Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.13091/880
Title: A novel metaheuristic algorithm by efficient crossover operator for land readjustment
Authors: Koç, İsmail
Çay, Tayfun
Babaoğlu, İsmail
Keywords: Subdivision
Land Readjustment
Land Reallocation
Zoning Application
Metaheuristic Algorithms
Intelligent Parcel-Based Crossover
Expert System
Differential Evolution
Optimization
Issue Date: 2022
Publisher: PERGAMON-ELSEVIER SCIENCE LTD
Abstract: Land readjustment and reallocation (LR) applications are complex and difficult real-world problems involving many different criteria. By considering these criteria, it is very difficult and takes a long time to be solved manually by an expert. Since the search space of these problems is very large, solution of these problems requires meta-heuristic optimization algorithms instead of classical methods in order to acquire more robust, acceptable and qualified solutions. Considering the meta-heuristic approaches, the algorithm needs an objective function that can make the right decision and evaluate the solutions most reasonably among the candidate solutions. Using the proposed objective function, the quality of the distribution and subdivision plans will be automatically evaluated and compared without the need for an expert. In this study, an objective function which considers all the criteria in the LR problems is proposed. In addition, unlike the available crossover operators used in metaheuristic algorithms in the literature, two different parcel-based crossover operators called Classical (CPC) and Intelligent (IPC) Parcel-Based Crossover Operators are proposed. While CPC performs the distribution of the owners to the predetermined parcel randomly, IPC makes this operation with a greedy approach rather than randomly. According to this approach, if the shareholder and distance values after the crossover operation would be better than the existing ones, the crossover operation is performed. Otherwise, this operation is cancelled. By using the proposed objective function and crossover operators, artificial bee colony (ABC), particle swarm optimization (PSO) and differential evolution (DE) algorithms are run under equal conditions on a real project site, and the obtained results are compared with the official results obtained by a technician in the study. In addition, since there will be so many zoning blocks of different sizes and shapes on a real project site, it is very possible to have gaps or overflows in the blocks of subdivision plans obtained from the algorithms. Therefore, the gaps and overflow areas in the blocks can be completely eliminated by utilizing an Expert System developed specifically for LR problems called LRES, and as a result, the solutions obtained from the algorithms can be directly applicable in real life by the LRES. It's clearly seen from the experimental studies that all of the results obtained by using the algorithms based on LRES are much more effective than the official results obtained by a technician in terms of both solution quality and speed. In addition, among the evaluated algorithms, it is observed that the PSO algorithm presents much more effective and robust results than results of the other algorithms. Moreover, as a consequence of the algorithms using the IPC presents much more successful results than the results of the algorithms using CPC, it can be used as a very effective alternative crossover operator for land use problems.
URI: https://doi.org/10.1016/j.eswa.2021.116082
https://hdl.handle.net/20.500.13091/880
ISSN: 0957-4174
1873-6793
Appears in Collections:Mühendislik ve Doğa Bilimleri Fakültesi Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collections
WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collections

Files in This Item:
File SizeFormat 
1-s2.0-S0957417421014196-main.pdf
  Until 2030-01-01
5.26 MBAdobe PDFView/Open    Request a copy
Show full item record

CORE Recommender

SCOPUSTM   
Citations

1
checked on Jan 28, 2023

WEB OF SCIENCETM
Citations

1
checked on Jan 30, 2023

Page view(s)

48
checked on Jan 30, 2023

Google ScholarTM

Check

Altmetric


Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.