Determination of location selection criteria of electric vehicle charging stations with GIS and machine learning

Muhammed Oğuzhan Mete

Makale detayları

Determination of location selection criteria of electric vehicle charging stations with GIS and machine learning

Yazarlar:

Muhammed Oğuzhan Mete

Yayıncı:

Nigde Omer Halisdemir University

Ana Başlıklar:

Makine Öğrenme (Diğer) | Coğrafi Bilgi Sistemleri ve Mekansal Veri Modelleme | Planlamada Coğrafi Bilgi Sistemleri (CBS)

Dergi:

Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.14 Issue 4, pp.1664-1676

Belge Tipi:

Makale

Yayın Tarihi:

10/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001360293

pISSN:

2564-6605

Anahtar kelimeler:

Özet:

 As the use of electric vehicles increases worldwide, the need for new charging station installations arises. The main objective of this study is to develop a suitable location selection model for electric vehicle charging stations, which are replacing vehicles with internal combustion engines nowadays. For this purpose, Munich, Germany, and Milan, Italy, have been selected as the study areas. The selection of these cities is based on the high number of electric vehicle charging stations and the similarity of these cities to Türkiye in terms of traffic and driving culture. Within the scope of the study, spatial analysis is performed with Geographic Information Systems GIS, regression analysis is performed with Explainable Artificial Intelligence XAI techniques, and a model that can be used to determine the most suitable locations for charging stations is developed. The weights of the criteria are determined using the Random Forest method, and the model accuracy is observed to be 27 in Milan and 87 in Munich. As a result, the criteria that are effective for selecting the most suitable locations for electric vehicle charging stations have been determined with inferences based on the utilization data of existing stations. Thus, it aims to facilitate the transition from fossil fuel vehicles to electric vehicles by ensuring the efficient use of resources by directing investments in this field with a decision support system and providing suitable transportation infrastructure for electric vehicle owners.

 As the use of electric vehicles increases worldwide, the need for new charging station installations arises. The main objective of this study is to develop a suitable location selection model for electric vehicle charging stations, which are replacing vehicles with internal combustion engines nowadays. For this purpose, Munich, Germany, and Milan, Italy, have been selected as the study areas. The selection of these cities is based on the high number of electric vehicle charging stations and the similarity of these cities to Türkiye in terms of traffic and driving culture. Within the scope of the study, spatial analysis is performed with Geographic Information Systems GIS, regression analysis is performed with Explainable Artificial Intelligence XAI techniques, and a model that can be used to determine the most suitable locations for charging stations is developed. The weights of the criteria are determined using the Random Forest method, and the model accuracy is observed to be 27 in Milan and 87 in Munich. As a result, the criteria that are effective for selecting the most suitable locations for electric vehicle charging stations have been determined with inferences based on the utilization data of existing stations. Thus, it aims to facilitate the transition from fossil fuel vehicles to electric vehicles by ensuring the efficient use of resources by directing investments in this field with a decision support system and providing suitable transportation infrastructure for electric vehicle owners.

 As the use of electric vehicles increases worldwide, the need for new charging station installations arises. The main objective of this study is to develop a suitable location selection model for electric vehicle charging stations, which are replacing vehicles with internal combustion engines nowadays. For this purpose, Munich, Germany, and Milan, Italy, have been selected as the study areas. The selection of these cities is based on the high number of electric vehicle charging stations and the similarity of these cities to Türkiye in terms of traffic and driving culture. Within the scope of the study, spatial analysis is performed with Geographic Information Systems GIS, regression analysis is performed with Explainable Artificial Intelligence XAI techniques, and a model that can be used to determine the most suitable locations for charging stations is developed. The weights of the criteria are determined using the Random Forest method, and the model accuracy is observed to be 27 in Milan and 87 in Munich. As a result, the criteria that are effective for selecting the most suitable locations for electric vehicle charging stations have been determined with inferences based on the utilization data of existing stations. Thus, it aims to facilitate the transition from fossil fuel vehicles to electric vehicles by ensuring the efficient use of resources by directing investments in this field with a decision support system and providing suitable transportation infrastructure for electric vehicle owners.

