Short-term crowd forecasting for SAIL2025
Acta Polytechnica CTU Proceedings · 16.06.2026 · 57 · 52-60
DOI: 10.14311/app.2026.57.0052 ↗Kurz gefasst
Fachartikel in der Zeitschrift Acta Polytechnica CTU Proceedings, online erschienen am 16.06.2026. Verfasst von Winnie Daamen und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Every five years, the SAIL event in the center of Amsterdam attracts millions of visitors. Managing large crowds in relatively small areas requires real-time insight into the traffic situation – particularly pedestrian flows – and forecasts thereof. To provide reliable short-term crowd forecasts without historical data, we developed an adaptive hybrid AI model that learns exclusively from live sensor, weather, and event data during the SAIL event. Our hybrid architecture integrates a periodically retrained LightGBM model with a Kalman Filter for real-time error correction and Conformalized Quantile Regression yields calibrated 90 % prediction intervals. The resulting system produces robust 4-hour forecasts and offers a generalizable solution to manage crowds at other large-scale events, particularly those infrequent events with limited historical data.
Zitieren
APA
Daamen, W., Hari, T., Xin, Y., Hoogendoorn-Lanser, S., & Hoogendoorn, S. (2026). Short-term crowd forecasting for SAIL2025. Acta Polytechnica CTU Proceedings, 57, 52-60. https://doi.org/10.14311/app.2026.57.0052
Vancouver
Daamen W, Hari T, Xin Y, Hoogendoorn-Lanser S, Hoogendoorn S. Short-term crowd forecasting for SAIL2025. Acta Polytechnica CTU Proceedings. 2026;57:52-60. doi:10.14311/app.2026.57.0052
BibTeX
@article{Daamen2026,
author = {Daamen, Winnie and Hari, Theivaprakasham and Xin, Yanan and Hoogendoorn-Lanser, Sascha and Hoogendoorn, Serge},
title = {Short-term crowd forecasting for SAIL2025},
journal = {Acta Polytechnica CTU Proceedings},
year = {2026},
volume = {57},
pages = {52--60},
doi = {10.14311/app.2026.57.0052}
}
EUVIMED