Study On The Cellular Automaton Simulation Of The Escaping Pedestrian Flow In Building

In recent years, the growth of population has brought us many social problems, in which thejam of pedestrian flow is catching more and more attentions of people. The jam of pedestrians cancause many accidents, and thus brings large losing to our lives. So to study the theory ofpedestrian flow is very urgent now. Pedestrian flow is a kind of many-body systems of interacting people, which has the similarcharacteristics of traffic flow. For example, when the density of system increasing, complexbehaviours such as self-organizations and jamming transitions would occur. In order to study anddescribe these phenomena, many models have been proposed by people. These models cansimulate the pedestrian flow from different points of view. In this article, some representativemodels, such as hydromechanical model, social force model, lattice gas model of biased-randomwalkers and cellular automata model of pedestrian flow are introduced. Comparison betweenthese models is also presented. The main contribution of this article lies in four aspects. First, a lattice gas model of biased-random walkers is used to simulate the escapingpedestrian flow in a corridor. The relationships between transition time and the parameters ofsystem sizes are discussed. It is found that the transition time tc closely depends on the width ofcorridor W and the strength of drift D , it scales as tc ∝ W ?0.85±0.04, and tc ∝ D . However,the widths of doors and the intervals of doors have little influence on the transition time, but makedifference on the flow rate at the jamming state. Second, a multi-speeds cellular automata model of pedestrian flow is used to simulate thesame process of evacuation as above-mentioned. The relationships between transition time and theparameters of system sizes are discussed again with this model. The same conclusions asabove-mentioned are also found. Furthermore, the relationship between the transition time tc IIIand the maximal speed Vmax is found as tc ∝ Vmax ?α . Third, by adapting the method of calculating the transition probabilities in the lattice gasmodel, and the rules of collision avoidance in the cellular automata model, a new cellularautomata model of pedestrian flow is presented to simulate the escaping flow of crowd outside aroom. It is found that the width of door is the most important parameter that influent the transitiontime. The patterns of pedestrians are also found by simulation, which seems to be more realisticthan that of the research of Tajima et al. Fourth, the above-mentioned cellular automata model is used to simulate the process ofevacuation in a building. The building is made up of one corridor and several rooms. It is foundthat the parameters which make great influence on the transition time tc is not the width of doors,but the width of corridor and the maximal velocity of pedestrian. It is also found that the timepeople cost to leave the room is not the same in different rooms. In the room most far away fromthe exit of corridor, people cost the least time to leave that room. However, in the other rooms, thetime people cost to leave is decrease with the distance between the door of room and the exit. Italso shows that to optimize the distribution of the building, and try to use the longer corridor isgood to the evacuation of pedestrian.