Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Evacuation planning optimization based on a multi-objective artificial bee colony algorithm

Niyomubyeyi, Olive LU ; Pilesjö, Petter LU and Mansourian, Ali LU (2019) In ISPRS International Journal of Geo-Information 8(3).
Abstract

Evacuation is an important activity for reducing the number of casualties and amount of damage in disaster management. Evacuation planning is tackled as a spatial optimization problem. The decision-making process for evacuation involves high uncertainty, conflicting objectives, and spatial constraints. This study presents a Multi-Objective Artificial Bee Colony (MOABC) algorithm, modified to provide a better solution to the evacuation problem. The new approach combines random swap and random insertion methods for neighborhood search, the two-point crossover operator, and the Pareto-based method. For evacuation planning, two objective functions were considered to minimize the total... (More)

Evacuation is an important activity for reducing the number of casualties and amount of damage in disaster management. Evacuation planning is tackled as a spatial optimization problem. The decision-making process for evacuation involves high uncertainty, conflicting objectives, and spatial constraints. This study presents a Multi-Objective Artificial Bee Colony (MOABC) algorithm, modified to provide a better solution to the evacuation problem. The new approach combines random swap and random insertion methods for neighborhood search, the two-point crossover operator, and the Pareto-based method. For evacuation planning, two objective functions were considered to minimize the total traveling distance from an affected area to shelters and to minimize the overload capacity of shelters. The developed model was tested on real data from the city of Kigali, Rwanda. From computational results, the proposed model obtained a minimum fitness value of 5.80 for capacity function and 8.72 × 10 8 for distance function, within 161 s of execution time. Additionally, in this research we compare the proposed algorithm with Non-Dominated Sorting Genetic Algorithm II and the existing Multi-Objective Artificial Bee Colony algorithm. The experimental results show that the proposed MOABC outperforms the current methods both in terms of computational time and better solutions with minimum fitness values. Therefore, developing MOABC is recommended for applications such as evacuation planning, where a fast-running and efficient model is needed.

(Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Evacuation planning, Geographic information system (GIS), Multi-objective artificial bee colony, Spatial optimization, Swarm intelligence, Geospatial Artificial Intelligence (GeoAI), Artificial Intelligence (AI), Operational research
in
ISPRS International Journal of Geo-Information
volume
8
issue
3
article number
110
pages
23 pages
publisher
MDPI AG
external identifiers
  • scopus:85063689122
ISSN
2220-9964
DOI
10.3390/ijgi8030110
language
English
LU publication?
yes
id
d4cdeac4-f65e-4137-a520-c4533dc70f9b
date added to LUP
2019-04-07 10:35:37
date last changed
2023-10-07 00:05:53
@article{d4cdeac4-f65e-4137-a520-c4533dc70f9b,
  abstract     = {{<p>                                                         Evacuation is an important activity for reducing the number of casualties and amount of damage in disaster management. Evacuation planning is tackled as a spatial optimization problem. The decision-making process for evacuation involves high uncertainty, conflicting objectives, and spatial constraints. This study presents a Multi-Objective Artificial Bee Colony (MOABC) algorithm, modified to provide a better solution to the evacuation problem. The new approach combines random swap and random insertion methods for neighborhood search, the two-point crossover operator, and the Pareto-based method. For evacuation planning, two objective functions were considered to minimize the total traveling distance from an affected area to shelters and to minimize the overload capacity of shelters. The developed model was tested on real data from the city of Kigali, Rwanda. From computational results, the proposed model obtained a minimum fitness value of 5.80 for capacity function and 8.72 × 10                                                         <sup>8</sup>                                                          for distance function, within 161 s of execution time. Additionally, in this research we compare the proposed algorithm with Non-Dominated Sorting Genetic Algorithm II and the existing Multi-Objective Artificial Bee Colony algorithm. The experimental results show that the proposed MOABC outperforms the current methods both in terms of computational time and better solutions with minimum fitness values. Therefore, developing MOABC is recommended for applications such as evacuation planning, where a fast-running and efficient model is needed.                                                 </p>}},
  author       = {{Niyomubyeyi, Olive and Pilesjö, Petter and Mansourian, Ali}},
  issn         = {{2220-9964}},
  keywords     = {{Evacuation planning; Geographic information system (GIS); Multi-objective artificial bee colony; Spatial optimization; Swarm intelligence; Geospatial Artificial Intelligence (GeoAI); Artificial Intelligence (AI); Operational research}},
  language     = {{eng}},
  month        = {{03}},
  number       = {{3}},
  publisher    = {{MDPI AG}},
  series       = {{ISPRS International Journal of Geo-Information}},
  title        = {{Evacuation planning optimization based on a multi-objective artificial bee colony algorithm}},
  url          = {{http://dx.doi.org/10.3390/ijgi8030110}},
  doi          = {{10.3390/ijgi8030110}},
  volume       = {{8}},
  year         = {{2019}},
}