Towards planning for Emergency Activities in LargeScale Accidents An Interactive and Generic AgentBa

1
Towards planning for Emergency Activities in
Large-Scale Accidents An Interactive and Generic
Agent-Based simulator
Narjès Bellamine-Ben Saoud
Julie Dugdale Bernard Pavard GRIC-IRIT France

• Mohamed Ben Ahmed
• Tarek Ben Mna,Néjia Ben Touati
• RIADI-GDL Laboratory Tunisia

2
Large Scale Accidents
An accident is an unfortunate, harmful event,
caused unintentionally where one or many persons
become victims
DISASTER

• Many civilians victims

• Hostile Environnement

NEED
Rapid HELP Minimize disaster damage

• Optimal RESCUING PLANS strategies

3
MOTIVATIONS

• Model to understand rescuing process and plan
• Assist rescuing decision making ?
• Off-line
• design new /adapted/ adequate organization
  rescue plans
• Real time
• manage resources support real time decisions
• Design new collaborative situations
• Provide realistic software training tools

4
ISSUE

• PROVIDE a virtual platform which is
• Realistic
• modeling real components and behaviors
• Flexible - Easy to use
• To provide rapid testing of various scenarios
  (real and hypothetical what if scenarios)
• modeling simulating
• Agent based Generic Interactive Simulator

5
Resources used

• Written documentation
• P. Ecollan. Mise en oeuvre de soins médicaux
  immédiats en presence dun nombre important de
  victims à Paris  LE PLAN BLANC, Thèse en
  médecine 1989
• CDSP91 et SAMU 91. Plan destiné à porter secours
  à plusieurs victimes  PLAN ROUGE , Rapport,
  Novembre 2000
• Observation and analysis of Real Simulations and
  exercices of the RED plan

6
Rescue is a complex situation

• where various agents (victims, doctors, fireman,
  policeman, ) with heterogeneous competences and
  having various roles, organize themselves
  dynamically on teams, adapt their behaviour
  (individual, collaborative, walking, searching,
  hearing, treating), process continuously all
  received information and react in an
  unpredictable way to their environment.

Their main objectives are rescuing victims,
performing better evacuation results and mainly
reducing delays and minimizing number of dead
victims.
7
Model of evacuation and sort center
8
Emergency activities rescuing
Disaster rescue is one of the most serious social
issues involving

• Heterogenous agents
• Victims
• Rescuers
• Medical Team Doctors, nurses
• Fire Brigade
• Police Force

• On site rescue activities
• Searching for people and evaluation of sanitary
  needs
• Participation in recognizing
• victims lifting
• their provision with medical care
• Medical sort and victims orientation
• Evacuation and transport of victims
• Elaboration of retrospective medical Summary

• Strategies
• Treat first
• Evacuate first

9
Simulator architecture

• Actors hierarchy

Actors
10
main components of our model

• autonomous Agents representing victims with
  evolving illness and rescuers (doctors, nurses,
  fireman) moving on site, perceiving victims,
  collaborating to rescue the first ones
• Environment representing the accident site
  having obstacles and dangerous areas and where
  the victims are initially spread and the doctors
  move to explore perceive - treat and helpers
  evacuate
• Interactions between rescuers exploring
  collectively, evacuating by pairs, communicating
  directly or via artefacts (paper forms or
  electronic)
• Organization of actors as distributed
  independent sub-teams in various site
  sub-zones or as a centralized whole team
  conducted by the rescue chief
• User interfaces allowing mainly initial
  configuration of the simulations (e.g. number of
  victims and states, followed strategies, rescuers
  behaviours), continuous visual control of the
  process of rescuing (e.g. site overview with
  acting-interacting agents, graphics, text
  descriptions), dynamic changes of parameters of
  an on-going simulation (e.g. adding new victims,
  adding new rescuers, or adding dangerous zones or
  new obstacles on sites) as well as step-by-step
  simulation.

11
Modeling victims

• Victims Gravity Scale
• According to White Plan the following scale is
  adopted
• 1 Extremely serious
• 2 Very Serious
• 3 Serious
• 4 Not Serious
• By extention, 0 dead
• Victims gravity evolution
• using probabilities of transition from one state
  to the other

12
Sample of Medical Form used for evacuated victims
Diagnostic précis avant triage
Heures
Traitement mis en oeuvre
13
Victims evolution
Propabilities in Normal Areas
14
Simulator Input-Output
BD
Fichiers
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20

• Genericity
• Rescue strategies centralised or distributed
• Agents behavior explore, treat, evacuate
• Communication modes and artefacts used direct
  or mediated, paper form or Medicard LAN
• For environment properties obstacles and
  dangerous sectors
• Interactivity
• Step by step simulation
• Changing accident site adding obstacles and/or
  damages
• Change status of agents rescuers becoming
  victims
• Add new agents

21
CONCLUSIONS

• A virtual collaboration environment with
  autonomous agents evolving in a dynamic
  environment
• Where we can evaluate (run and compare) different
  cooperative rescue scenarios
• study the activities of the group (under various
  strategies and rescue behaviours)
• to assess the best configuration (number of
  doctors, resources, etc) for better rescue
  results
• to assess the use of different communication
  devices (mainly cooperation between rescuers
  using traditional or electronic communication
  artefacts)

22
Ongoing and Future work

• Hospitals
• Ambulances
• Routes
• Traffic simulation
• - GIS connection

Simulation extented to a whole city
Virtual reality

• 3D Context and  natural  interactivity
• Interoperability

Decision support system
Training
Planning