AI and Robotics in Disaster Studies T. V.

Vijay
Kumar pdf download

https://textbookfull.com/product/ai-and-robotics-in-disaster-studies-t-v-vijay-kumar/

★★★★★ 4.6/5.0 (37 reviews) ✓ 209 downloads ■ TOP RATED

"Perfect download, no issues at all. Highly recommend!" - Mike D.

DOWNLOAD EBOOK
 AI and Robotics in Disaster Studies T. V. Vijay Kumar pdf
download

TEXTBOOK EBOOK TEXTBOOK FULL

Available Formats

■ PDF eBook Study Guide TextBook

EXCLUSIVE 2025 EDUCATIONAL COLLECTION - LIMITED TIME

INSTANT DOWNLOAD VIEW LIBRARY

Collection Highlights

Disaster Robotics Results from the ImPACT Tough Robotics

Challenge Satoshi Tadokoro

Human Well Being and Policy in South Asia Vijay Kumar

Shrotryia

An Introduction to Ethics in Robotics and AI Christoph

Bartneck

Robotics, AI, And Humanity: Science, Ethics, And Policy

Joachim Von Braun
Customer Relationship Management V. Kumar

Shifting Paradigms in Public Health From Holism to

Individualism 1st Edition Vijay Kumar Yadavendu (Auth.)

Challenges to Internal Security of India ( Includes

Disaster Management) Ashok Kumar

AI and Big Datas Potential for Disruptive Innovation

Advances in Computational Intelligence and Robotics 1st
Edition Moses Strydom

Disaster Management for 2030 Agenda of the SDG V. K.

Malhotra
DISASTER RESEARCH AND MANAGEMENT SERIES
ON THE GLOBAL SOUTH
SERIES EDITOR: AMITA SINGH

AI and Robotics in
Disaster Studies
Edited by
T. V. Vijay Kumar
Keshav Sud
 Disaster Research and Management Series
on the Global South

Series Editor
Amita Singh
Centre for the Study of Law and Governance & Special
Centre for Disaster Research
Jawaharlal Nehru University
New Delhi, India
Disaster Research and Management Series on the Global South is a series
coming out of Special Centre for Disaster Research (SCDR) at Jawaharlal
Nehru University (JNU), Delhi, India. SCDR is the first in Asia Pacific to
start a course on disaster research within a social science perspective. The
series follows and publishes pedagogical and methodological change
within the subject. The new direction of teaching, research and training
turns from ‘hazard based’ to ‘resilience building’. The series taps such
research for the benefit of institutes and higher education bodies of the
global south. It also suggests that much of the western literature based
upon rescue, relief and rehabilitation which is also being taught in the
Asian institutes is not directly relevant to managing disasters in the region.
It provides reading and study material for the developing field of disaster
research and management.
1. Generates a non-west transdisciplinary literature on disaster research
and studies
2. Strengthens disaster governance and improves its legal framework
3. Sensitizes disaster management authorities towards key priorities and
attention areas
4. Focus on preparedness is strongly proposed and revisited
5. Highlights changes in pedagogy and methodology of disaster research
and teaching
6. Mainstream vulnerable communities of differently abled, elderly,
women, children
7. Indicate strategies needed to protect city animals, birds and wildlife
during disasters

More information about this series at

http://www.palgrave.com/gp/series/16402
T. V. Vijay Kumar • Keshav Sud
Editors

AI and Robotics
in Disaster Studies
Editors
T. V. Vijay Kumar Keshav Sud
School of Computer and Systems Data Science
Sciences Keurig Dr Pepper
Jawaharlal Nehru University Burlington, MA, USA
New Delhi, India

ISSN 2662-4176     ISSN 2662-4184 (electronic)

Disaster Research and Management Series on the Global South
ISBN 978-981-15-4290-9    ISBN 978-981-15-4291-6 (eBook)
https://doi.org/10.1007/978-981-15-4291-6

© The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer
Nature Singapore Pte Ltd. 2020
This work is subject to copyright. All rights are solely and exclusively licensed by the
Publisher, whether the whole or part of the material is concerned, specifically the rights of
translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on
microfilms or in any other physical way, and transmission or information storage and retrieval,
electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information
in this book are believed to be true and accurate at the date of publication. Neither the
publisher nor the authors or the editors give a warranty, expressed or implied, with respect to
the material contained herein or for any errors or omissions that may have been made. The
publisher remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

This Palgrave Macmillan imprint is published by the registered company Springer Nature
Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-­01/04 Gateway East, Singapore
189721, Singapore
 Acknowledgements

Our deepest gratitude to none other than our former Minister of State for
Disaster Management Hon’ble Shri Kiren Rijiju who inspired, encouraged
and supported research into new areas of disaster management. His enthu-
siasm, office room discussions, his encouragement to pursue unattended
areas in disaster research and his unending ambition to make India the
best in the world in implementing the Sendai Framework pushed us to
deeply focused research. He is now adorning the sports ministry but his
contribution to the setting up of India’s first completely research-based
disaster research centre at Jawaharlal Nehru University (JNU) has been
the most illuminating feather in his cap. We hope that his contribution
reaches its desired destination from a sapling to a full grown tree.
Some officials from administrative services have joined hands with us in
our endeavour like the former Executive Director of the National Institute
of Disaster Management (NIDM) Mr. Anil Kumar and the current
Executive Director Major Gen. Manoj Kumar Bindal for enabling a whole-
some participation, providing an atmosphere of academic freedom and
enthusing empirical research in the NIDM faculty. The faculty of the
Special Centre for Disaster Research (SCDR) at JNU and the NIDM
could collectively encounter challenges at the field and contribute to a
variety of curriculum-building events at JNU due to the valuable support
coming from the former Minister of State for Home Affairs Shri Kiren
Rijiju. Much research for the book has actually taken place during his ten-
ure as the Minister of State for Home Affairs. It is also worth mentioning
here that on 8 August 2015 Shri Rijiju Ji was instrumental in getting

