Cybernetic Incident Management
duration: 01/2003 - 12/2007 /// subsidy: TSIT program of Senter Novum)
Big incidents like the planecrash in Amsterdam Bijlmer or the fireworkdisaster in Enschede showed us that many cities don't have an efficient disasterplan. Designing and updating a disasterplan is a very complex activity that doesn't fit directly into the range of administrative activities of municipalities or other governmental institutions.
Human interaction plays a keyrole in calamity suppression and can be seen as a dynamic system. Humans can experience some events during a reallife disaster as completely unexpected. The problem with a disasterplan is that it is a plan on paper, not a tried and tested plan.
The main goal of the CIM project is to shape the plan on paper into an active structure. The specific challenge which calamities put to the administrative apparatus are:
• a calamity might take place, but nobody knows when and how
• no scenario can cover completely what happens during a calamity
• it's very expensive and complex to invest in all operating municipal care services so help has to be provided from national careservices, which brings extra logistic and coordinating problems.
The CIM project aimed at generating sufficient knowledge to allow the industrial partners of the project to build the first generation of an intelligent and adaptive agent-based ICT-infrastructure for incident management, within 5 years from the start of the project,. The agent technology used should also allow simulation of the dynamic and often non-hierarchical nature of disaster management by the various involved organizations.
In CIM, Almende introduced representing software agents to support people in reacting quickly and in setting up the right communication between people. The interaction between people and software agents will also be used for training (real-life or simulated), which is used to increase effectiveness of protocols and efficiency of communication in case of a real incident.
Knowledge in the system is contained in the communication structures and the supporting software in the form of distributed agents that are able to obtain and weigh information dynamically. The maintenance and evolution of the system is achieved by performing simulations and training sessions, both virtually as well as real-life. Measuring the effectiveness of a response and using this as feedback can improve the quality of the protocols and the system itself.
D.N. Ferro and A. H. Salden, Self-organizing mobile surveillance security networks, The 2nd International Conference on Bio Inspired mOdels of NEtwork, Information and Computing Systems BIONETICS 2007, Budapest, Hungary, December 10 - 13, 2007.
A.H. Salden and D.N. Ferro, Self-Organized Critical Networks, International Conference on Complex Systems, October 28-November 2, 2007.
D. N. Ferro, A.H. Salden and J. M. Valk, A Robust Coalition Formation Framework for Mobile Surveillance Incident Management, In: Proceeding of the 4th Intelligent Human Computer Systems for Crisis Response and Management ISCRAM'07 (Eds. B. Van de Walle, P. Burghardt and C. Nieuwenhuis), 2007, pp. 479-488.
J.M. Valk and D. N. Ferro, A generalized matching framework, combining matchmaking and coalition formation, In: Proceedings of the 18th Belgian-Netherlands Conference on Artificial Intelligence BNAIC‘06(Eds. P. Schobben, W. Vanhoof and G. Schwanen), 2006, pp. 315–322.