<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>A cloud based framework for Emergency Management: experiences in Calabria Region</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Extended Abstract</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Giuseppe Tradigo</institution>
        </aff>
      </contrib-group>
      <abstract>
        <p>We present an ongoing project aimed to support the management of the procedures, mainly the handling of emergency calls, for the centralized regional emergency management agency (Protezione Civile Regionale in Italian). The main goal consists in enabling the sharing of information among di erent active locations that are in charge of acting in any emergency which occurs on the territory. We present a prototype version of the system that has been implemented by following the Calabrian Emergency Management agency speci cations. The system is in charge of collecting information regarding various types of emergency such as earthquakes, oods, stormy seas and other extreme weather conditions. In addition, the system o ers some functionalities for the management of emergency teams as well as volunteers and available resources (e.g., vehicles) by means of the application of geographical optimization paths. The requirements include also the design and development of a mobile application allowing citizens notify in real-time emergency conditions (e.g., forest res during summer time).</p>
      </abstract>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>We here present an ongoing project and a system prototype that has been
de ning with PCR. Recently, PCR decided to make a reorganization of its
telecommunications infrastructure, application procedures and communication
ows through regional sites and, as a consequence, it has to integrate all the
communication facilities through the de nition of a centralized structure. The
aim is to de ne a datacenter where information regarding: (i) emergency calls
and catastrophe monitoring; (ii) source data management (iii) protocols
applications to save life and people as in case of earthquakes or oods or landslides due
to heavy rains. The PCR is in charge of apply intervention even in case of car
accidents or problems in reaching populations. ICT protocols and infrastructures
are in charge of being updated in Calabria Region. We here report about the
de nition of a new work ow implementing what is currently managed without
automatic procedures.</p>
      <p>The idea is to de ne a system able to manage the emergency calls by means
of pages and multimedia information that has to be organized by the central
operative unit. For instance in the following we report a possible scenario:
1. a citizen is in trouble because of extreme weather conditions;
2. he calls the PCR with an help request;
3. the incident room gets the help request which is assigned an emergency level;
4. a new intervention request is generated and sent to other operative rooms;
5. the nearest operative room identi es a team and the needed resources;
6. the team intervenes and sends feedback back to the incident room;
7. the incident room shares the information with other rooms;
8. eventually the incident is closed and archived.</p>
      <p>An e cient emergency response is strongly related to a careful and attentive
prevention activity and emergency planning, which are preliminarily carried out
before any emergency occurs. In order to predict possible aws which could
occur on the territory, the PCR has many systems and operational sources already
working. Some of the preliminary goals of our work with the PCR have been: (i)
identi cation and analysis of the existing components of the PCR information
system already in use; (ii) de nition and planning of the extraction,
transformation and loading procedures from the operational sources towards an integrated
database; (iii) de nition and implementation of a new information system for
the emergency management.</p>
      <p>PCR is a Regional Public Administration Unit which is part of a network
of other Regional units who refer to the Presidency of the Council of Ministers.
It is composed by 180 employees, 91 of which work in the incident rooms (21
for each turn). Each of the incident room dislocated on the territory, is in need
to share information about resources, events and updates in real time.
