Examining Spatial Concordance of Knowledge Aggregated AHP TOPSIS and Random Forest for Electric Vehicle Charging Site Suitability
Applied Research Advances · 2026
DOI: 10.65069/ara21202636 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Research Advances, 2026. Verfasst von Shahrier Main Mridul. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Electric vehicle (EV) charging infrastructure planning in data-scarce megacities is often constrained by limited spatial data and expert-driven decision models whose underlying assumptions are rarely validated. This study presents an integrated GIS-based decision-support framework for identifying optimal EV charging locations in Dhaka, Bangladesh — a 22-million-resident megacity targeting 30% EV penetration by 2030, by combining expert-elicited Analytic Hierarchy Process (AHP) weighting with the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). Unlike conventional GIS-MCDM studies, the proposed framework systematically evaluates criterion collinearity and assesses the spatial concordance between expert-derived suitability patterns and a machine-learning model using Random Forest with spatial cross-validation. Fourteen spatial criteria derived from open-access datasets and weighted by eleven domain experts were used to generate a 100-m suitability map. The analysis identified fifteen priority charging zones, with the Tejgaon–Hatirjheel corridor ranking highest. Results showed strong agreement between expert- and machine-learning-derived suitability surfaces (Spearman's ρ = 0.968; top-zone overlap = 73.8%), while revealing substantial collinearity between transit and fuel station proximity, highlighting a potential source of bias in conventional MCDM applications. Built entirely on open data, the proposed framework offers a transparent, reproducible, and transferable methodology for EV charging infrastructure planning in data-constrained cities.
Zitieren
APA
Mridul, S. M. (2026). Examining Spatial Concordance of Knowledge Aggregated AHP TOPSIS and Random Forest for Electric Vehicle Charging Site Suitability. Applied Research Advances. https://doi.org/10.65069/ara21202636
Vancouver
Mridul SM. Examining Spatial Concordance of Knowledge Aggregated AHP TOPSIS and Random Forest for Electric Vehicle Charging Site Suitability. Applied Research Advances. 2026. doi:10.65069/ara21202636
BibTeX
@article{Mridul2026,
author = {Mridul, Shahrier Main},
title = {Examining Spatial Concordance of Knowledge Aggregated AHP TOPSIS and Random Forest for Electric Vehicle Charging Site Suitability},
journal = {Applied Research Advances},
year = {2026},
doi = {10.65069/ara21202636}
}
EUVIMED