Research And Simulation Of Individual Behavior Decision Model On Spatial Effect

With the development of urban modernization,large buildings,large shopping malls,cinemas,music venues and other relatively closed space gradually increased,the population density is growing.These relatively closed spaces have the characteristics of complex structure,lack of light,limited exports,large numbers of personnel and diverse staffing.In such a space,staff would be under huge pressure in physiology and psychology.The crowd led to the frequent occurrence of accidents in recent years,which has caused a serious threat to people's lives and social property.Stampede is largely the result of the massive crowd behavior is out of control in public places.So the study of large-scale swarm behavior in public places has become the hot spot.Based on the relatively closed space for research scene,on the basis of social force model.The paper studies the individual behavior decision-making mechanism from the perspective of microcosmic individual,focus on the scene familiarity and emotional factors impact on individual behavior decision-making and the whole crowd behavior,use mathematical model to describe the influence of emotional factors on individuals and add this model to the social force model.As a result,the improved social force model containing emotion factor is constructed.The crowd behavior simulation platform is developed by using the Repast simulation toolkit.Simulation experiments are designed on the platform and the subway platform emergency evacuation is simulated.The simulation shows that some individuals in the group have the characteristics of conformity,clustering and so on.It also indicates that mood has an impact on group behavior and group behavior under the influence of the mood is more random.The subway platform at the same time the actual filming of the cluster phenomenon has carried on the simulation and comparison.The results show that the improved model can simulate the behavior of the group in the actual situation and verify the rationality of the model.