An agent-based testbed for simulating large scale accident rescue heuristics
Abstract
The scope of this study is to present our study of the complex social problem of large scale accident rescue by applying an agent-based approach. Our field of study concerns situations involving a large number of victims over a wide area (which may or not be hostile) and where rescuers have to act rapidly to rescue the greatest number of victims in the shortest time by optimizing both their human and material resources. Based on real life observations and rescue plans on one side and designing new rescuing strategies on the other side we have built a generic and interactive user-friendly simulator. Modeling and simulation provide us with a virtual environment where we can easily develop and test a large number of "what-if" heuristic scenarios of different rescue organizations. These organizations may be compared and assessed in order to find efficient configurations and strategies for organizing a rescue.