GIS-based Evacuation Routing using Capacity Aware Shortest Path

Evacuation Routing Algorithm and Analytic Hierarchy Process

for Flood Prone Communities

Cinmayii Manliguez

1,2

, Zarah Jean Diche

, Maria Jezebel Jimenez

Maureen Agrazamendez

1,2

and Joseph Acosta

1,2

Department of Mathematics, Physics, and Computer Science, College of Science and Mathematics,

University of the Philippines Mindanao, Davao City, 8022, Philippines

Phil-LiDAR 1.B.13 LiDAR Data Processing and Validation in Mindanao: Davao Region,

College of Science and Mathematics, University of the Philippines Mindanao, Davao City, 8022, Philippines

Keywords: Decision Theory, Flood, Uncertainty, Evacuation Routing.

Abstract: Evacuation routing is one of the fundamental instruments for flood risk mitigation. In this study, features

extracted from LiDAR data are used to create dynamic network composed of buildings and roads. Flood prone

areas identified through flood models from Phil-Lidar 1 Project are considered by Capacity Aware Shortest

Path Evacuation Routing algorithm to determine optimal routes. Road capacity and location of building

features were also considered. Uncertainties among possible paths taken are evaluated through Decision

Theory. Specific DT technique implemented to generate alternative routes for the possibility of detours is the

Analytic Hierarchy Process. This study can help city governance in terms of planning and disaster risk

reduction management.

1 INTRODUCTION

The Philippines experiences an average of 20

typhoons every year making it as one of the most

flood-prone countries in the world (Ortega, 2014).

One of the most destructive typhoons to have ever

been recorded in history, Super Typhoon Yolanda

(Haiyan) entered the Philippine archipelago on 08

November 2013 leaving 6,300 people dead, 28,688

injured, and 1,062 missing (NDRRMC, 2014).

Despite the early warnings issued by the Philippine

Atmospheric Geophysical Astronomical Services

Administration (PAGASA) to the general public,

these were not translated into appropriate actions in

every coastal village in the Central Philippine Region

(Lagmay et al., 2015), which resulted to the loss of

lives by the thousands. This can be prevented if the

warnings were also accompanied by appropriate

information to the people concerning with what to do

in case the signal is raised. Evacuation planning is one

fundamental instrument for risk mitigation that is

taken into focus in this study.

Routing is part of the evacuation planning process

that determines the best routes to relocate the affected

population to the nearest shelters—usually, within the

shortest amount of time possible for an individual.

However, it is not enough for an evacuation routing

problem to just generate the best routes for each

evacuee, it should also be able to adjust to the

dynamic changes that might occur within the road

network while the evacuation process is ongoing.

These dynamic changes can affect the road network

and consequently, the evacuation process. This

uncertainty can be dealt best with decision theory

(DT).

In many real world situations like evacuation

routing, decision can be settled through various

mathematical tools. Decisions based on mathematical

reckoning present a quantified proof on why such

decision was made with the assurance that

subjectivity, bias and rationality of decision maker

and the decision making process itself is not

magnified. DT is used to select the optimal decision

given a set of alternatives to choose from. It also

covers the concepts of negativity and selection of the

worst among choices which can be taken into

consideration in routing process especially if a sudden

change in the road network occurs.

The general aim of this study is to implement a

Manliguez, C., Diche, Z., Jimenez, M., Agrazamendez, M. and Acosta, J.

GIS-based Evacuation Routing using Capacity Aware Shortest Path Evacuation Routing Algorithm and Analytic Hierarchy Process for Flood Prone Communities.

DOI: 10.5220/0006327402370243

In Proceedings of the 3rd International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2017), pages 237-243

ISBN: 978-989-758-252-3

237

decision theory technique to cope with the

uncertainties in the dynamic road network.

Specifically, the study aims to:

1. Introduce the Analytic Hierarchy Process to

the Capacity Aware Shortest Path Evacuation

Routing (CASPER) Algorithm to allow the

algorithm to adjust to the dynamic changes

within the road network; and

2. To apply the proposed CASPER-AHP

algorithm to a dynamic network, composed

of buildings and roads, which was created

using LiDAR data and the ArcGIS Network

Analyst Tool.

