Research On Indoor Evacuation Behavior Based On Cellular Automata And Game Theory

In recent years,the number of large buildings at home and abroad has increased sharply,with frequent accidents.The safety of personnel in buildings has attracted considerable attention.Over75% of the world’s population lives in cities and towns,and approximately 90% of their time is spent indoors.Emergency situations such as fires 、 earthquakes 、 and gas leaks can have dire consequences in the indoor environment.Therefore,studying the efficient evacuation of indoor personnel is crucial to saving lives and ensuring personnel safety.In response,this paper focuses on using cellular automata models and game theory to study the self-organization behavior of indoor personnel and the evacuation problem caused by the competitive behavior of evacuees.The main work is as follows:Firstly,starting from the formation mechanism of personnel evacuation,two main behaviors were defined: internal self-organization and external control intervention.Collect emergency evacuation videos from various sources both domestically and internationally,with a focus on analyzing the evacuation characteristics exhibited by these two behaviors when real dangerous events occur.The analysis shows that:(1)In real emergencies,evacuees mainly show four kinds of self organizational behavior: conformity behavior,cooperation/competition behavior,clustering behavior,impulsive fluke behavior,and leader follower intervention behavior,of which the proportion of conformity behavior under panic psychology is as high as 60%.(2)The self-organized conformity and competitive behavior of evacuees have caused congestion at the escape exits,and unreasonable personnel exit selection has occurred during the escape process.(3)Controlling intervention behavior greatly affects personnel’s exit selection and evacuation efficiency.Secondly,the self-organizational behavior of personnel is studied in detail.A field evacuation model for indoor personnel based on cellular automata was established,introducing inertial effects and wall fields.The research shows that:(1)as the population density increases,the clustering behavior of evacuees gradually strengthens,and the time required for escape increases.(2)Reasonable self-organizing conformity behavior can help evacuees find escape routes,but excessive conformity behavior has certain negative effects on evacuation.(3)Indoor exit obstacles can hinder personnel evacuation,and compared to horizontally placed obstacles,vertical obstacles have higher evacuation efficiency.Then,aiming at the phenomenon of unreasonable export selection caused by competitive behavior in the process of evacuation,a model of indoor export selection combining game theory and cellular automata is proposed.The research shows that:(1)The game theory indoor exit model can achieve the evacuation of people in the room more efficiently.The patient level of export selection has a certain impact on personnel’s export selection behavior,and the degree of influence continuously increases with the increase of the number of people.(2)For the selection of individual indoor building exits,when the exit distribution is located in the middle of the wall,the evacuation time for personnel is the smallest and the evacuation efficiency is the highest.(3)For the case of two exits,when the two exits are located on the same wall,the distance between the exits should be0.4～0.5 times the length of the wall where they are located,and the evacuation effect of personnel is the best.When the exit is located on different wall surfaces,whether it is a long wall or a short wall,the symmetrical distribution of the exit structure significantly improves its evacuation efficiency.Finally,the indoor personnel evacuation model was optimized,and external human intervention measures were used to guide personnel evacuation behavior.A cellular automaton model for indoor leaders was proposed to simulate the evacuation dynamics under leader follower behavior.Research has found that:(1)the presence of leaders in single and double exit indoor rooms can accelerate the evacuation process,but having too many leaders does not have a significant impact on crowd evacuation.A small number of leaders can guide personnel to evacuate.(2)For rooms with multiple exits,when leaders randomly exist,some of them cause pedestrians to move towards the same exit,resulting in overcrowding and insufficient utilization of the exits.(3)Fixed leader positions have a better effect on evacuation compared to random leader positions.That is to say,the position planning of leaders is more important than their quantity.This result provides new insights for leaders on the impact of crowd evacuation.This article mainly simulates evacuation behavior using cellular automata models and game theory methods,and draws some meaningful conclusions.These conclusions are helpful for understanding the mechanism of individual behavior,and can also provide scientific theoretical basis and technical support for formulating emergency strategies for indoor locations.