Optimization of building land parcellation using a genetic algorithm
Tehnika · 2026 · 81 (5) · 441-449
DOI: 10.5937/tehnika2605441i ↗Kurz gefasst
Fachartikel in der Zeitschrift Tehnika, 2026. Verfasst von Aleksa Isailović und 4 weiteren Autor:innen. Die Arbeit nennt 16 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This paper presents the theoretical framework and a demonstration of Genetic Algorithm (GA) application for the optimization of urban land subdivision. The problem is formulated as a combinatorial optimization model with two discrete variables: the number of columns (n) and rows (k) of the parcellation grid. The objective function maximizes either the total number of building plots or the Gross Floor Area (GFA), while strictly adhering to urban planning constraints, including minimum plot area, frontage, and depth. The implementation of a penalty function based on Deb's approach ensures high robustness and the feasibility of solutions within the search space. Model testing on four typologically diverse blocks confirmed the operational efficiency of the algorithm and rapid convergence, typically achieved within the early evolutionary stages (between the 5th and 20th generation). The proposed model serves as a methodological foundation for the development of advanced automated parcellation systems based on real-world geospatial data.
Zitieren
APA
Isailović, A., Stanić, M., Dašić, T., Šoškić, M., & Višnjevac, N. (2026). Optimization of building land parcellation using a genetic algorithm. Tehnika, 81(5), 441-449. https://doi.org/10.5937/tehnika2605441i
Vancouver
Isailović A, Stanić M, Dašić T, Šoškić M, Višnjevac N. Optimization of building land parcellation using a genetic algorithm. Tehnika. 2026;81(5):441-449. doi:10.5937/tehnika2605441i
BibTeX
@article{Isailovi2026,
author = {Isailović, Aleksa and Stanić, Miloš and Dašić, Tina and Šoškić, Mladen and Višnjevac, Nenad},
title = {Optimization of building land parcellation using a genetic algorithm},
journal = {Tehnika},
year = {2026},
volume = {81},
number = {5},
pages = {441--449},
doi = {10.5937/tehnika2605441i}
}
EUVIMED