As previously mentioned, changes in the road

network are likely to happen at the time of a disaster.

Therefore, it is entirely possible that these changes

could affect some evacuation routes where no other

feasible routes can be achieved at that point,

indicating a dead end and the possibility of a

bottleneck occurring on the affected routes, leaving

the population extremely vulnerable when the

disaster strikes. The proposed algorithm can identify

such bottlenecks. In addition, if a change in the road

network occurs such that the affected roads are those

roads that connect to the destination point, instead of

re-routing the evacuees, this information can aid the

city officials by finding an alternative destination

point for the evacuees in order to minimize

evacuation time. Also, the generation of the optimal

evacuation routes can help in determining which

roads should be made available for the evacuees by

the time the disaster occurs.

This study only concerns the generation of the

optimal evacuation routes for the population residing

near the Lipadas River in Davao City. Their

geographical location, being close to the river

channels, leave them highly susceptible to flood

hazards. The residential areas close to the Talomo

River are considered in this study as a means to test

the effectiveness and the scalability of the proposed

solution.

2 BACKGROUND AND RELATED

LITERATURE

In the face of a natural disaster, evacuation planning

refers to the process of relocating the endangered

population to safer areas as soon as possible. The

study of Shahabi (2015) proposed a new taxonomy to

the evacuation planning problem in which the

problem is divided into three stages: preparation,

evacuation execution, and post-disaster response.

The evacuation execution stage refers to those

actions that are taken immediately, once the nature of

the disaster becomes known, to reallocate the

potentially endangered population to safety. At this

stage the location and time of the disaster is the key

information (Shahabi, 2015). The post-disaster

response stage is characterized by emergency

personnel caring for the affected population, shelter

maintenance, and disaster assessment. For example,

Eguchi et al. (2008) studied post-disaster damage

assessment using integrated GPS sensor network and

GIS. Lue et al. (2014) studied disaster damage

assessment by utilizing geo-tagged videos taken from

the affected areas.

Disasters can either be static or dynamic. Dynamic

disasters are characterized by their changing behavior,

location, or severity. Disasters such as tsunamis and

terrorist attacks are considered as static whereas

hurricanes, wildfires, and flood are dynamic (Shahabi,

2015). These dynamic disasters would evidently add

to the complexity in evacuation planning and has to

be considered in a realistic evacuation solution.

The evacuation routing problem refers to the

process of relocating the potentially affected

population towards safe destination points such as

shelters and hotels. The objective of this problem is

usually to minimize exposure, global evacuation time,

average travel time, or traffic congestion. An

effective routing solution should consider the

characteristics of the transportation network,

available transportation vehicles, as well as the

capacity of the potential destinations (Shahabi, 2015).

2.1 Flood Hazard Maps

Flood hazard maps are designed to identify the areas

that are at a risk of flooding and to increase the

awareness of the likelihood of flooding among the

public. They also encourage the population residing

in flood-prone areas to find out more about the local

flood risk and to take the appropriate action (Linham

and Nicholls, 2010).

To benefit from the flood hazard mapping, it is

important to provide the local community residing in

the flood hazard zone with the appropriate

information about emergency procedures and ways of

reducing the risk of flood, otherwise, presenting the

flood hazard maps may only serve to increase the fear

and anxiety of the residents.

2.2 Application of GIS Techniques

A geographical information system (GIS) is a

computer system that is designed to support the

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

238

capture, management, manipulation, analysis, and

modeling and display of spatially-referenced data

suitable for solving complex planning and

management problems (Cole et al., 2005).

All stages of the evacuation planning problem

identified by Shahabi (2015) can greatly benefit from

the application of GIS. Gaining access to the

appropriate data is the key. In an emergency, it is

critical to have the right data, at the right time,

displayed logically, to respond and take the

appropriate action. By utilizing GIS, this data can be

shared throughout different agencies or departments

with the use of spatial databases held in one central

location. GIS provides a mechanism to centralize and

visually display critical information in the midst of an

emergency (Cole et al., 2005).

With regards to preparation, GIS can be used to

provide answers to particular questions such as

identifying the safest location for the critical facilities,

selecting of evacuation routes based on the

anticipated or actual flood, or determining whether

the transportation network can handle the sudden

increase in traffic flow.