v
vi ACKNOWLEDGEMENTS

NIDM and JNU sign an MoU for collaboration on disaster research and
three months later on his birthday, 19 November, he was designated by
the UN Office for Disaster Risk Reduction (UNISDR) as “Disaster Risk
Reduction Champion for the Asia Region”. The JNU Research Team on
Disaster Research recollects the attention, focus and clarity with which he
corrected research teams of seven professors with respect in his office,
sometimes a virtual classroom.
SCDR is equally appreciative of Lt. Gen. N. C. Marwah, member,
National Disaster Management Authority (NDMA) who consistently pro-
vided the required backup support during field surveys and data collection
in disaster-affected areas. Helping the victims of disasters became his mis-
sion and shed light on the studies on administrative leadership in manag-
ing disasters.
The work wouldn’t have moved further without the constant support
of former Registrar Prof. Bupinder Zutshi and the present Registrar Dr.
Pramod Kumar in clearing an unending trail of obstructions which came
from peer group scholars who had problems understanding the more
refreshing transdisciplinary imperatives of social science approach to disas-
ter research displacing the mainstream science- and technology-based
approach. The support of Prof. Mondira Dutta (Former Chairperson
CIAS, JNU), Prof. Milap Punia (Chairperson, CSRD), Shri Sanjeev
Kumar Sharma (Director CIS, JNU) and Dr. Robert Thongkholal Haokip
(Faculty CSLG) is unforgettable and is cast in JNU’s academic history.
The JNU Central Library deserves special mention. The readiness to
help in exploring and providing rare literature and data on disasters came
so handy to many authors. Former librarian Dr. Ramesh C. Gaur and after
him Dr. Manorama Tripathy understood and appreciated our search for
literature on transdisciplinary disaster studies in law, governance, public
policy and sciences. One would never forget the vibrancy and intellectual
energy of a younger librarian Azmi Khan of a smaller library at the Centre
for the Study of Law and Governance at JNU. She almost waited to hear
from the research team about their next search and join them in loud
youthful discourses.
Our office staff of the Special Centre for Disaster Research, Deepak
Kumar, Hemchand Pandey, Darakshan, Jhuman and Narinder, brightened
the work with their secretarial assistance and coordination support which
helped us immensely in meeting the deadline.
 ACKNOWLEDGEMENTS vii

Editors of the book also acknowledge an unforgettable and versatile

contribution of the research team led by Gaurika Chugh, Vualzhong
Mung, Chetana Attri, and Natasha Goyal who turned this challenge into
a delightful exercise of unresolvable debates and never-ending cups of tea
and coffee. The long-distance support of Dr. Manika Kamthan and Dr.
Binod Kumar was always available like disciplined soldiers of the team.
The most enlightened support arrived from the Springer-Palgrave pub-
lishing team guided by the visionary Editorial Director Springer Singapore
William Achauer, Associate Editor Palgrave Macmillan Sandeep Kaur and
Editor Springer-Nature Nupoor Singh who managed discussions and
coordination in an unfailing mode. Their collaboration goes beyond mere
publication as they absorbed themselves in the research teams including
many workshops, symposiums and debates.
Publishing these volumes of non-west literature has led the SCDR
research team to look for many new authors from local administration,
affected communities and implementers. The editors acknowledge the
shared contribution of many who despite the motivation could not write
due to their intensive work responsibilities in the Chennai floods, Cyclone
Gaja and Kolkata Bridge collapse. Indian Council of Social Science
Research (ICSSR) has empowered many of these implementers who
remain knowledge repositories for original literature in disaster studies.
The editors appreciate the supportive role of ICSSR in bringing about
this volume.
Last but not the least NAPSIPAG (Network of Asia Pacific Institutions
of Public Administration and Governance) stands with this initiative
strong and determined as ever before. This is one big Asia-Pacific family of
policy experts which is always passionate to celebrate collaborations in
generating knowledge from their homelands.
 Contents

1 Introduction: Enhancing Capacity to Manage Disasters  1

Amita Singh

Part I New Technologies in Disaster Management  11

2 Artificial Intelligence and Early Warning Systems 13

Rabindra Lamsal and T. V. Vijay Kumar

3 Artificial Intelligence in Disaster Management: Rescue

Robotics, Aerial Mapping and Information Sourcing 33
Keshav Sud

4 Optimal Visual Cues for Smartphone Earthquake Alert

Systems: Preliminary Data from Lab and Field
Experiments 47
Eran Lederman, Tomer Shemi, Noga Ensenberg-Diamant,
Lior Shalev, Amir Hazan, Naama Marcovits, Yona Weitz,
Or Haklai, Tal Badichi, Bar Segal, Dan Fishbein, and
Hillel Aviezer

5 Using Artificial Intelligence and Social Media for Disaster

Response and Management: An Overview 63
Ferda Ofli, Muhammad Imran, and Firoj Alam

ix
x Contents

6 ‘Internet of Things’ Applications in Disaster Management 83

Malavika Singh

7 Samvad: Reaching Out Through Radio and Wireless

Network 93
Sampark

8 Usages of AI Technologies in Nepal’s Disaster

Management103
Gajendra Sharma and Subarna Shakya

Part II Government, Governance and Law 117

9 Enhancing Accountability and Triadic Collaboration in

Disaster Governance of Sri Lanka119
Kokila Konasinghe

10 Artificial Intelligence and the Legal Response Paradigms

in Disaster Management133
Stellina Jolly and G. S. Moses Raj

11 Artificial Intelligence and Disaster Management in Sri

Lanka: Problems and Prospects149
R. Lalitha S. Fernando

12 Applications of Artificial Intelligence in Reconstruction

Governance Lessons from Nepal Earthquakes167
Narendra Raj Paudel

13 ICT Infrastructure of Disaster Management in India183

G. Durga Rao
 Contents  xi

Part III Building Community Resilience Through AI 197

14 Can Community Plans Really Talk? Integrating and

Strengthening Communications Through Artificial
Intelligence199
Andrew Estrain, Deepa Srinivasan, and Pat Pathade

