Simulation of Pedestrian Evacuation in University Canteen Based on Cellular Automata

Abstract

In this study, we firstly build up an exit choice model to describe the pedestrians' exit choice behavior in a specific place, i.e., university canteens. The pedestrians are assumed to minimize their evacuation time in choosing exits by explicitly considering the estimated evacuation time and effects of pedestrians' destinations. In addition, the behavior of food tray return is also captured in the proposed route choice model for pedestrians. Fusing the exit choice model and the route choice model into a cellular automata (CA)based pedestrian simulation model, we simulate the pedestrian evacuation process in university canteen and the results illustrate the effectiveness of the modeling framework in terms of describing the evacuation process in university canteens. Specifically, destination is identified as a key influence factor for pedestrian evacuation, and pedestrians tend to select the exit that is closer to their destinations. If it is required to return the food tray to designated spots before leaving the canteen, the evacuation efficiency is significantly reduced. The simulation result can also assist in determining the optimal number of tray-return spots if maximum evacuation efficient is to be achieved. In presence of the specific parameter settings and canteen layout in the case study example, it is recommended to deploy four tray-return spots in the university canteen.INDEX TERMS University canteen, exit choice, route choice, pedestrian evacuation, cellular automata.

Human Behavior During Emergency Evacuation: Cell Transmission Model

“…There are two categories for pedestrian evacuation research, i.e., empirical methods [1][2][3][4][5][6][7][8][9][10][11][12][13] and simulation methods [14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32]. The results of the experiments can provide the basic data for the prediction of pedestrian evacuation.…”

“…In the simulation process of microscopic model, each pedestrian is regarded as a discrete individual. Typical representatives of the microscopic model are the social force, lattice gas, and cellular automaton models [19][20][21][22][23][24][25][26][27][28][29][30][31][32]. In the social force model, the pedestrians can be considered as structureless particles and the movement is based on Newtonian mechanics.…”

“…Kirchner et al [26,27] suggested the concept of static floor field and dynamic floor field in the cellular automaton model. In addition, a variety of extended cellular automaton models for pedestrian evacuation have been proposed, such as Zheng et al [28], Guo et al [29], Zhao et al [30], Liu et al [31], and Li et al [32].…”

Direction-Decision Learning Based Pedestrian Flow Behavior Investigation

“…The most popular discrete models used in recent research include the lattice gas model (LGM) [4], [5] and the cellular automata model (CAM) [6]- [9], in which the space is discretized to approximate real pedestrian movement. The evolution of the pedestrians in time is determined by physical and social laws that describe the interaction among the particles as well as their interactions with the physical surroundings.…”

“…Similarly, the heuristic method was also proposed to compute the desired walking direction, which was a trade-off between avoiding obstacles and minimizing detours from the most direct route [28], [29]. With the development of artificial intelligence, the machine learning method has been used to simulate pedestrian flow due to the development of pedestrian detection [9]. For example, a multiagent reinforcement learning-based framework was proposed to simulate pedestrian groups [30], and a data-driven neural network approach and an artificial intelligence-based approach were proposed to simulate the pedestrian flow [16], [31].…”