Lastly, in the post-disaster response stage, GIS

can play a role in the disaster damage assessment and

information management. With the use of GPS and

telecommunication devices, assessments of the

damages can be geo-referenced and transmitted back

to the emergency headquarters for real-time update of

the recovery (Cole et al., 2005).

2.3 Evacuation Routing Methods

The existing evacuation routing methods can be

divided into the following classifications: simulation,

network flow, and heuristic methods (Shekhar et al.,

2012).

2.3.1 Simulation Methods

Simulation methods are solutions that visually

simulate an emergency situation. It tries to visualize

what could possibly happen as realistically as

possible. Flow-based modeling, agent-based

modeling, and cellular automaton modeling are just

some of the methods that would fall into this category

(Santos and Aguirre, 2004). These tend to focus more

on the individual evacuees' movements and their

interaction with one another (Mahmassani et al.,

2004).

2.3.2 Network Flow Methods

Many research works have been done to model the

evacuation problem as a network flow problem and to

find the optimal solution using the linear

programming (LP) methods. Hamacher and Tjandra

(2002) gave an extensive literature review on the

models and algorithms used in these linear

programming methods. It initially models the

evacuation network into a network graph (denoted by

G), then it requires the user to enter an estimated

upper bound T of the evacuation egress time. Second,

it converts the network graph G into a time-expanded

network (denoted by GT), by duplicating the

evacuation network G for each discrete time unit t =

0, 1, 2, ..., T. Then, it defines the evacuation problem

as a minimum cost network flow problem (Ford and

Fulkerson, 1962) on GT. Lastly, it feeds the GT to the

minimum cost network flow solvers, such as

NETFLO (Kennington and Helgason, 1980), to

obtain the optimal solution.

2.3.3 Heuristic Methods

Unfortunately, in a real-world urban evacuation

scenario, the evacuation demand can easily

overwhelm the capacity of the evacuation routes,

resulting to traffic congestion (Bish et al., 2013).

Congestion is not only inconvenient, but can also

cause potentially dangerous situations because it

discourages evacuation from potentially affected

areas and it can leave the evacuees extremely

vulnerable if they are trapped in the affected areas and

it can leave the evacuees extremely vulnerable if they

are trapped in the affected roadways when the disaster

strikes. Furthermore, congestion can make the entire

evacuation process itself hazardous (Bish et al., 2013).

One of the limitations of CCRP is that it assumes that

the maximum capacity of an edge does not depend on

the traffic flow amount on the edge (Lu et al., 2005).

In other words, it does not consider the traffic

congestion realistically.

2.3.4 Casper Algorithm

The Capacity Aware Shortest Path Evacuation

Routing (CASPER) algorithm is a heuristic

evacuation routing method that connects each source

node (evacuee) to its nearest destination while taking

into account the capacity of the transportation

network and the traffic flow in order to minimize

traffic congestion and system-wide transportation

times (Shahabi and Wilson, 2014). The algorithm first

sorts the evacuees based on their distance from the

closest destination area. Then, starting from the

evacuee with the longest distance, it finds the shortest

path and assigns the evacuee to that path. It iteratively

continues this process until there are no more

evacuees left, indicating that the affected population

GIS-based Evacuation Routing using Capacity Aware Shortest Path Evacuation Routing Algorithm and Analytic Hierarchy Process for

Flood Prone Communities

239

has successfully been removed from the hazard area.

During the analysis, CASPER dynamically updates

the edge travel costs based on the number of assigned

evacuees and the capacity of the edge (Shahabi, 2012).

Each source point s is metered (interval(s)) so it

will generate a different flow on each edge. For

example, evacuees leaving from a source point at 20s

intervals have interval(s) = 20. Each source point also

has only one path P

assigned to them. A path P

is an

ordered set of edges that will guide all the population

from source point s to safety (t). From here, the total

flow on edge e can be calculated by summing up all

flows of all paths that pass through e (Shahabi and

Wilson, 2014).