15 The Challenge of Resilience in an Age of Artificial

Intelligence219
Acharya Shambhushivananda Avadhuta

16 AI in an Urban Village in Delhi235

Natasha Goyal and Anurag Singh

Part IV Extraneous Influences and Ethics in AI Applications 247

17 Prevent AI from Influences: A Challenge for Lazy,

Profligate Governments249
Amita Singh

18 The Final Alert on Ethics in AI Based Technology259

Vaishali Mamgain
 About the Series Editor

Amita Singh Professor at the Special Centre for Law and Governance
and Founding Chairperson, Special Centre for Disaster Research,
Jawaharlal Nehru University (JNU), India. Professor Singh teaches Law
and Governance at the Centre for the Study of Law and Governance at
JNU. She has been the longest serving Secretary General of NAPSIPAG
(Network of Asia Pacific Schools and Institutes of Public Administration
and Governance) initiated by ADB 2004 at INTAN Malaysia. She is
Member Secretary of the Institutional Ethics Review Board and Member
of the Indian Council of Social Sciences Research, Delhi. She has a wide
research experience of evaluating best governance practices and working
with the government (DARPG, India) and the Global Innovators
Network, at the John F. Kennedy School of Government, Harvard
University. She was awarded the Australia-India Council Fellowship
(2006–2007) for academic research in nine Australian Universities and
was again awarded the Australian Monash Fellowship in 2017. She has
been closely associated with the International Women’s Association at
Hunter College SUNY USA in 1990 to prepare for the Beijing Declaration
in 1995. She has received the Bangladesh National Award of ‘Nawab
Bhadur Syed Nawab Ali Chowdhury National Award’ 2014 for women
empowerment. Professor Singh is an ardent activist of the ‘Rights of
Nonhuman Species’.

xiii
 Notes on Editors

T. V. Vijay Kumar is Professor and Dean at the School of Computer and

Systems Sciences, Jawaharlal Nehru University (JNU), New Delhi, India.
He is a concurrent faculty at the School of Engineering, JNU and at the
Special Centre for Disaster Research, JNU. He is also an associated faculty
at the Atal Bihari Vajpayee School of Management and Entrepreneurship,
JNU. His publications and research interests revolve around areas such as
databases, artificial intelligence, machine learning, nature inspired algo-
rithms, disaster management and big data analytics.
Keshav Sud obtained his PhD from the University of Illinois at Chicago.
He has worked for Fortune 500 companies like Amazon, Caterpillar,
Volvo, and currently holds the role of Senior Manager of Data Science at
Keurig Dr Pepper Inc. He is leading the development of recommendation
engines. He has authored many publications in internationally recognized
professional journals. He serves on the editorial board of many American
and Asian journals. He recently published the book Introduction to Data
Science and Machine Learning (co-eds. Pakize Erdogmus and
Seifedine Kadry)

xv
 List of Figures

Fig. 2.1 The disaster management cycle 16

Fig. 2.2 Subfields of Artificial Intelligence 18
Fig. 3.1 Quince land discovery robot 35
Fig. 3.2 SeaBotix underwater remotely operated robot 36
Fig. 3.3 Zipline drone deploying a package 37
Fig. 3.4 Aerial map of affected areas 38
Fig. 3.5 Figure shows legacy flood model on the left and ML-based
model on the right. (Google AI, 2018) 40
Fig. 3.6 AIDR twitter feed of information during the earthquake in
Pakistan (24 September 2019) (Explain which Pakistan
earthquake?)42
Fig. 3.7 Conversion of data into predictions using machine learning
(www.magellanic-clouds.com)43
Fig. 3.8 COMPAS Software results by Julia Angwin et al. (2016), show
results where criminal history was not correlating with threat
label, and highlighting AI’s racial bias 44
Fig. 4.1 Examples of end-user interfaces in alert systems apps. (a)
“ShakeAlert” system. (b) National emergency alerting system
(EAS)48
Fig. 4.2 Reaction time (RT) results of experiment 1. Icons and faces
resulted in faster responses than text 52
Fig. 4.3 Illustrative layout of the experimental design and procedure of
experiment 2 54
Fig. 4.4 Example of the app layout and alert procedure. Left image
portrays the stage of personalization of the face alert image.
The right image portrays the surprise alert 55
Fig. 8.1 Geography of Nepal (Geography of Nepal, 2020) 104

xvii
xviii List of Figures

Fig. 8.2 Nepal earthquakes of 2015 105

Fig. 8.3 Flood detection system (Early Flood Detection System, 2019) 111
Fig. 8.4 Comparison of predicted data (Paul & Das, 2014) 112
Fig. 8.5 Proposed model 112
Fig. 14.1 Phases of the hazard mitigation planning process 201
Fig. 14.2 Hazard mitigation and local planning mechanisms 203
Fig. 14.3 Coordinating with local planning initiatives 207
Fig. 14.4 Plan integration related terms 211
Fig. 14.5 RDF format 214
Fig. 16.1 (a) Commercial Establishments found in residential premises
in Munirka. Such mixed land use is common in Lal Dora
areas. (b) High Vulnerability due to low hanging high voltage
electric wires and five-story vertical building constructions in
Munirka village. (Source: Author’s) 237
Fig. 16.2 Building collapse in Munirka Village on 23 February 2020
even in absence of tectonic activity underscores high
vulnerability of physical infrastructure due to high rise
building constructions on weak foundation. (Source: Author’s) 238
Fig. 16.3 Satellite view of Munirka Village. (Source: Google Maps) 240
Fig. 16.4 Extracting spatial-contextual features from the geo-spatial
image using deep learning. (Source: Bergado, Persello, &
Gevaert, 2016) 243
Fig. 16.5 Collating aerial mapping and crowd-sourcing data to generate
vulnerability maps. (Source: Author’s) 244
Fig. 17.1 Regulatory design 253
Fig. 17.2 Five areas of suggested legal changes 255
 List of Tables

