A mesoscopic approach to modeling and simulation of pedestrian traffic flows

The paper discusses the simulation of pedestrian traffic flows in public buildings from a practical point of view: The models developed are intended for use in an operative early-warning system. They must be able to forecast how the situation will evolve in each room of the building for the near future. By evaluating possible measures they can be used as a tool that pro-actively avoids critical system states. A mesoscopic approach to creating these models is described and partially realized. It envisions groups of persons as modeling and simulation objects and thereby distinguishes between physical and logical (virtual) groups. The paper presents novel solutions to the modeling of both types of person groups. An example of a model described in XML and implemented with Java is demonstrated.