Time Series Foundation Models are Flow Predictors
This provides accurate and scalable flow prediction for mobility planning, especially in data-limited scenarios, but is incremental as it applies existing foundation models to a specific domain.
The paper tackled crowd flow prediction using time series foundation models (Moirai and TimesFM) in a zero-shot setting without spatial information, achieving up to 33% lower RMSE, 39% lower MAE, and 49% higher CPC compared to state-of-the-art baselines on mobility datasets.
We investigate the effectiveness of time series foundation models (TSFMs) for crowd flow prediction, focusing on Moirai and TimesFM. Evaluated on three real-world mobility datasets-Bike NYC, Taxi Beijing, and Spanish national OD flows-these models are deployed in a strict zero-shot setting, using only the temporal evolution of each OD flow and no explicit spatial information. Moirai and TimesFM outperform both statistical and deep learning baselines, achieving up to 33% lower RMSE, 39% lower MAE and up to 49% higher CPC compared to state-of-the-art competitors. Our results highlight the practical value of TSFMs for accurate, scalable flow prediction, even in scenarios with limited annotated data or missing spatial context.