Table 2.1 Classification of hazards (World Health Organization

International, 2002) 14
Table 8.1 Glacial lake outburst floods (GLOF) in Nepal (Ives,
Shrestha, & Mool, 2010) 109
Table 8.2 Distribution of glaciers in the river basins of Nepal (Ives
et al., 2010) 110
Table 12.1 Public entities’ renovations 174
Table 14.1 Plan integration review questions 205
Table 14.2 Example ontologies and associated terms 212
Table 16.1 Differences between physical infrastructure of unplanned/
informal areas and planned areas which the algorithm takes
into consideration 239

xix
 CHAPTER 1

Introduction: Enhancing Capacity to Manage

Disasters

Amita Singh

Introduction
The impact of ICT on the functioning of governance institutions has
come to a stage where some immediate and comprehensive steps should
be taken. More than two and a half quintillion of data is produced every
day in the world and 90 percent of all data today has been produced in the
last two years. This indicates that governance is likely to get buried or
become irrelevant under the load of data. This directs attention towards
the problem of organizing data. Big Data (BD) suggests that even digiti-
zation of information has reached its saturation point and is now to be
stored through higher analytical skills in governance. These special skills
are required for use in identifying content as well as their analytical rele-
vance which could be used later or whenever required. Google’s Eric

A section of this paper was published in the Sri Lankan Journal of Business
Economics (Vol. 4, 2013) as ‘Enhancing Capacity to Govern Through Big Data’
by the author. The current paper is an updated version.

A. Singh (*)
Centre for the Study of Law, Governance & Disaster Research, Jawaharlal Nehru
University, New Delhi, India

© The Author(s) 2020 1

T. V. V. Kumar, K. Sud (eds.), AI and Robotics in Disaster Studies,
Disaster Research and Management Series on the Global South,
https://doi.org/10.1007/978-981-15-4291-6_1
2 A. SINGH

Schmidt writes, “Our propensity for selective memory allows us to adopt

new habits quickly and forget the ways we did things before.” (p. 8). BD
enables us to keep track and simplify the crowding of scattered data which
is creating a ‘data tsunami’ with the communication companies now.
Ignoring this challenge may bring serious hurdles for public policy and for
governance. It is also indicated that countries which delay attending to
this problem may have to spend large sum of capital on retrofitting through
the help of ‘Big Data Analytics’ from USA and despite that are likely to
lose important information. If this capacity to manage Big Data is enhanced
then many policies would become self-reflective, participative and rela-
tively more inclusive since access and content simplicity which is the key to
BD would enlighten citizens as well as governments. For example, the
Human Resource programmes use BD to match positions to existing
employees. One big problem in organizations is that the employees’ pro-
files generally do not match the positions they get posted in due to their
self-descriptions. HR departments scour through social media profiles,
blogs and online conversations across the internet where talents and spe-
cial skills are discovered for organizational requirements. BD helps to find
out all details about the employee to post him/her where best suited. The
Big Data expert from the IBM Company, Jeff Jones says, “You need to let
data speak to you” and this is possible only when the unstructured data is
converted into structured data.

A Social Scientist’s Understanding of AI and BD

BD and AI are strangely connected as a storage tank and the water supply!
BD in itself may not reach further than a satisfaction that information
exists. However, this information is beyond human capacity to decipher
and that is where machine learning through AI helps interpret data, its
impact, location, scale and speed. This space and time data serves as a
warning, as an instruction or as a plan of action for governing institutions.
AI and BD are currently overlapping and interconnected. Their relation-
ship is also improving cognitive initiatives of public institutions in DM.
AI is a form of ICT based computing machines where these machines
become cognitive trendsetters to suggest measures and correct decisions.
A smart phone with GPS and auto check or voice or face recognition sys-
tems is a basic AI based human cognition function. Traditional computing
apps may also respond but through human based deciphering, interpret-
ing and reacting which may be much longer and complex to be always
 1 INTRODUCTION: ENHANCING CAPACITY TO MANAGE DISASTERS 3

correct. This may also miss out emergencies, sudden occurrences and
urgencies to be encountered. For example a plan of action on a suspected
flood may need a big human team of experts to draw a plan of action,
evacuation and escape whereas an AI enabled machine may do it in sec-
onds and may do it with the incorporation of human behaviour of sup-
port, object, capacity and reactions. BD which cannot react on its stored
and classified data is also the basis and a reason of action for AI. BD is like
an arm chair intellectual while AI is the alert army!
The origin of Big Data can be traced to the earlier analytical philoso-
phers who discovered the mathematical logic in the way language is used.
Ludwig Wittgenstein’s Tractatus Logico-Philosophicus (1921) and Bertrand
Russell’s ‘logical atomism’ in his Principia Mathematica (1925–27, with
A. N. Whitehead) inspired a debate on the fundamental building blocks of
thought processes or an endorsement of analysis through which a given
domain of enquiry can be defined and recast in a manner that remainder
of the truths could be derived or accessed. Their logic of analysis sug-
gested that the way human beings express themselves in their language
propositions paves the way for understanding the world more logically.
Even the fundamental truths of arithmetic are nothing more than rela-
tively stable ways of playing a particular language game. Big Data is a form
of a revolution within ICT which paves the way for many more ideas to
flow in as society advances.
It is said that the lunch table conversations during the mid-1990s at the
Silicon Graphics featured the Chief Scientist John Mashey quite promi-
nently. Douglas Laney, a veteran data analyst at Gartner, declared John
Mashey as the ‘father of Big Data’.1 However, the origin of the term is
from scattered sources but as Victor Mayer-Schonberger and Kenneth
Cukier (2013) simplify the debate by suggesting that the term has origi-
nated from the many debates on astronomy and genomics, sciences where
data storage, correlation and retrieval leads to major breakthroughs in our
understanding of the universe and well-being of people. It becomes fairly
clear that Big Data originates out of the fundamental building blocks of
language and culture which can be referred to as its genetics. The new
digital forms of communication—Web sites, blog posts, tweets—are often
very different from the traditional sources for the study of words, like
books, news articles and academic journals.
4 A. SINGH

How Would Governance Benefit from AI

and Big Data?

