• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.3.2013.tde-18082014-120718
Document
Author
Full name
Renata Carolina Barreiro Rodrigues
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2013
Supervisor
Committee
Mendes, André Bergsten (President)
Brinati, Marco Antonio
Ono, Rosaria
Title in Portuguese
Resolução de um problema de evacuação predial faseada.
Keywords in Portuguese
Evacuação predial faseada
GRASP
Programação inteira
Relaxação lagrangeana
Abstract in Portuguese
O trabalho apresentado nesta dissertação é referente ao estudo da evacuação de pessoas, mais especificamente, a evacuação predial faseada. O objetivo é determinar, para instâncias de até 25 andares, os instantes de liberação de cada grupo de pessoas, a fim de minimizar o tempo total de evacuação do edifício. No entanto, a determinação destes instantes deve considerar o risco ao qual os diferentes grupos estão submetidos, priorizando a evacuação do andar afetado. Além disso, os conflitos de diferentes grupos por espaço nas rotas de evacuação também devem ser evitados, já que, são nessas situações que acontecem grande parte dos acidentes. Para atingir tal objetivo, foi elaborado um modelo matemático de programação linear inteira. Devido à alta complexidade do modelo, fez-se necessária a aplicação de métodos heurísticos para a obtenção de soluções. Dessa maneira, foram desenvolvidas uma heurística de busca baseada em GRASP e uma heurística lagrangeana. Apesar da heurística lagrangeana atestar a qualidade da solução (a partir da comparação do resultado obtido com o limitante inferior), a heurística de busca mostrou-se mais adequada para o problema, pois forneceu resultados de qualidade com pouco esforço computacional.
Title in English
Solving a problem of building phased evacuation.
Keywords in English
Building phased evacuation
GRASP
Integer programming
Lagrangean relaxation
Abstract in English
This dissertation studies the evacuation of people, more specifically, building phased evacuation. The objective of this study is to determine, for buildings of up to 25 floors, in which instants each group of people has to be released, in order to minimize the total evacuation time. Furthermore, the determination of these instants has to consider the risk to which each group of people is submitted, thus the affected floor has to be the first group to be released. In addition, conflicts for space between groups should be avoided, since such situations increase the occurrences of accidents. To achieve this goal, an integer linear programming model was designed. Due to the high complexity of the model, it was necessary to apply heuristics to obtain solutions for some instances. Therefore, a search heuristic based on GRASP and a lagrangian heuristic were developed. Despite the fact that the lagrangian heuristic attests to the quality of the solution (when it is compared to the lower bound), the search heuristic was considered more suitable for this problem because it provided quality results with lower computational efforts.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2014-09-01
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.