Numerical analysis focused on each agent´s moving on refuge under congestion circumstances by using cellular automata

The objective of this paper is studying about people´s behavior on evacuation under congestion circumstances. In the event of natural disasters such as earthquakes, typhoon hits, fire, and so on, people try to take refuge to safe place. Under these circumstances, if congestion or accidents by panic occur, it may take time for people to get out of dangerous spaces depending on the degree of the congestion. In underground shopping areas or in skyscrapers, taking evacuation time causes higher risk of death. It is necessary to achieve efficient refuge taking account of various conditions and possible situations. In this study, the authors suggest a new computing process of simulation to focus on how people move in the evacuating situation. This new process proposes how people behave in or near the congestion. Furthermore, we see the validity of the simulation and the results of analysis.