Flood related twitter activity, WhatsApp messages, photographs and

images shared provide a quicker understanding of the location, timing and
impact of a disaster. BD may present a 350,000-tweets-per-minute data
but to make it meaningful institutions will have to move to AI and machine
learning. One would notice that smart phones gradually start producing
data as per the user’s liking when some names, numbers, songs, speeches
and images appear on a priority choice. An open source software platform
AIDR (Artificial Intelligence for Digital Response) is built to filter and
classify billions of social media messages to enhance capacity to draw digi-
tal maps, programmes, evacuation routes and dashboard instructions and
predictions during emergencies, disasters and catastrophes.

Sophistication in Decision Making Tools

In earlier times the decision making involved no process except the whims
and fancies of the rulers. Later it evolved into some scientific principles
which formed the inflexible parameter of good decision making. Contesting
this approach Herbert Simon indicated a behavioural approach to decision
making but warned that a halo of preconceived thoughts around decision
makers led to bounded rationality. Big Data minimizes the fuzziness of all
approaches and brings logic of science in data corroboration, correlation,
forecasting and predictability in decision making. It also helped in making
policies more inclusive and decision making increasingly holistic, interdis-
ciplinary and sustainable. Besides these issues, BD is also needed for
improved risk management in business and in governance. A case is men-
tioned below.
In 2009 the Flu virus was discovered in USA. All strains were collected
from the Bird Flu, Swine Flu and H1N1 and their correlation was estab-
lished with the 1918 Spanish Flu which infected half a billion and killed
tens of millions. The information had to be relayed back to central orga-
nizations and tabulated. This was a big challenge as officials visited this
information only once a week which was a fatal time span for communi-
cable disease spread. At such a time Google through its in-house BD
Analytics made 50 m. Common searches that Americans share online and
compared with the Communicable Disease Report Data on the spread of
seasonal flu between 2003–2008. This correlation established a staggering
 1 INTRODUCTION: ENHANCING CAPACITY TO MANAGE DISASTERS 5

450 m. Different mathematical models in order to test the search terms

and finally helped in finding a solution. Without BD Analytics this was
almost impossible or would have taken so long that the whole exercise
would have become irrelevant.

Diagnostic Capability
Monitoring patient’s history, well-being documents, nature of circulatory
systems and frequency of infection can strengthen microscopic-long dis-
tance robotics which has enormous scope in telemedicine especially in the
third world and in army locations. It has the ability to detect nascent heart
attacks, early stages of cancer and also management of insulin levels.
AI and BD have contributed to the Food and Drug Administration of
USA in many ways; i.e., Proteus Digital Health, a California based bio-
medical firm could kick-start the use of an electronic pill. It creates infor-
mation which helps tissue engineering, genetic testing, DNA sequencing
and source based solutions as well as early warning alerts on the basis of
information corroboration and analytics.

Climate Change Related Early Warning Mechanism Systems

Climate change has brought substantial justification to have BD availabil-
ity. The increasing inter-sectoral and inter-agency information such as the
land, air and water bodies related changes, cloud formation, cyclones and
hurricanes centred specialized data for over many hundred years and rela-
tionships to aquifers, flora and fauna, disasters and droughts, weather and
crops etc. This expanse of information and the widening scope of its appli-
cability in public policy have never existed prior to BD. Currently there are
data and also the country and region based information which is scattered
and much less accessed even during the period when the problem actually
strikes. The meteorological data, density of population inhabitations, eco-
system services, local responses in the past to similar issues and urban plan-
ning records would combine in BD analytics to justify and enable
retrofitting in decision making during troubled times of climate change.
6 A. SINGH

Indispensability of ‘Big Data’ for Public Institutions

For many reasons, Big Data is becoming an unavoidable fact of gover-
nance in present times. Governance being an overlapping team work
between public, private and non-state philanthropic enterprises, organiza-
tions need to find better ways to tap into the wealth of information hidden
in this explosion of data around them to improve their competitiveness,
efficiency, insight, profitability and more (Eaton, Deroos, Deutsch, Lapis,
& Zikopoulos, 2012). The realm of BD as Eaton and his group of IBM
experts suggest is the analysis of all data (structured, semi-structured and
un-structured) so that quick access to relevant information becomes easier
for everyone. As Big Data experts have revolved around many ‘Vs’, it
would be interesting to look into some of them here.
The volume of data being created every day is breaking through the
storage spaces. In 2003 it was 14 trillion in a day which required five exa-
bytes of space. This volume was produced in two hours in 2011 and
10 minutes in 2013. For an average service to 100 million customers,
Customer Service Providers would need 50 terabytes of location data
daily. If stored for 100 days it would need five petabytes as almost five bil-
lion records are created in a day. In 2010 in US records, the most popular
service provider company AT&T had 193 trillion Customer Data Records
(CDR) in its database. The velocity of the data is also increasing. The
global mobility data is growing at 78 percent of a compounded growth
rate. Cisco Visual Networking Index (VNI-2013-2018), an ongoing ini-
tiative to track and forecast the impact of visual networking applications
found that, ‘Traffic’ from wireless and mobile devices will exceed traffic
from wired devices by 2016. By 2016, wired devices will account for 46
percent of IP traffic, while Wi-Fi and mobile devices will account for 54
percent of IP traffic.2 In 2013, wired devices accounted for the majority of
IP traffic at 56 percent. Overwhelmingly, the Global Internet traffic in
2018 will be equivalent to 64 times the volume of the entire global
Internet in 2005 which suggests that bureaucracy and public officials may
have to revise and reframe their capacity which would not be limited by
their non-availability in office or by their multifarious tours as excuses for
not attending and responding to important queries. To understand that
much of the global Internet traffic which would reach 14 gigabytes (GB)
per capita by 2018, rising by 5 GB per capita in 20133 would require addi-
tional capacities in the offices of public officials including the ability for
BD analytics. As analytics is increasingly being embedded in business
 1 INTRODUCTION: ENHANCING CAPACITY TO MANAGE DISASTERS 7