Furthermore, due to the inherent complexity and unpredictability of emergency events,
the interaction for a human operator with the system has to be minimal and
extremely e cient. We decided to design and implement an information system
with a lightweight web-based interface implementing an integrated and
interactive dashboard for the incident rooms, which allows to perform:
{ scenario analysis and planning - allowing a PCR operator to acquire and
visualize all the data about the event and to plan and enumerate the resources
around it;
{ environmental monitoring network - able to retrieve and visualize real time
information from the regional sensor network;
{ models from intervention scenarios - a system module able to analyze use
cases from the past, ranking them by similarity with the current emergency
event;
{ inspection reports - with which PCR operators can obtain detailed reports
about present and past emergency events;
{ vehicles and resources management - a dedicated module for the e cient
planning and management of emergency vehicles.</p>
      <p>A rst conceptualization of the integrated dashboard for emergency data
management is depicted in Figure 1, which reports a simulation of an emergency
event with its related data aggregated and updated in real time.</p>
      <p>From the initial analysis phase we identi ed a set of existing systems and
information sources in the PCR: (i) SITGE (Sistema Informativo Territoriale
per la Gestione delle Emergenze) system, an old and unmaintained web-based
geographical information system developed with ESRI technology, containing
partial data for the emergency management with particular reference to re
events visualization, (ii) GIOVE (Portale per la Gestione Integrata delle
Organizzazioni di Volontariato nelle Emergenze di Protezione Civile) a database
containing voluntary associations informations about vehicles and resources, (iii)
an hardware/software system for the management of re emergencies composed
of 12 stations with high de nition cameras, thermal cameras, humidity,
temperature and wind direction sensors implementing a prediction model and a decision
support system in the event of a re; (iv) a regional isofrequency microwave
radio network, used for emergency radio communication, covering 80% of the
territory with nodes at 10-15 km from each other.
2</p>
    </sec>
    <sec id="sec-2">
      <title>Related Work</title>
      <p>
        In [
        <xref ref-type="bibr" rid="ref3">3</xref>
        ] a two levels architecture for an Emergency Management System has been
presented. It consists of one back-end level for control rooms and one front-end
level for teams sent to the a ected area. Each of two level is a peer-to-peer
(P2P) system. The lack of a xed infrastructure for P2P systems makes this
kind of network suitable in emergency management scenarios where it is needed
to quickly deploy a network, but the presence of access points is not guaranteed.
This architecture has been developped studying the organizational requirements
of Civil Protection Department of Calabria and the Augustus method (https:
//it.wikipedia.org/wiki/Metodo_Augustus) that it adopts. As well, since
the available bandwidth (roughly 11 Mbps) is enough, MANETs can guarantee
a good QoS level. As far as communication of front-end teams with back-end,
TETRA seems to be a good choice.
      </p>
      <p>TErrestrial Trunked RAdio (TETRA) is an ETSI standard to build specialist
Professional Mobile Networks (http://www.etsi.org/technologies-clusters/
technologies/tetra). The low frequency used by TETRA guarantees an
excellent geographic coverage. TETRA is speci cally intended for Police Fire Brigades,
Army, etc., and its relays are generally arranged so that, even if some of them
go down, the others can cover most of the area.</p>
      <p>
        In [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ] an upgrade of the old SITGE system has been presented. The updated
system had new features for the computation of indices allowing to identify areas
in re risk.
      </p>
      <p>Each italian region has its own Emergency Management Information
System. Usually, a platform of this type has a set of modules allowing to manage
the objects of the domain (events, volunteers, vehicles, maps, etc.) and their
interactions (e.g. http://www2.regione.piemonte.it/protezionecivile/index.php/
sistema-informativo ).</p>
      <p>
        Similarly, in [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ] a model for management of bioemergency is presented. Tools
for path optimization and emergency camp positioning is presented. In [
        <xref ref-type="bibr" rid="ref6">6</xref>
        ]
algorithms for emergency intervention and rescue planning are reported.
      </p>
      <p>
        The system, here proposed, starts from requirements of the new PCR and
it is in plan with respect to new italian reorganization due to recent natural
disasters. We plan to include logistic and algorithms to plan rescues, resources
according to land orography and streets. The idea is to create an emergency
support system for the PCR able to feed a continuously growing database of
events and interventions which will be valuable for research because data contains
both spatial and temporal dimensions, which can be exploited through spatial
and temporal techniques and models (see [
        <xref ref-type="bibr" rid="ref7">7</xref>
        ]).
3
      </p>
    </sec>
    <sec id="sec-3">
      <title>System Architecture</title>
      <p>We designed a rst system prototype by using deployment of components. Figure
2 shows the components layers: (i) the visual part of the emergency management
prototype (presentation layer) consists of the JSP pages that contain the views
of the data; (ii)the middleware level hosts the web application and data access
module and (iii) the database level which is an instance of MySQL database.