Traffic Model. The traffic model is defined as a

function with two parameters T(f,c). The traffic model

predicts the congestion on an edge based on its

capacity and total flow. From there, the cost of

traversing an edge, and consequently the cost of

traversing a path, can be calculated (Equation 1 and

2).







































(1)













 















  













(2)

The main objective here is to minimize the cost of

the path with the highest cost.



















































(3)

The calculation in Equation 3 considers both the

previously reserved paths and the new population

flow (i.e. flow(s,e)) so there is a need to record all the

reserved paths. Lastly, the costs of all the paths are re-

calculated. This step is important since the record of

the reserved paths is not complete during the path

finding process and therefore the costs are just a lower

bound. Once all the paths are reserved, their costs

need to be re- calculated to find the most accurate

global evacuation time (Shahabi, 2015).

2.4 Dynamic Evacuation Routing

Problem

The dynamic evacuation routing problem describes

the problem of generating as well as maintaining

evacuation routes in a dynamic environment. In a

realistic evacuation scenario, unpredictable situations

can occur within the environment whilst the

evacuation process is still ongoing. For example,

changes in the road network such as road blockages,

car accidents, and flooded underpass can affect the

road network and consequently the evacuation

process. If a change in the transportation network

occurs, the evacuation routing system should be able

to detect that change, inform those evacuees whose

routes might be affected from the change, and update

these routes in a timely manner (Shahabi, 2015).

2.5 Analytic Hierarchy Process (AHP)

The Analytic Hierarchy Process (AHP) method was

developed by Saaty (2008) and has been widely used

to solve multi-criteria decision making problems.

Some examples of multi-criteria decision making

problems are: choosing a telecommunication system,

choosing a product marketing strategy, etc.

(González-Prida et al., 2012). These decision

problems are decomposed into a hierarchy of criteria

and alternatives.

In AHP, values like price, weight, or time, or

even subjective opinions such as feelings, preference,

or satisfaction, can be translated into measurable

numerical relations. The core of AHP is that it does

comparison of pairs instead of sorting (ranking),

voting (e.g. assigning points), or free assignment of

priorities. Individuals and groups use the AHP

preference scale to formulate the comparison

matrices (Alexander, 2012)).

Saaty (2008) provided a measure of consistency,

called the Consistency Index (CI), as a degree of

consistency. Once the consistency index is calculated,

its value is compared with the appropriate one. The

appropriate consistency index is called the Random

Consistency Index (RI) which was obtained from

associated random matrices of order n to compute the

error due to inconsistency (Nieto et al., 2015). This

comparison of values is called the Consistency Ratio

(CR).

If the value of the Consistency Ratio is smaller

than or equal to 0.10, then the inconsistency is

considered acceptable, otherwise, it must be reviewed

to improve its consistency (González-Prida et al.,

2012).

3 METHODOLOGY

In this study, the Analytic Hierarchy Process will be

introduced to the Capacity Aware Shortest Path

Evacuation Routing (CASPER) algorithm. This

proposed algorithm will be applied to the road

networks along Lipadas River. The results will be

compared to those results obtained from using

previous evacuation methods.

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

240

3.1 Benchmark Data Sets

The flood hazard maps of Lipadas and Talomo will

be used in this study to identify the areas within the

road network that are at a risk of flooding. Road

networks and building features will also be used for

the dynamic network. All data were generated

through the use of LiDAR data obtained from the

Phil-LiDAR 1 Office of the University of the

Philippines Mindanao.

3.2 Road Network

In this study, the ArcGIS Network Analyst Tool will

be used to build the road network. It provides

network-based spatial analysis including point-to-

point routing, travel directions, closest facility, and

service area analysis. By using an advanced network

data model, networks from their geographic

information system (GIS) data are built. It enables

users to dynamically model realistic network

conditions (i.e. turn restrictions, speed limits, traffic

conditions) at different times of the day (ESRI, 2014).

3.3 Source Points and Destination Points

The flood hazard maps of Lipadas and Talomo will

be used in this study to identify those areas within the

road network that are at a risk of flooding. Areas that

are marked red indicate a high flood hazard (>1.5m).

Those that are marked orange indicate a moderate

flood hazard (0.5 – 1.5m). Areas that are marked

yellow indicate a low flood hazard (<0.5m). Figure 1

shows the 100-year return period flood hazard map of

the Talomo area (Project NOAH, 2016).

Figure 1: 100 Year Flood Hazard Map of Talomo (Project

NOAH, 2016).

