3D TECHNOLOGY FOR IMPROVING DISASTER MANAGEMENT:
GEO-DBMS AND POSITIONING
S. Zlatanova, P. van Oosterom and E. Verbree
Delft University of Technology, Jaffalaan 9, Delft, The Netherlands, (s.zlatanova|oosterom|e.verbree @otb.tudelft.nl
Commission VII, WG VII/S
KEY WORDS: Hazards, Databases, Data Structures, Interoperability, Query, Integration, Urban, Real-time
ABSTRACT:
An increasing number of disasters (natural and man-made) with a large number of victims and significant social and economical
losses are observed in the last years. Although particular events can always be attributed to fate, it is improving the disaster
management that have to contribute to decreasing damages and ensuring proper care for citizens in affected areas. Some of the
lessons learned in the last several years give clear indications that the availability, management and presentation of geo-information
play a critical role in disaster management.
The paper discusses requirements for a system supporting *urban and urgent' disaster management. The aim of the system is to
facilitate the work of police forces, fire departments, ambulances and government coordinators in disaster situations. The (geo)
information provision addresses important aspects such as data discovery (finding the needed information based on a user location
and user profile defined on the-fly), data preparation (in a form appropriate for the requesting front-end) and data export
(considering the state-of-the-art in standardisation).
The focus of the paper is on the required developments at a database level (or middleware) to be able to provide context-aware
solutions assuming given positioning (to determine the location p
art of the context) and communication (to exchange the context-
aware information). Three important aspects are addressed: a) Integrated management of multimedia information (graphics, text,
video, raster), b) Ontology and semantic translators, c) 3D positioning to cope with emergency situations.
1. INTRODUCTION
The terrorist attacks in the US and Spain, summer floods of
2002. fires and earthquakes of 2003 in Europe have tragically
demonstrated, the whole disaster management sector needs
extended and more sophisticated means for facing man-made,
natural and industrial risks. This issue gains a high priority in
the political agenda of many governments in Europe. New
systems have to be developed that allow different service units
to operate together (to understand each other) in any critical
situation. The cooperation across different sectors, involved in
disaster management such as the Health Sector, Police and Fire
Brigade and Civil Protection, has to be extended beyond their
specific services. They have to be open to cooperate, coordinate
and understand other organisations and the information they
may provide. The final goal is limiting the number of casualties
by a) facilitating the work of the emergency services, making it
safer and more efficient, and b) ensuring citizens (in the area of
disaster and outside it) will receive high-quality care, on-time
information and instructions.
Amongst all, geo-information is becoming a key issue in the
achievement of these goals. Geo-information collections
consisting of maps, images, plans and variety of schema's are
already widely used in many of the Disaster Management
Phases, e.g. Mitigation, Preparedness, Recovery (Zlatanova
and Holweg 2004). The use of geo-information in the urgent
(time-critical) Response phase is still limited. Several major
problems can be considered:
e Most of the information available is designed, stored, and
managed by organisations that normally have distinct
authorizations. In normal circumstances these
organisations operate independently of each other. They
are only partly designed to work in a multidisciplinary
environment, and their systems reflect this status with
known limitations to their interoperability.
e QGeo-data is managed by different systems (CAD, GIS,
DBMS) in specific details, resolutions, object definition,
schemas and formats. Exchange of data is based on
creating a copy of data sets in a specific format that is
readable by the systems of the other party. Preparations of
such files may require days and storage space of several
hundreds megabytes. This manner of work is definitely not
appropriate for dealing with emergency situations.
The experience suggests that the real barriers are not lack of
data or insufficient technical capabilities. The current
development of Geo-Information Infrastructures (GII's) at
several levels (e.g. national NCGI, European INSPIRE, etc.)
based on open standards (OpenGIS, ISO TC211, OMG, W3C)
further improves this situation. The bottlenecks are in most
cases related to the ‘information’ about the information, Le.
finding the most appropriate data and making data available.
The lack of interoperability, due to the explosion of many
standards and developments, delays systems to be connected
and updated without massive investments (often unaffordable
for organisations). This results into a partial automation capable
of dealing dedicated tasks but unable to deliver intelligence to
multi-user groups.
This paper addresses the developments in geo-DBMS (with
focus on the 3D and semantic aspects) to support Response
phase in disaster management. Since providing appropriate
information in the Response phase is highly related to the
location of rescue teams and/or citizens on the field, current and
future possibilities for 3D positioning are discussed in detail.
The paper concludes with important directions for research and
development
628
Interna
Disaste
wide g
users
(relate
genera
workin
outdoo
manag
Holwe
subdiv
(contrc
(advisi
inform
centres
be ass
need i1
the im
worket
analysi
a varie
enviroi
the Int
routing
visuali
Most «
suppor
is of s
gco-in;
3.1 Si
The 11
relatio:
now .
archite
consor
(Le. Si
FC21H
inform
DBMS
Oracle
non-co
Postgri
