GOOGLE MAPS FOR CROWDSOURCED EMERGENCY ROUTING 
Simeon Nedkov * *, Sisi Zlatanova* 
* GISt, OTB Research Institute for the Built Environment, Delft University of Technology, Delft, The Netherlands — 
s.b.nedkov@student.tudelft.nl, s.Zlatanova@tudelft.nl 
Commission IV, WG IV/5 
KEY WORDS: Internet/Web, Disaster, Infrastructure, Observations, Mobile 
ABSTRACT: 
Gathering infrastructure data in emergency situations is challenging. The affected by a disaster areas are often large and the needed 
observations numerous. Spaceborne remote sensing techniques cover large areas but they are of limited use as their field of view may 
be blocked by clouds, smoke, buildings, highways, etc. Remote sensing products furthermore require specialists to collect and 
analyze the data. This contrasts the nature of the damage detection problem: almost everyone is capable of observing whether a street 
is usable or not. The crowd is fit for solving these challenges as its members are numerous, they are willing to help and are often in 
the vicinity of the disaster thereby forming a highly dispersed sensor network. 
This paper proposes and implements a small WebGIS application for performing shortest path calculations based on crowds ourced 
information about the infrastructure health. The application is built on top of Google Maps and uses its routing service to calculate 
the shortest distance between two locations. Impassable areas are indicated on a map by people performing in-situ observations on a 
mobile device, and by users on a desktop machine who consult a multitude of information sources. 
1. INTRODUCTION 
Effective management of a disaster depends on knowledge about 
the health and condition of the infrastructure. An important 
analysis in the response phase of a disaster management 
operation is the calculation of shortest routes between two 
locations. Routing, although a simple task on itself, becomes 
difficult during disaster relief operations where a lack of 
information is the norm. Collecting information about the state 
of the infrastructure is a complex task due to the extent of the 
area and quantity of needed observations. Fortunately, making 
these observations does not require much specialization as 
almost everyone can judge whether a street is passable. At the 
same time numerous people on the ground are not able to 
directly aid the relief operation due to lack of equipment or 
knowledge, but nonetheless are willing to help. It comes 
therefore to mind to leverage the knowledge and large numbers 
of the crowd in collecting information about the state of the 
infrastructure and more. 
Crowdsourcing and web mapping are becoming increasingly 
common. Crowdsourcing has shown its strengths in endeavours 
such as Wikipedia. Web mapping platforms such as Google 
Maps and Bing Maps have revolutionized cartography and 
have brought it to the masses. New web technologies have made 
creating dynamic and intelligent websites easier. The combining 
of crowdsourcing and web mapping technologies have produced 
OpenStreetMap. The field of Disaster Management has also 
benefited from this combination in the form of Ushahidi , a 
mapping web application used in disaster relief operations to 
map hazards and people's needs in disaster stricken areas. Using 
Ushahidi the "crowd" is able to collect, store and share 
information about events and points of interest in the disaster 
  
* Corresponding author. 
area. They can identify blocked roads, shelter locations, people 
in need of immediate medical attention, etc. Rescue workers use 
this information to quickly get an overview of the needed aid 
and plan their operations around that information. However, 
Ushahidi does not support geospatial analyses. More 
specifically, it is not possible to perform shortest route 
calculations. 
This paper presents our investigations on crowdsourcing for 
disaster management purposes. What is crowdsourcing? What 
constitutes crowdsourcing emergence? The found constituents 
are implemented in a small WebGIS application that uses the 
"knowledge of the crowd" to aid the routing challenges rescue 
workers face during the response and recovery phases of 
disaster management. The goal of proposed application is 
twofold: 1) automate the way finding process in a disaster 
stricken area by 2) enabling volunteers to act as sensors and 
report on the infrastructure health. 
The paper is organized as follows: section 2 discusses what 
crowdsourcing is and what stimulates its emergence and growth. 
The section continues by discussing what a WebGIS is and what 
functionality is expected from it. Section Error! Reference 
source not found. outlines the used technologies. Section 
Error! Reference source not found. show highligths the 
methodology and obtained results. Section 5 presents some 
conclusions while section 6 identifies possible research 
directions. 
2. CROWDSOURCED DIS AS TER MANAGEMENT 
The sudden change of urban infrastructure configuration and 
health immediately after a disastrous event renders much of 
446 
existi 
At tl 
asa 
and t 
natur 
funct; 
infras 
inforr 
the lai 
Airbo 
a solt 
of ren 
to ma 
becau: 
the gr 
consid 
persor 
netwo 
infrast 
routing 
Two t 
the tas 
people 
the crc 
essence 
networ 
that th 
Citizen 
of gro 
tools. ] 
local m 
inform: 
familiar 
21 In 
Disaste 
organiz; 
governn 
instituti 
Kodrich 
technolc 
commur 
consists 
organisa 
financial 
The sec 
immedia 
Organiza 
they are 
tier is e 
that wis 
operatioi 
media to