3.4 Evacuation Framework

The evacuation framework is initially specified which

includes all the discussed requirements. It will also

include some steps needed for the experimental

evaluations. The pseudo-code, as shown in Figure 2,

outlines the overall system.

input: G(V,E), S, t, T(f,c)

P ← GetEvcRoutes(G, S, t, T )

EvcTime ← max{P

| P

 P}

SimTime ← Simulation(G, S, P )

return P, EvcTime, and SimTime

Figure 2: Evacuation Framework.

The system inputs the road network in the form

of a directed graph G. Let set S be the source

population with their intervals interval(s). The

function T denotes the chosen traffic model. The

GetEvcRoutes function processes the inputs to

generate one evacuation route for each source point.

All the paths end at vertex t, which serves as the

destination point for all the source points. The

EvcTime variable serves as the predicted global

evacuation time based on the routes generated for the

source population. Based on the lengths of the

generated paths and the traffic predictions, it takes

EvcTime time to get everyone to the destination point

(safety). SimTime, on the other hand, is the simulated

global evacuation time. This is only included for

evaluation purposes. The Simulate function takes the

graph, source points, and the generated paths as

inputs. Instead of predicting the time, it simulates

every person (vehicle) moving from s to t on the road

map (graph) whilst taking into account the

interactions between evacuees (vehicles). SimTime is

recorded time for the last evacuee to reach his or her

destination. The accuracy of the traffic model is

measured by comparing the two times (Shahabi and

Wilson, 2014).

3.5 CASPER-AHP Algorithm

The CASPER algorithm will be hybrid with the

Analytic Hierarchy Process to allow the evacuation

routing algorithm to adjust to the dynamic changes

that might occur within the road network whilst the

evacuation is still ongoing. The changes to the road

network are not initially known; hence, the algorithm

can only adjust once it learns about them. In this study,

road blocks will be introduced to the road network

during evacuation. The selection of the paths in which

the road blocks occur will be done at random.

The AHP will be embedded into the CASPER

algorithm once a change has been detected in the road

GIS-based Evacuation Routing using Capacity Aware Shortest Path Evacuation Routing Algorithm and Analytic Hierarchy Process for

Flood Prone Communities

241

network. As mentioned earlier, each edge e has a

positive nonnegative impedance (imp) and road

capacity (cap) associated to it. Neither of these values

are constant since the road network can change during

evacuation. In other words, the graph edges are

allowed to change their values at some time after the

evacuation starts.

Once a road block is introduced on the edge, the

edge becomes no longer accessible. When this

happens, the algorithm should be able to backtrack to

the previous edge. The embedded AHP will decide

which alternate path the evacuees should take

(detours). This study will implement a two-level

hierarchy AHP. The first level deals with the factors

to consider in choosing the alternative route for the

evacuees whose original path is affected by the road

block. The second level deals with the alternative

routes.

The steps in formulating the solution to the

decision problem using AHP as summarized by

Goepel (2013) are answered as follows:

1. Define the goal of the decision — the purpose

of this decision is to have the evacuees adjust

to the changes within the road network by

selecting the best alternative route to them.

2. Model the decision problem into a hierarchy

— the following criteria are considered in

deciding the best alternative route for the

evacuees: road capacity, road type, road status

(in terms of flood risk), and the traffic

congestion on that road.

3. Pair comparison of criteria in each category —

In level, there will be one comparison matrix

corresponding to the pair-wise comparisons

between the four criteria with respect to the

goal. Thus, the comparison matrix of level 1

has a size of 4 by 4. Since each choice is

connected to each factor, then there is a total of

four comparison matrices at level 2. The size

of the matrices on that level depends on the

number of choices.

4. Calculate the priorities and consistency index

— Use the Consistency Index (CI) and the

Consistency Ratio (CR) to check for the

consistency of the comparisons. The Random

Consistency Index (RI) value in this case is

0.90.

5. Evaluate alternatives according to the

calculated priorities — Compute the overall

weight of each alternative choice based on

the weight of level 1 and level 2. The overall

weight is the normalization of linear

combination of multiplication between

weight and priority vector.

In the proposed CASPER-AHP algorithm, it is

assumed that the changes in the road network are not

initially known. Therefore, the beginning steps are

similar to that of the CASPER algorithm with a static

problem. It loops over all the evacuee source points.