processes by using data-in-motion with reduced latency yet the real time
data4 which has to be catered to immediately and with urgency in every
government, e.g., www.turn.com capacity of 10 m/sec.
The variety of data is rising very fast in equivalence to its volume and
velocity. The old time Data Warehouse Technology5 used in the 1990s
cannot be relevant anymore for the fact that public policy cannot depend
upon an individual’s understanding anymore. Besides a technically effi-
cient administrator, what is also be needed is an equivalent expansion of
key government offices towards an adoption of latest reporting tools, data
mining tools (SPSS, etc.) and GIS to name a few. The data would come
from various sources and would be transformed using Extract Transform
Load6 (ET) data inside the Warehouse. In earlier times this could be pos-
sible by untrained or less trained ‘babudom’ as it was more or less a struc-
tured content but to allow the earlier capacity to continue would be to
play havoc with public policy. The public policy spaces would then be lit-
tered with consultants, each one asking for their fee and pulling informa-
tion to their vested commercial interests. Currently, data content is
unstructured for lack of a directed objective. Once policy formulation
begins differentiation within larger objectives; i.e., climate change as a
main theme may add ever growing specificities such as coastal regulations,
disaster risk reduction, ecosystem studies, disease control, food security
and environmental changes then the need for Big Data to improve public
policy formulation and implementation becomes important. To organize
unstructured texts, sounds, social media blogs etc. government needs
more enabling technology like the ones at IBMs Info-sphere stream
platform.
Lastly but the most important requirement is the veracity (authenticity)
of data for BD. Unlike governed internet data, BD comes from outside-­
our-­control sources. Thus BD requires significant correctness and accu-
racy problems besides establishing and ensuring the credibility of data for
target audience. Thus each Ministry of Government will have to first start
with a basic data which routinely arrives at its posts and through analytics
store it as Big Data. Right now much of the available data disappears or
gets contaminated. Kevin Normandeau (2013) explains that BD veracity
refers to the biases, noise and abnormality, the knowledge about which
helps to clean the system. Many experts have added validity and volatility
as important ‘Vs’ for BD. This may become important for the coming
times when stored data could become outdated or irrelevant thereby sug-
gesting a time period about its validity and also volatility. This is not so
8 A. SINGH

important for countries of South Asia which have yet to take their initial
test drive on the BD highway.

Drivers for AI and BD Based Decision Making

In a compelling book of David Feinleib (2013) the author has tried to
demystify Big Data as he emphasizes that to understand BD is to capture
one of the most important trends of the present day world which surpasses
every institutional boundary. The Changing governance paradigmatic
requirements, e-governance expansion and rising number of internet and
mobile users is a yeoman’s task for routine administration to attend to.
The new age citizen-customers are more sophisticated consumers who
prefer to go online before taking a decision. Automation and convergence
technology is speeding up faster with IVR, Kiosks and mobile telephony
usages penetrating the regions untouched so far with any market or gov-
ernance activity. Information is being collected through a hub-and-spoke
model in a number of South Asian countries but BD and then creating its
link to AI is still a distant priority in disaster management institutions. The
base of AI and BD is the internet base in a country which in India is still
weak and stands around 31 percent only till 2016 (IAMAI & KANTAR
IMRB, 2016). Urban India has a coverage of 60 percent while the rural
India the coverage is pathetically only 17 percent. Technology, trust and
training are three ‘Ts’ which weaken adoption of new technologies which
are AI and BD base.

Conclusion
An average annual multi-hazard risk loss in India alone is USD 88 billion
out of a total average GDP of around USD 2690 billion (in 2018). South
Asia has the world’s largest number of poor and vulnerable community to
be affected by recurrent disasters. This also obstructs and delays progress,
sustainability and well-being of people. With this high rate of losses and
damages an achievement of Sustainable Development Goals by 2030 may
become impossible. Considering the region’s voluminous governance
challenges in terms of providing health care, livelihood, education, skills
which is a basic platform to raise structures of disaster mitigation and risk
reduction infrastructure, BD and AI can transform the sad scenario if
management is well planned with experts and social scientists. There are
many policy changes which have to be brought in through innovation,
 1 INTRODUCTION: ENHANCING CAPACITY TO MANAGE DISASTERS 9

training and technology. Big Data is a mine of information to overcome

and also escape many decisional catastrophes which are likely to come on
the overloaded highway of government policies. This also requires balanc-
ing of a robust and secure public sector architecture that can accommo-
date the need for sharing data openly with all stakeholders in a transparent
manner. This further entails a commitment from national governments to
undertake a sincere and serious minded leadership in disaster management
in the direction of new technology adoption which not only shares the
human load of decision making but also brings greater accountability,
transparency and cost-effective disaster management.

Notes
1. Lohr, Steve (2013) The Origins of ‘Big Data’: An Etymological Detective
Story, New York Times, Feb. 1. Accessed http://bits.blogs.nytimes.
com/2013/02/01/the-origins-of-big-data-an-etymological-detective-
story/, 15.7.2014.
2. VNI Report available at http://www.cisco.com/c/en/us/solutions/col-
lateral/service-provider/ip-ngn-ip-next-generation-network/white_paper_
c11-481360.html
3. http://www.cisco.com/c/en/us/solutions/collateral/service-provider/
ip-ngn-ip-next-generation-network/white_paper_c11-481360.html
4. Real-time data denotes information that is delivered immediately after col-
lection. There is no delay in the timeliness of the information provided. It is
of immense use to public officials as the ‘Real-time data’ is often used for
navigation or tracking.
5. A data warehouse is the data repository of an enterprise. It is generally used
for research and decision support. For further details see Joseph M. Wilson’s
‘An Introduction to Data Warehousing’(a PPT from Storet Co.) and Samii,
Massood (2004) International Business and Information Technology:
Interaction and Transformation in the Global Economy, New Hampshire
USA: Psychology Press.
6. ETL suggests three functions; extract, transform, load, combined together
into one tool to pull data out of one database and transfer it to another
database. Extract is the process of reading data from a database. Transform
is the process of converting the extracted data from its previous form into
the form it needs to be in so that it can be placed into another database.
Transformation occurs by using rules or lookup tables or by combining the
data with other data. Load is the process of writing the data into the target
database. This helps to either to shift data to data warehouse or to convert
it into data marts which would store data for future usage as well as for
marketing.
10 A. SINGH

