Advanced

Dynamic urban land-use change management using multi-objective evolutionary algorithms

Masoumi, Zohreh ; Coello Coello, Carlos A. and Mansourian, Ali LU (2019) In Soft Computing
Abstract

Frequent land-use changes in urban areas require an efficient and dynamic approach to reform and update detailed plans by re-arrangement of surrounding land-uses in case of change in one or several urban land-uses. However, re-arrangement of land-uses is problematic, since a variety of conflicting criteria must be considered and satisfied. This paper proposes and examines a two-step approach to resolve the issue. The first step adopts a multi-objective optimization technique to obtain an optimal arrangement of surrounding land-uses in case of change in one or several urban land-uses, whereas the second step uses clustering analysis to produce appropriate solutions for decision makers from the outputs of the first step. To present and... (More)

Frequent land-use changes in urban areas require an efficient and dynamic approach to reform and update detailed plans by re-arrangement of surrounding land-uses in case of change in one or several urban land-uses. However, re-arrangement of land-uses is problematic, since a variety of conflicting criteria must be considered and satisfied. This paper proposes and examines a two-step approach to resolve the issue. The first step adopts a multi-objective optimization technique to obtain an optimal arrangement of surrounding land-uses in case of change in one or several urban land-uses, whereas the second step uses clustering analysis to produce appropriate solutions for decision makers from the outputs of the first step. To present and assess the approach, a case study was conducted in Tehran, the capital of Iran. To satisfy the first step, four conflicting objective functions including maximization of consistency, maximization of dependency, maximization of suitability and maximization of compactness were defined and optimized using non-dominated sorting genetic algorithm. Per-capita demand was also employed as a constraint in the optimization process. Clustering analysis based on ant colony optimization was used to satisfy the second step. The results of the optimization were satisfactory both from a convergence and from a repeatability point of view. Furthermore, the objective functions of optimized arrangements were better than existing land-use arrangement in the area, with the per-capita demand deficiency significantly compensated. The approach was also communicated to urban planners in order to assess its usefulness. In conclusion, the proposed approach can extensively support and facilitate decision making of urban planners and policy makers in reforming and updating existing detailed plans after land-use changes.

(Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
epub
subject
keywords
Clustering, Decision support, Dynamic urban land-use change, Multi-objective optimization, NSGA-II, Soft computing
in
Soft Computing
pages
26 pages
publisher
Elsevier
external identifiers
  • scopus:85069749776
ISSN
1432-7643
DOI
10.1007/s00500-019-04182-1
language
English
LU publication?
yes
id
8f818da1-d1df-4f4c-9982-e11409ab043b
date added to LUP
2019-08-07 12:03:58
date last changed
2020-01-22 07:44:41
@article{8f818da1-d1df-4f4c-9982-e11409ab043b,
  abstract     = {<p>Frequent land-use changes in urban areas require an efficient and dynamic approach to reform and update detailed plans by re-arrangement of surrounding land-uses in case of change in one or several urban land-uses. However, re-arrangement of land-uses is problematic, since a variety of conflicting criteria must be considered and satisfied. This paper proposes and examines a two-step approach to resolve the issue. The first step adopts a multi-objective optimization technique to obtain an optimal arrangement of surrounding land-uses in case of change in one or several urban land-uses, whereas the second step uses clustering analysis to produce appropriate solutions for decision makers from the outputs of the first step. To present and assess the approach, a case study was conducted in Tehran, the capital of Iran. To satisfy the first step, four conflicting objective functions including maximization of consistency, maximization of dependency, maximization of suitability and maximization of compactness were defined and optimized using non-dominated sorting genetic algorithm. Per-capita demand was also employed as a constraint in the optimization process. Clustering analysis based on ant colony optimization was used to satisfy the second step. The results of the optimization were satisfactory both from a convergence and from a repeatability point of view. Furthermore, the objective functions of optimized arrangements were better than existing land-use arrangement in the area, with the per-capita demand deficiency significantly compensated. The approach was also communicated to urban planners in order to assess its usefulness. In conclusion, the proposed approach can extensively support and facilitate decision making of urban planners and policy makers in reforming and updating existing detailed plans after land-use changes.</p>},
  author       = {Masoumi, Zohreh and Coello Coello, Carlos A. and Mansourian, Ali},
  issn         = {1432-7643},
  language     = {eng},
  month        = {07},
  publisher    = {Elsevier},
  series       = {Soft Computing},
  title        = {Dynamic urban land-use change management using multi-objective evolutionary algorithms},
  url          = {http://dx.doi.org/10.1007/s00500-019-04182-1},
  doi          = {10.1007/s00500-019-04182-1},
  year         = {2019},
}