The algorithm finds the shortest path to the

destination point and is assigned to the source point.

The changes in the road network are known once a

path has already been assigned to the source point.

The changes in the road network are known once a

path has already been assigned to the source point.

The selection of the edge to where the road block is

introduced done at random. As previously mentioned,

once a road block is placed on the edge, that edge

becomes no longer accessible, so it is logical to

remove that edge from the road network. If the

selected edge is part of the reserved path for source

point s, the AHP function is called to allow the

evacuees at that source point to backtrack to the

previous edge (remember that a path P

is defined as

an ordered set of edges that will guide all the

population from s to safety). From there, the AHP

function will decide which alternate path the evacuees

should take from that edge. This allows the algorithm

to adjust to the dynamic changes within the road

network.

4 EXPECTED RESULTS AND

FUTURE DIRECTION

The proposed combined algorithm is expected to be

used during the flood event given that there is already

a flood hazard map downloaded in a specific device.

In the future, this proposed method will be optimized

so that it can be integrated into an application that can

be used on any devices that is portable during flood

event such as smartphones, tablets, etc.

ACKNOWLEDGEMENTS

This study is under the Phil-Lidar 1.B.13 research

project of the University of the Philippines Mindanao

that is funded by the Department of Science and

Technology (DOST) and the Philippine Council of

Industry, Energy and Emerging Technology Research

and Development (PCIEERD) of the Philippines.

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

242

REFERENCES

Alexander, M., 2012. Decision-Making using the Analytic

Hierarchy Process (AHP) and SAS/IML. SESUG-2012.

1-12.

Bish, D., Sherali, H. and Hobeika, A., 2013. Optimal

evacuation planning using staging and routing. Journal

of the Operational Research Society. 65(1), 124-140.

doi:10.1057/jors.2013.3.

Caldara, F. R., Garcia, R. G., Almeida Paz, I. C. L., Nääs, I.

A., Santos, R. K. S. and Foppa, L., 2013. Assessing the

most suitable floor system for growing finishing

piggery under tropical conditions using the analytic

hierarchy process. Italian Journal of Animal Science.

13(1), 136-141. doi: 10.4081/ijas.2014.2830.

Cole, J. W., Sabel, C. E., Blumenthal, E., Finnis, K., Dantas,

A., Barnards, S. and Johnston, D. M., 2005. GIS-based

Emergency and Evacuation Planning for Volcanic

Hazards in New Zealand. Bulletin of the New Zealand

Society for Earthquake Engineering. 38(3), 149-164.

Eguchi, R. T., Huyck, C. K., Ghosh, S. and Adams, B. J.,

2008. The Application of Remote Sensing

Technologies for Disaster Management, in: The 14th

World Conference on Earthquake Engineering, Beijing,

China.

ESRI, 2014. What is GIS? [WWW Document].

Environmental Systems Research Institute. URL

https://www.esri.com/what-is-gis/overview (accessed

10.06.16).

Ford, F. R. and Fulkerson, D. R., 1962. Flows in networks.

Princeton University Press, Princeton, New Jersey.

Georgiadou, P. S., Papazoglu, I. A., Kiranoudis, C. T. and

Markatos, N. C., 2007. Modeling Emergency

Evacuation for Major Hazards and Industrial Sites.

Reliability Engineering and System Safety. 92(10),

1388-1402.

González-Prida, V., Barberá, L., Viveros, P. and Crespo, A.,

2012. Dynamic analytic hierarchy process: AHP

method adapted to a changing environment. IFAC

Proceedings Volumes (IFAC-PapersOnline). 25–29.

doi: 10.3182/20121122-2-ES-4026.00005.

Hamacher, H. W. and Tjandra, S. A., 2002. Mathematical

Modeling of Evacuation Problems: A State of the Art,

in: Pedestrain and Evacuation Dynamics. Springer

Berlin Heidelberg, pp. 227-266.

Hart, P. E., Nilsson, N. and Raphael, B., 1968. A formal

basis for the heuristic determination of minimum cost

paths. IEEE Transactions on Systems Science and

Cybernetics. 4(2), 100–107. doi:

10.1109/TSSC.1968.300136.