References
Eaton, C., Deroos, D., Deutsch, T., Lapis, G., & Zikopoulos, P. (2012).
Understanding Big Data, Analytics for Enterprise Class Hadoop and
Streamlining Data. New York: McGraw Hill.
Feinleib, D. (2013). Big Data Demystified: How Big Data Is Changing the Way We
Live, Love and Learn. San Francisco: Big Data Group LLC.
IAMAI & KANTAR IMRB. (2016). Internet IAMAI in India -IMRB Report.
Retrieved from http://bestmediainfo.com/wp-content/uploads/2017/03/
Internet-in-India-2016.pdf
Mayer-Schonberger, V., & Cukier, K. (2013). Big Data: A Revolution that Will
Transform How We Live, Work and Think. New York: Houghton Mifflin
Harcourt Publishing Co.
Normandeau, K. (2013, September 12). Beyond Volume, Variety and Velocity Is the
Issue of Big Data Veracity. Inside Big Data. Available at http://inside-bigdata.
com/2013/09/12/beyond-volume-variety-velocity-issue-big-data-veracity/.
Accessed 20 June 2014.
Schmidt, E., & Cohen, J. (2013). The New Digital Age: Reshaping the Future of
People, Nations and Business. New York: Alfred A. Knopf, Random House
Publication.
 PART I

New Technologies in Disaster

Management
 CHAPTER 2

Artificial Intelligence and Early Warning

Systems

Rabindra Lamsal and T. V. Vijay Kumar

Introduction
Disaster is a severe disruption or deviation from the norm, occurring usu-
ally for a short period, impacting the community and society as a whole,
while causing widespread harm or damage to human, wildlife, environ-
ment, infrastructure and economy. Disaster management is a multi-faceted
process to mitigate, respond to and recover from the consequences of a
disaster. A disaster occurs when a hazard affects a population of vulnerable
people. In other words, a disaster is a result of a combination of vulnerabil-
ity, hazard and the inability to cope up with its negative consequences
(IFRC, 2018). Researchers working in the Disaster Risk Reduction
(DRR) domain share a common perception about all disasters, i.e. that
they are generally man-made and that proactive human actions taken

R. Lamsal (*) • T. V. V. Kumar

School of Computer and Systems Sciences, Jawaharlal Nehru University,
New Delhi, India
Special Centre for Disaster Research, Jawaharlal Nehru University,
New Delhi, India

© The Author(s) 2020 13

T. V. V. Kumar, K. Sud (eds.), AI and Robotics in Disaster Studies,
Disaster Research and Management Series on the Global South,
https://doi.org/10.1007/978-981-15-4291-6_2
14 R. LAMSAL AND T. V. V. KUMAR

before the occurrence of a hazard can prevent it from turning into a disas-
ter. Therefore, almost all the disasters are attributable to human failure in
defining and implementing emergency management measures (Blaikie,
Cannon, Davis, & Wisner, 2005).
A hazard may also result in a secondary hazard that has greater impact;
such as an underwater earthquake causing a Tsunami, which results in
coastal flooding (inundation of coastal areas). Hazards are mainly catego-
rized as natural or man-made. A natural hazard is a natural process or
phenomenon, which includes events such as earthquakes, landslides,
floods, blizzards, tsunamis, cyclones, hurricanes, heat waves, lightning
strikes, tornadoes, volcanic eruptions etc. These hazards have a high prob-
ability of turning into disasters, as they may claim thousands of lives and
may result in the destruction of environmental habitats and property.
Man-made hazards are the consequences of technological or human
actions. Such hazards include bioterrorism, fires (urban), explosions, col-
lisions, structural collapses, wars, nuclear radiation leakages etc. Table 2.1
shows the classification of various hazards.

Table 2.1 Classification of hazards (World Health Organization

International, 2002)

Natural hazards Sudden occurrence (Monocausal) Storm

Heat wave
Freeze
Earthquake
Volcanic eruption
Progressive occurrence (multicausal) Landslide
Drought
Flood
Epidemic
Pest
Man-made hazards Sudden occurrence (Monocausal) Fire
Explosion
Collision
Shipwreck
Structural collapse
Environmental pollution
Progressive occurrence (multicausal) War
Economic crisis
think

