Operations Research 2025
Abstract Submission

2208. Charger Location Assessment and Utilization Prediction Using Cluster Analysis and Machine Learning – A case study in Germany

Invited abstract in session TD-9: Optimization in Energy Infrastructure Planning, stream Energy and Sustainability.

Thursday, 14:30-16:00
Room: H15

Authors (first author is the speaker)

1. Nils Katzke
Information Systems, Freie Universität Berlin
2. David Rößler-von Saß
Wirtschaftsinformatik, Freie Universität Berlin
3. Natalia Kliewer
Information Systems, Freie Universitaet Berlin
4. Christian Hein
Nationale Leitstelle Ladeinfrastruktur, NOW GmbH

Abstract

The shift to sustainable transportation makes the electrification of individual motorized travel crucial. However, adoption remains slow due to high vehicle costs and, more significantly, range limitations. To mitigate range anxiety, public charging infrastructure must be strategically planned to ensure high availability and demand orientation, allowing electric vehicles to meet travelers’ needs as reliably as conventional vehicles.
To support this goal, this study presents a data-driven approach to identifying critical charging locations based on multiple performance indicators. Feature selection techniques and clustering algorithms are applied to classify charging sites based on technological and contextual attributes.
The analysis is based on charging data from Germany’s public-funded infrastructure, provided by NOW GmbH. Our experiments explore various feature selection strategies and clustering methods to determine their effectiveness in identifying utilization patterns. Associations between cluster characteristics and observed utilization are evaluated to derive insights into charging behaviors.
The results demonstrate the potential of clustering to inform charger utilization prediction and provide valuable implications for charging location management and policy development. This approach supports efficient infrastructure deployment and offers a framework for guiding future planning and incentivization strategies.

Keywords

Status: accepted

Back to the list of papers