Kennington, J. L. and Helgason, R. V., 1980. Algorithms

for Network Programming in: Networks. Wiley and

Sons, New York, 12, 87-88.

doi:10.1002/net.3230120107.

Lagmay, A. M. F., Agaton, R. P., Bahala, M. A. C., Briones,

J. B. L. T., Cabacaba, K. M., Caro, C. V. C., Dasallas,

L. L., Gonzalo, L. A. L., Ladiero, C. N., Lapidez, J. P.,

Mungcal, M. T. F., Puno, J. V. R., Ramos, M. M. A. C.,

Santiago, J., Suarez, J. K. and Tablazon, J. P., 2015.

Devastating storm surges of Typhoon Haiyan.

International Journal of Disaster Risk Reduction. 11, 1-

12. doi: 10.1016/j.ijdrr.2014.10.006.

Linham, M. M. and Nicholls, R. J., 2010. Flood hazard

mapping [WWW Document]. ClimateTechWiki. URL

http://www.climatetechwiki.org/content/flood-hazard-

mapping (accessed 10.01.16).

Lu, Q., George, B. and Shekhar, S., 2005. Capacity

Constrained Routing Algorithms for Evacuation

Planning: A Summary of Results, in: Advances in

Spatial and Temporal Databases. Springer Berlin

Heidelberg, 3633, 291-307. doi:10.1007/ 11535331_17

Lue, E., Wilson, J. P. and Curtis, A., 2014. Conducting

Disaster Damage Assessments with Spatial Video,

Experts, and Citizens. Applied Geography 52. Elsevier

Ltd: 46-54. doi:10.1016/j.apgeog.2014.04.014.

Mahmassani, H. S., Sbayti, H. and Zhou, X., 2004.

DYNASMART-P Version 1.0 User's Guide. Maryland

Transportation Initiative, College Park, Maryland.

NDRRMC. 2014. National Disaster Risk Reduction

Management Council. Philippines. URL

www.ndrrmc.gov.ph (accessed 09.04.16)

Nieto, V. M. O. S., Caldara, F. R., Nääs, I. A., Moi, M.,

Almeida Paz, I. C., Garcia, R. G., Seno, L. O. and

Santos, L. S., 2015. Applying multicriteria analysis for

choosing the best marination for pork. Revista MVZ

Cordoba. 20(1), 4396-4405.

Ortega, A., 2014. Philippines' Country Report 2014 [WWW

Document]. Country Report, Asian Disaster Reduction

Center (ADRC). URL http://www.adrc.asia/

countryreport/PHL/2014/PHL_CR2014B.pdf

(accessed 10.31.16).

Project NOAH, 2016. DOST - Nationwide Operational

Assessment of Hazards. URL http://noah.dost.gov.ph/

(accessed 09.04.16)

Saaty, T. L., 2008. Decision making with the Analytic

Hierarchy Process. International Journal of Services

Sciences. 1(1), 83-98.

Santos, G. and Aguirre, B. E., 2004. A Critical Review of

Emergency Evacuation Simulation Models, in:

Building Occupant Movement During Fire

Emergencies. Disaster Research Center, University of

Delaware.

Shahabi, K., 2012. Out of Harm’s Way: Enabling

Intelligent Location-Based Evacuation Routing

[WWW Document]. ArcUSER. URL

http://www.esri.com/news/ arcuser/0612/out-of-harms-

way.html (accessed 10.02.16).

Shahabi, K. and Wilson, J. P., 2014. CASPER: Intelligent

capacity-aware evacuation routing. Computers,

Environment and Urban Systems. 46, 12-24. doi:

10.1016/ j.compenvurbsys.2014.03.004

Shahabi, K., 2015. Scalable Evacuation Routing in

Dynamic Environments. University of Southern

California.

Zeng, M. and Wang, C., 2009. Evacuation Route Planning

Algorithm: Longer Route Preferential, in: Advances in

Neural Networks – ISNN 2009. Springer Berlin

Heidelberg. 5551, 1062–1071. doi: 10.1007/978-3-642-

01507-6_119.

GIS-based Evacuation Routing using Capacity Aware Shortest Path Evacuation Routing Algorithm and Analytic Hierarchy Process for

Flood Prone Communities

243