SIMULAÇÃO DE MULTIDÕES EM SITUAÇÕES DE RISCO IMINENTE

Abstract

There are many methods of simulating a crowd of people, each with its advantages and disadvantages. The increasing number of risk situations in big cities, such as fires or terrorist attacks, made it important to study the behavior of individuals in this type of situation. The purpose of this work is to develop a mathematical-computational model to simulate crowds in fire situations in order to establish relationships between different variables found in those scenarios. This work uses the instinctive concept of crowds in situations with high levels of stress to configure the interactions between people and the environment in a dynamic multiagent system, where each individual presents different cognitive and motor capacities, so that the safety of a closed environment can be diagnosed quicker than the current methods, as well as establishing relation between the fire stage, number of people and the percentage of survivors of the group, and also evaluate the possibility of applying the agent-environment interactions for the implementation of a function optimization system.