Genetic algorithm-based metacellular automata model for evacuation from complex buildings
[Objective]The metacellular automata evacuation model is widely used owing to its simplicity and efficiency.However,current research indicates that the behavioral characteristics of people during evacuation and the effects of interactions between individuals and the environment are not well understood.In particular,research on evacuating crowded areas in complex buildings and on the equalization of evacuation exits is still limited.Efficiently evacuating people from complex buildings has become a critical issue.This study presents an optimization model based on cellular automata theory for personnel evacuation in complex buildings.The model aims to minimize evacuation time by considering various factors such as evacuation paths,exit configurations,individual characteristics,and crowd behavior.The simulation is solved using a genetic algorithm fitness function.[Methods]We create a matrix model of the physical environment floor plan based on the two-dimensional layout of the Air Traffic Control Experiment Center at the Civil Aviation University of China.The entire map is divided into 39,200 metric cells to collectively form the evacuation area.Evacuees are categorized into four groups according to their behavioral characteristics:blind following,exit aware,environment aware,and experienced.The model considers various factors,including evacuation paths,exit arrangements,crowd density,and individual behaviors,to optimize the evacuation process in complex buildings.This study analyzes the impact of three optimization strategies on the evacuation process:guiding exit selection,guiding the use of connecting corridors,and applying a genetic algorithm-based fitness function.Furthermore,the results of this model are compared with those of the simulation software Pathfinder to verify the effectiveness of the model.[Results]The evacuation outcomes are analyzed under various experimental schemes from four perspectives:total evacuation time,extreme congestion areas,accuracy error in evacuation time between the cellular automata model and Pathfinder,and exit utilization balance.The results show that compared with the initial scheme,the optimized evacuation model reduces the total evacuation time by 64.4％,reduces extreme congestion areas by 50.7％,and improves the accuracy error in evacuation time between the cellular automata model and Pathfinder by 2.2％.In addition,the model reduces the standard deviation of evacuees at the three exits from 47.14 to 7.75,considerably improving the balance in exit utilization.The experimental findings demonstrate that the optimized evacuation model proposed in this study effectively enhances overall evacuation efficiency.[Conclusions]To optimize the evacuation of individuals from complex buildings,an evacuation model based on cellular automata is constructed and enhanced through the optimization of a genetic algorithm's fitness function.This optimization results in substantial reductions in total evacuation time and extreme congestion areas.In addition,the optimization leads to a balanced utilization of exits,considerably improving overall evacuation effectiveness.The optimized model successfully achieves its intended objectives,providing valuable insights for evacuating individuals in complex building structures.
NETL
NSTL
万方数据