Figure 2 reports components and their interactions.
The functionalities regard also the possibility of signalling emergency cases
by using a mobile app. A citizen can take a picture and de ne one of the
possible application emergency cases. Citizen position as well as personal data (in
case of availability in supporting rescuers) can be sent with simple steps. The
system prototype has been implemented in a cloud based environment, to
maximize system scalability, availability and redundancy. In order to preserve the
integrity and privacy of PCR's own data, we set up an instance of the OwnCloud
environment exclusively used by the PCR, to support the functionalities of the
presented system, some of which are: (i ) artifact exchange among incident rooms,
(ii ) end users' data acquisition and storage from the mobile app, (iii ) storage
of daily and periodic reports about PCR emergency management. Nevertheless
the adoption of a proprietary cloud environment has some drawbacks. For
instance, the limited number of nodes implies some potential problems regarding
the application scalability, while the management of the cloud architecture itself
can be a costly and complex task to be sustained by the PCR.</p>
      <p>Emergency can be managed in one of the operative units by accessing to a web
screen where the emergency is veri ed and validated, and intervention protocols
are selected and applied. Resources are also managed as well as the evolution
of rescues activities. Consider that in case of emergency the communication
activities and signals reach an high frequency in terms of number of call or app
instances, thus the system should be able to track the activities. PCR has to
both (i) avoid loosing emergency alerts as well as (ii) lter out the less relevant
ones. Algorithms to prevent and automatically manage the relevance of alerts
can be applied.</p>
      <p>Figure 3 reports an example of using the developed application. There are
di erent user pro les. A citizen can apply sending information about emergency
situation with pictures and giving the status (e.g., available, not available) in
staying on site to give information in case of emergency updates (e.g., re
conditions) in case of safe observation position.</p>
      <p>The work ow process has been designed by using a model called rasor model
(standing for Rapporto Attivita Sala Operativa Regionale) which contains
information regarding emergency protocols. It contains alerting messages classi
cation, municipality, location and closest emergency team for the event; type and
status of the event; rescue team reports and emergency updates; resource
details and emergency status time tracking; nal reports. The database has been
designed using such a schema (a snapshot in italian is reported in Figure 4).
It maps the rasor model where emergency signaling regarding reports on new
emergency are annotated by PCR operators.</p>
      <p>Finally in Figure 5 we report the database schema used to map the
information related to the resource and rescue management.</p>
    </sec>
    <sec id="sec-4">
      <title>Use Case and applications</title>
      <p>Firstly, we report a typical use case implemented by the PCR incident rooms
in an emergency, at present time: (i ) an emergency call arrives at one of the
incident rooms by phone; (ii ) the operator asks for details and compiles the
paper registry (which is a non-shared distributed resource); (iii ) the operator is
in charge of distributing the news by phone to the other incident rooms; (iv ) the
decisor (e.g., PCR's Director) plans the intervention(s) and orders the teams to
intervene on site; (v ) the various teams give feedbacks on the intervention status
back to the incident room via phone.</p>
      <p>We here report some use cases implemented with the system to automatize
the use case shown above and others encountered during the requirement analysis
for the PCR.</p>
      <p>Gathering emergency calls. Currently PCR has already implemented a number
of initiatives for the dissemination of competence information through the use
of social networks in addition to more traditional communication channels. In
order to ensure similar fruition and the maximum possible dissemination of
information, the system acquires calls and information coming from web based
communication channel (mobile one). Furthermore, the mobile interface will be
integrated with the most used social networks (i.e., facebook, twitter) but also
the backend will gather comments and enquiries coming from the PCR social
network pages and analyze them, rendering ad-hoc statistics on the system
dashboard for the PCR management.</p>
      <p>Managing resources. As part of the reorganization of the management of mobile
vehicles, the system o ers a eet management module to provide an additional
ow of information of great importance in the management of emergency
situations. Modeling of minimum risk routes is still an open research topic, as well
as the development of algorithms to simulate scenarios on networks.
