RIDE: An Open Dataset and Benchmark for Train Delay Prediction
Researchers have introduced RIDE, a comprehensive dataset and benchmark designed to standardize train delay prediction. This nationwide dataset, covering the Belgian railway network from 2023 to 2025, includes 94.5 million train events and 35.7 million weather records. The benchmark facilitates direct comparison of various prediction models, revealing that graph neural networks currently achieve the best performance, outperforming traditional statistical and non-learning methods. AI
IMPACT Standardizes evaluation for train delay prediction models, potentially accelerating adoption of advanced machine learning techniques.