 Elektrikli araç kullanımının dünya genelinde artması, yeni şarj istasyonu kurulumlarının ihtiyacını da ortaya çıkarmaktadır. Bu çalışmanın temel amacı günümüzde içten yanmalı motora sahip araçların yerini alan elektrikli araçların şarj istasyonları için uygun yer seçimi modelinin geliştirilmesidir. Bu amaç doğrultusunda Almanyanın Münih ile İtalyanın Milano şehirleri çalışma bölgesi olarak belirlenmiştir. Bu şehirlerin seçilmesinde elektrikli araç şarj istasyonlarının sayısının fazla olması ve ülkelerin trafik ve sürüş kültürü açısından Türkiyeye benzerliği öne çıkmıştır. Çalışma kapsamında Coğrafi Bilgi Sistemleri CBS ile konumsal analizler yapılmış, Açıklanabilir Yapay Zeka XAI teknikleri ile regresyon analizi gerçekleştirilerek şarj istasyonları için en uygun yerlerin belirlenmesinde kullanılabilecek bir tahmin modeli geliştirilmiştir. Rastgele Orman yöntemiyle kriterlerin ağırlıkları belirlenmiş ve model doğruluğu Milano şehrinde 27 iken Munih şehrinde 87 olarak gözlemlenmiştir. Sonuç olarak elektrikli araç şarj istasyonları için mevcuttaki istasyonlara ait kullanım verisine dayalı çıkarımlar ile en uygun konumların seçiminde etkin kriterler belirlenmiştir. Böylelikle bu alandaki yatırımların karar destek sistemi ile yönlendirilmesi sağlanarak kaynakların etkin kullanılması, elektrikli araç sahiplerine uygun ulaşım altyapısının sunulması ile fosil yakıtlı araçlardan elektrikli araçlara geçişin kolaylaştırılması amaçlanmıştır.

Makale detayları

Determination of location selection criteria of electric vehicle charging stations with GIS and machine learning

Yazarlar:

Muhammed Oğuzhan Mete

Yayıncı:

Nigde Omer Halisdemir University

Ana Başlıklar:

Makine Öğrenme (Diğer) | Coğrafi Bilgi Sistemleri ve Mekansal Veri Modelleme | Planlamada Coğrafi Bilgi Sistemleri (CBS)

Dergi:

Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.14 Issue 4, pp.1664-1676

Belge Tipi:

Makale

Yayın Tarihi:

10/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001360293

pISSN:

2564-6605

Explainable AI

Geographic Information Systems

Electric Vehicle Charging Station

Machine Learning

Site Selection AnalysisAçıklanabilir Yapay Zeka

Coğrafi Bilgi Sistemleri

Elektrikli Araç Şarj İstasyonu

Makine Öğrenmesi

Yer Seçim Analizi

Sayı

Makale detayları

Yazarlar: Muhammed Oğuzhan Mete

Yayıncı: Nigde Omer Halisdemir University

Ana Başlıklar: Makine Öğrenme (Diğer) | Coğrafi Bilgi Sistemleri ve Mekansal Veri Modelleme | Planlamada Coğrafi Bilgi Sistemleri (CBS)

Dergi: Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.14 Issue 4, pp.1664-1676

Belge Tipi: Makale

Yayın Tarihi: 10/2025

Belge Dili: Türkçe

Document ID (Org No.): 2001360293

pISSN: 2564-6605

Sayı

Yazarlar:

Muhammed Oğuzhan Mete

Yayıncı:

Nigde Omer Halisdemir University

Ana Başlıklar:

Makine Öğrenme (Diğer) | Coğrafi Bilgi Sistemleri ve Mekansal Veri Modelleme | Planlamada Coğrafi Bilgi Sistemleri (CBS)

Dergi:

Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.14 Issue 4, pp.1664-1676

Belge Tipi:

Makale

Yayın Tarihi:

10/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001360293

pISSN:

2564-6605