مدل انتخاب خروجی مبتنی بر نظریه بازی برای تخلیه

How do evacuees find their way to escape from a fire zone? This will be a significant question that should be considered for modeling the evacuation process. Generally, a building consists of enclosure areas such rooms, walkways and stairs. The principal problem in wayfinding is to select the way out when occupants egress from a multi-exit area. The choice of exits will depend on how groups of evacuees interact. Non-cooperative game theory deals largely with how intelligent individuals interact with one another in an effort to achieve their own goals—to leave the fire zone as fast as possible. This article presents a game theory based exit choice model for evacuation. It has been integrated in an evacuation model and demonstrates that the evacuees’ interaction can affect the evacuation pattern and clearance time of a multi-exit zone.

Evacuation of occupants from the hazardous region(s) is per se a way to reduce the ill effects of a fire disaster in a building and predicting evacuation pattern is useful in emergency management. Indeed, it has been considered by building maintenance professionals [1] as well as building control officers [2] that the evacuation system is the most important means for fire safety protection in buildings. In the circumstances, many evacuation models, such as EXODUS [3], SIMULEX [4] and [5], EGRESS [6] and [7], SGEM [8], [9] and [10] etc have been developed to assist building designers to predict the evacuation pattern. Most of the models have focused on modeling the flow of evacuees and the behavior of crowd flow has not been comprehensively studied, especially the behavioral reaction of the evacuees during their movement. One of the critical behavioral reactions of people that may affect the escape process is the choice of exit. Choice of exit is one of the most complex aspects of people's movement. Generally, a building consists of enclosure areas such rooms, walkways and stairs. Besides the final exits of the building, the term ‘exit’ here also refers to the openings of an area through which people can escape from one area to another inside the building. The principal problem in wayfinding is to select the way out when occupants egress from a multi-exit area. Therefore, choice of exit is the selection of routes from one point to another. In case of emergency, an individual's choice of escape route can be regarded as a wayfinding which involves perception and cognition. His or her decision will be affected by what has been seen in the environment and what has been formulated in his or her mind—the cognitive map. When several people are finding their ways to leave a hazardous zone, an individual's decision may be affected by other people's actions. In other words, interaction of people will be a process that should be considered in modeling the evacuation pattern in a zone of multi-exits. Non-cooperative game theory [11] and [12] deals largely with how intelligent individuals interact with one another in an effort to achieve their own goals—to leave the fire zone as fast as possible. This article presents a game theory-based method that can be incorporated within an evacuation model and effectively models the exit selection process in an evacuation process.

A novel approach to model the dynamic exit selection process of evacuees has been presented in this article. Evacuees perceive the actions of other evacuees and the environmental situations, and respond to their cognition to decide their escape route. In a space with high crowd density where occupants can notice the actions of other occupants, an individual will formulate his or her strategy of leaving on the basis of the crowd's actions, the travel distance to the exits, the effect of the environmental stimuli and the cognitive map (familiarity of the exits). At present, a non-cooperative game theory model has been established to examine how the rational interacting behavior of the evacuees will affect the evacuation pattern. The mixed-strategy Nash Equilibrium for the game describes the equilibrium for the evacuees and the congestion states of exits. The influence of other factors, such as the familiarity, has not been considered in this study. Nevertheless, the approach presented can rationalize the interaction of the evacuees with the environment. Further works will be required to examine the effect of familiarity and environmental stimuli as well as other ‘grouping’ effect on exit selection process.