EUVIMEDEUVIMED entdecken
Crossref

Sustainability assessment of electric vehicle charging infrastructure using deep learning, Analytic Network Process (ANP), and TOPSIS

Omer Kaya

International Journal of Applied Resilience and Sustainability · 2025

DOI: 10.70593/deepsci.0101005 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Applied Resilience and Sustainability, 2025. Verfasst von Omer Kaya.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The need to develop rapidly sustainable charging infrastructure for electric vehicles (EVs) presents a complex problem, as it requires consideration of numerous economic, environmental, technological and social elements. As a result, conventional decision-making frameworks typically are unable to adequately account for interdependent relationships between evaluation criteria and dynamic usage data, resulting in underdeveloped and less-than-optimal placement and construction of EV charging stations. To address this challenge, this paper proposed a novel integrated methodology using Deep Learning, Analytic Network Process (ANP) and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to evaluate the sustainability of EV Charging Infrastructure projects on a comprehensive level. Deep Learning models developed to forecast key usage metrics from real-world usage data to provide an objective method for comparing different options. Criteria weightings determined via ANP to reflect the many complex relationships between the various sustainability criteria and TOPSIS used to identify which of the alternatives for charging station placement or deployment plan is most similar to an ideal sustainable solution. We found that the integrated Deep Learning - ANP - TOPSIS methodology was able to correctly identify the most sustainable option, which represented a balance between cost savings, reduced greenhouse gas emissions, increased user convenience, and minimized negative impacts on the electrical grid. Compared to other alternatives, the highest ranked option had about 20 percent greater TOPSIS closeness coefficient and thus performed significantly better than the remaining options when compared across all four criteria. Results of sensitivity analysis indicated that the relative positioning of alternatives were relatively stable to changes in weighting of the criteria, which provided further evidence for the validity of the decision model. The methodology integrates predictive analytics with multi-criteria decision making to evaluate both a quantifiable measure of the performance for each option and a qualitatively assessed measure of the priority for that option, creating a sustainable and resilient electric vehicle charging system.

Zitieren

APA

Kaya, O. (2025). Sustainability assessment of electric vehicle charging infrastructure using deep learning, Analytic Network Process (ANP), and TOPSIS. International Journal of Applied Resilience and Sustainability. https://doi.org/10.70593/deepsci.0101005

Vancouver

Kaya O. Sustainability assessment of electric vehicle charging infrastructure using deep learning, Analytic Network Process (ANP), and TOPSIS. International Journal of Applied Resilience and Sustainability. 2025. doi:10.70593/deepsci.0101005

BibTeX
@article{Kaya2025,
  author = {Kaya, Omer},
  title = {Sustainability assessment of electric vehicle charging infrastructure using deep learning, Analytic Network Process (ANP), and TOPSIS},
  journal = {International Journal of Applied Resilience and Sustainability},
  year = {2025},
  doi = {10.70593/deepsci.0101005}
}
Kostenlos starten →