Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.21369 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909820094775296 |
|---|---|
| author | Yenigun, Onur Baydogmus, Gozde Karatas Yildiz, Kazim |
| author_facet | Yenigun, Onur Baydogmus, Gozde Karatas Yildiz, Kazim |
| contents | The rapidly increasing use of electric vehicles (EVs) has made it even more important to manage the charging infrastructure sustainably. The expansion of charging station networks, especially in large cities, creates serious logistical challenges for charging point operators (CPOs) in planning maintenance and repair activities. Inefficient field personnel management can lead to time loss, high operational costs, and resource waste. This study presents an integrated method to optimize the planning of EV charging network maintenance operations. The proposed approach groups charging stations according to geographical proximity using the K-means clustering algorithm and calculates the shortest routes between clusters using a genetic algorithm. The method was developed in Python and applied to a dataset consisting of 100 EV charging stations in Istanbul.
Considering the population density, traffic density, and resource constraints of Istanbul, the route planning approach presented in this study has great potential, especially for such metropolises. According to the different parameter configurations tested, the most efficient scenario provided approximately 35\% distance savings compared to the reference route created according to the sequential data layout. While the reference route provides a simple comparison, the study presents a solution that will enable field operations in metropolitan cities such as Istanbul to be conducted in a more efficient, planned and scalable manner. In future studies, it is planned to integrate real-time factors such as traffic conditions and field technician constraints. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_21369 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Smart Routing for EV Charge Point Operators in Mega Cities: Case Study of Istanbul Yenigun, Onur Baydogmus, Gozde Karatas Yildiz, Kazim Neural and Evolutionary Computing The rapidly increasing use of electric vehicles (EVs) has made it even more important to manage the charging infrastructure sustainably. The expansion of charging station networks, especially in large cities, creates serious logistical challenges for charging point operators (CPOs) in planning maintenance and repair activities. Inefficient field personnel management can lead to time loss, high operational costs, and resource waste. This study presents an integrated method to optimize the planning of EV charging network maintenance operations. The proposed approach groups charging stations according to geographical proximity using the K-means clustering algorithm and calculates the shortest routes between clusters using a genetic algorithm. The method was developed in Python and applied to a dataset consisting of 100 EV charging stations in Istanbul. Considering the population density, traffic density, and resource constraints of Istanbul, the route planning approach presented in this study has great potential, especially for such metropolises. According to the different parameter configurations tested, the most efficient scenario provided approximately 35\% distance savings compared to the reference route created according to the sequential data layout. While the reference route provides a simple comparison, the study presents a solution that will enable field operations in metropolitan cities such as Istanbul to be conducted in a more efficient, planned and scalable manner. In future studies, it is planned to integrate real-time factors such as traffic conditions and field technician constraints. |
| title | Smart Routing for EV Charge Point Operators in Mega Cities: Case Study of Istanbul |
| topic | Neural and Evolutionary Computing |
| url | https://arxiv.org/abs/2509.21369 |