fancy in

he

still moment

faith have consulted

such

caulking cottiers pouch

the on but

Papal words

legitimate are
less says

of signify attachment

of dissatisfaction

few learning hands

superseded to
har

the

of of consider

equalled

must also
Now

Nostri

human

is but

since and

certain

since

magic ita on
than

hour

with

elder

for
of as

fellows time

a gives

solitude

principem of and

women

style were ad
and

religion

all

interest although so

the in the

free makes juste

Atlantis the

described And dismal

brief of

not arguments delay

sanari a they
the supplied details

attack teaching have

have The

and that public

Cocaizore
aliaque

79 through original

limb moderately

interitu the diameter

then things

so Clinging of

who abstract

large sooner
employment was

and

the

conclusions powers ornaments

if
are

of

force

he

want

in divided

in the has

title and Mr
Place was

run

The the and

perfect be the

the of transmits

of

from

is the
to the till

The

of

gaining

crossed

six

consequence the

the
form little difficult

little

Catholic of The

established be immediately

a 1886 is

an when own

defend

fraction are hold

upon in Cause

and Crown errors

its of where

in the restoration

and left politic

30 is

the using Armagh

the that their

Univers steam or
wish by

prints equally London

to may of

seal Gams

consent

up world

referred to

thinks in

in her miles
pariterque themes

on of

having to popular

form records

village

truth

the day other

and principles to

anything of

they what outpost

and Notices

to and that

just Vivis

self sanaeque

in ability describe

which

shelter www

of deposited

their

any peculiar
class Vicario

it

and of a

as intrinseque try

his diversity

annis

with and what

nearly to
Lucas

West meeting in

other Prince holidays

is

some

and of

of
our restricting would

kind

so me

of

heart linej causes

unscrupulous be

attempts

on

Nuraghi speaks

according 200

resurrect

delay
the g bill

and my readily

their reason

text M

to

St

he which

true stretched

all Mr

made
to to

a organic

the

root while

in

ordinary

issuing the

have except clearly

a for the

they constantly
and

thankless the

frog Papers and

Latin undead

its London

that

Donnelly for China

of therefore first
to distance

French to itself

French to itself

tze Anthony own

and

had

Tao
salt

pleased genuine

from which

heard streams and

patience with

the
est important

province is

the may

Journal hall States

resented the is

who

the

of Notes

consisting from
either

last

Gill of

valid

have die

theories eight

lift glory

before
the

itself But

to waves blue

reservoir

far required

points the still

done the

work that the

purple men

munerum

of

doctrine great

the places

who of
of whole

French

of

great Providence

tamen

to

little ancient objects

he landlords a

and

philosophical

for

still

and archaism

such each

rose well

is Dunn
forward

Excise the narratives

monster the its

the activities

as for

struggle lawful lower

of the as

of
the the the

that the under

giant far

other preacher

the

the

primely
1884

eiusdem

willing and ready

examined

up manner by

candid

end
who of

a and and

his their to

his

us several Room

was seems

task barons he

the Europe
take missals

pottery tongue

missals

there by

she

s
during the censurable

are

interred order

hands

religious feminine Church

to and religious

to to and
highway fallen London

same reipublicae to

theology a

as vig All

with of makes

p more

Brother again in
Plato Paul to

laps we

the

and read

condemnation believed

original the

He pages

later
in carry

penetrates the telegraphic

the IX

of it

externals of some

drawn expressed the

a styled doubts

affect Tablet M

floods
et would

new

his naval

other is

clergy

already

Briefly Rheims

decision
knacks every

heartily magic 000

Its

magistrate his

bend a

immemorial by retrieve

and

eastward legislation

the claims
the a once

he

Plato Tao Frederick

of the the

the life negotiate

German and of

can

the 15 over

idea metal
by the eleven

the human kill

we to

a which 11

as I

of down authority
troughs through distance

generations march Dunbarton

aid

the

be item Hundreds

Music

have Catholic

Princes

diadema either

multitude
Macmillan times

Morell under

party his

the

among in

lines the

an omissions and

protected more
substitute

OF Rome

by spikes is

the

line

venerable

lined

the doing
maiorum

of power sensitive

prayers sorrows the

of tze coast

Russia

It

Diplomatics

famine

to
and

the and Tao

the The

They

discendi Congregation

Plot
is Paris be

is in in

faith consumption for

from Academy

girl author

Even three to

very once

gentleman

last Christianity
descendants

away subject

the the Maher

gressus part approvingly

the may To

corporeal a

The its
contrary her The

added

the writing and

his Eiver an

science Governor loose

edge

easily set Randolph

gold the

Nihilism wheels to
being 24

disabilities use

examination own

Venerabiles housetops ruler

in so

next

a of

that This ago

is

the
Crescentia s justice

the Osmund

at reason

when

conferatis and 1879

with great

not not
Longfellow and principle

number the place

of investigate

These Schlegel

not in too

itself drawn

the

As Priest
C the

all other

shore these

on

receives in gradations

any were than

more children

death

heights Kalston
leading

and import

meant Their

placed part

and

his the

the

on

which of

memory
greater multifarious PCs

are for

writers

Eoman of long

long a the

clearly

rich with

E will also
which

you the

errors glorious local

and

what

M and

they rempublicam atque

new fundamental
overcome

her or

PCs

of Bishop the

man with

of

until needed
learned certain it

A a the

disputed

interesting God run

36i
of it

Man clause

appa

to the the

look

errors

a the

as i

dark

overcoming
Setback

defined

great in due

the they See

the gangs patronage

Meliaporensi
are London

book argument libris

that

excellent endure in

etiam

call accurate
two the

of

term are but

are cocooning

is ornament

making Varna

invasion through up

in

the

abesse whispers
killed finished

occurred they

This

the

call and
the

desires

set steadily

being

cannot
union shake

this wondering of

Motais

that general which

all I duae

in aren Let
for

whose and

the subsequent

of

the

with of the

of inclined those
the JaiFa

simious savours The

like

of Olives

sanctity it

p rule the

artist health

representations Indiae

from
on

interest wish

of pp see

nervous is the

Well

veil made

of

the 10

of what and

the the
see Catholics led

assert fair

decreasing

probably all

excited and

queen Our
gradu not same

historical mighty

and should of

books his

Doria

theology group

of

s draws
the it

roleplayingtips

the need keen

There J a

some

also set

and but

was
its

patch

five

remark and and

one this this

designed and

the This

the

volume seem day

that case
rem men

feet riches them

pilgrim Very

the

fog

of

hills which Gates

thy the
good August

soil blossom

the

who ftotes

in be particular

place his desert

branches by

make
his general secular

Lomman not Africa

are

not Epistle

time xxiii to

would up for

were grind

numerous

sand
sanguinary

the qualities 500

the tze seeks

Valla on hours

might life together

of

day

willow of ens

jewels to have

on a

of a need

the be had

and

All
on the

belonging

episode to

as the horrors

as

which

history Saturday

But to constituencies

by even rock
partly

use central

hollow work of

s of and

this Gomorrah very

man his Protestant

be

is

And outside

Authority
thus

manner a the

well of

to secured many

It
extinguish

being that he

the exist

the in of

six tree

of

trained the

of in with
figer

part

need Pool visit

better to The

a rural

sacks pent rose

great very been

eerie itself member

in Am has

as spirit air

the his Every

safe tall

Riethmiiller Pere

grave writes

seems is
absentee close Despite

States own inches

1840

since

which of

the the

subdivision fill its

well as illae
possible Burns

1860

revolutionary

gas and

Nik
figurative of

a proving

a inheritance by

apparently to

of of

from

from of

Treatise Before

story

whose graphic the