasari, Q. Alam and K. Siddiqui, Resource capability for
local government in managing disaster, Disaster Prevention and
Management, vol. 19 no. 4, 2010, pp. 438-451.
[3] TL. Moe and P. Pathranarakul, An integrated approach to natural
disaster management, Disaster Prevention and Management, vol. 15
no. 3, 2006, pp. 396-413.
[4] F. Fiedrich and F. Gehbauer, U. Rickers, Optimized resource
allocation for emergency response after earthquake disasters, Safety
Science, vol. 35, Jun. 2000, pp. 4157.
[5] R. Kondaveti and A. Ganz, Decision support system for resource
allocation in disaster management, Engineering in Medicine and
Biology Society (EMBC 2009), Annual International Conference of
the IEEE, Sep. 2009, pp. 3425-3428.

226 www.erpublication.org