Fire management System. Information regarding re events, are acquired from
an already existing module and the information is mapped on the dashboard
with the location as well as the sequence of intervention updates.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          1.
          <string-name>
            <surname>Gunter</surname>
          </string-name>
          , T. D.,
          <string-name>
            <surname>Terry</surname>
            ,
            <given-names>N. P.</given-names>
          </string-name>
          (
          <year>2005</year>
          ).
          <article-title>The emergence of national electronic health record architectures in the United States and Australia: models, costs, and questions</article-title>
          .
          <source>Journal of Medical Internet Research</source>
          ,
          <volume>7</volume>
          (
          <issue>1</issue>
          ),
          <year>e3</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          2.
          <string-name>
            <surname>Yang</surname>
            ,
            <given-names>C. C.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Veltri</surname>
            ,
            <given-names>P.</given-names>
          </string-name>
          (
          <year>2015</year>
          ).
          <article-title>Intelligent healthcare informatics in big data era</article-title>
          .
          <source>Arti cial intelligence in medicine</source>
          ,
          <volume>65</volume>
          (
          <issue>2</issue>
          ),
          <fpage>75</fpage>
          -
          <lpage>77</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          3.
          <string-name>
            <surname>De Leoni</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>De Rosa</surname>
            ,
            <given-names>F.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Marrella</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Mecella</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Poggi</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Krek</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Manti</surname>
            <given-names>F.</given-names>
          </string-name>
          (
          <year>2007</year>
          ).
          <article-title>Emergency management: from user requirements to a exible p2p architecture</article-title>
          .
          <source>International Conference on Information Systems for Crisis Response and Management (ISCRAM).</source>
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          4.
          <string-name>
            <surname>Samarelli</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Manti</surname>
            ,
            <given-names>F.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Span</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>P.</surname>
          </string-name>
          , (
          <year>2009</year>
          ).
          <article-title>Il SIT di Protezione Civile della Regione Calabria: lutilizzo operativo per la raccolta dati relativi agli incendi di interfaccia e alle Ordinanze di emergenza</article-title>
          .
          <source>Conferenza Nazionale ASITA.</source>
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          5.
          <string-name>
            <surname>Schneider</surname>
            ,
            <given-names>T.</given-names>
          </string-name>
          ,
          <string-name>
            <given-names>A. R.</given-names>
            <surname>Mikler</surname>
          </string-name>
          , and
          <string-name>
            <given-names>M.</given-names>
            <surname>ONeill</surname>
          </string-name>
          .
          <article-title>"Computational tools for evaluating bioemergency contingency plans." Disaster Management and Human Health Risk: Reducing Risk</article-title>
          ,
          <source>Improving Outcomes</source>
          <volume>110</volume>
          (
          <year>2009</year>
          ):
          <fpage>33</fpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          6.
          <string-name>
            <surname>Martin</surname>
            ,
            <given-names>O.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Armin</surname>
            <given-names>R.</given-names>
          </string-name>
          <string-name>
            <surname>Mikler</surname>
            , and
            <given-names>Tamara</given-names>
          </string-name>
          <string-name>
            <surname>Schneider</surname>
          </string-name>
          .
          <article-title>"An Extensible Software Architecture to Facilitate Disaster Response Planning."</article-title>
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          7.
          <string-name>
            <surname>Tradigo</surname>
            ,
            <given-names>G.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Pierangelo</surname>
            <given-names>V.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Greco</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          <article-title>"Geomedica: managing and querying clinical data distributions on geographical database systems</article-title>
          .
          <source>" Procedia Computer Science 1.1</source>
          (
          <year>2010</year>
          ):
          <fpage>979</fpage>
          -
          <lpage>986</lpage>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>