New Publication in "HMD Praxis der Wirtschaftsinformatik"

20 June 2025, by Marten Borchers

Photo: UHH

We, Eva Bittner, Martin Semmann, Enrico Milutzki, Hauke Volquardsen, Jan Willruth, and Marten Borchers, are pleased to announce the publication of our paper "Development of ML-based Forecasting Systems for the Energy Consumption of Electric Buses in Public Transport", created in collaboration with Hamburger Hochbahn AG, the University of Hamburg, and the Hub of Computing & Data Science.

The electrification of public transport is a key measure for achieving international and European climate goals, such as those outlined in the European Green Deal. However, the transition to battery-electric buses introduces new operational challenges — especially in terms of reliable planning and the highly variable energy consumption.

To bridge the gap between real-world operational requirements and technical implementation, we analyzed operational driving and consumption data over a period of two years, conducted fifteen expert interviews with professionals from the e-bus sector, and developed several machine learning models. Using these models, energy consumption can be predicted with an average deviation of less than 0.4%.

In addition, we provide eight specific design recommendations to support the conception, development, and integration of such forecasting systems into operational practice.

We would like to thank all project partners, and especially Johanna Ahrens, Felix Schmidt, and Jan Krause for their valuable support, as well as all the experts for sharing their insights and experiences.

The full publication is available online at: https://link.springer.com/article/10.1365/s40702-025-01185-5