# RSTGCN: Railway-centric Spatio-Temporal Graph Convolutional Network for Train Delay Prediction

arXiv: https://arxiv.org/abs/2510.01262 · HTML: https://mlsanity.com/paper/2510.01262/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 64 references that are in the map.
- First posted on arXiv 26 Sep 2025.

## Builds on (most cited first, 7 of 7)

- [Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition](https://mlsanity.com/paper/1801.07455/) (2018)
- [Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting](https://mlsanity.com/paper/1709.04875/) (2017)
- [T-GCN: A Temporal Graph ConvolutionalNetwork for Traffic Prediction](https://mlsanity.com/paper/1811.05320/) (2018)
- [Structured Sequence Modeling with Graph Convolutional Recurrent Networks](https://mlsanity.com/paper/1612.07659/) (2016)
- [Estimating Train Delays in a Large Rail Network Using a Zero Shot Markov Model](https://mlsanity.com/paper/1806.02825/) (2018)
- [A Novel Markov Model for Near-Term Railway Delay Prediction](https://mlsanity.com/paper/2205.10682/) (2022)
- [Railway Network Delay Evolution: A Heterogeneous Graph Neural Network Approach](https://mlsanity.com/paper/2303.15489/) (2023)

## Built on by (most cited first, 1 of 1)

- [RIDE: An Open Dataset and Benchmark for Train Delay Prediction](https://mlsanity.com/paper/2606.05070/) (2026)

## Abstract (arXiv)

Accurate prediction of train delays is critical for efficient railway operations. While earlier approaches have largely focused on forecasting the exact delays of individual trains, studies on station-level delay prediction are somewhat sparse. To address this gap, we propose the Railway-centric Spatio-Temporal Graph Convolutional Network (RSTGCN), designed to forecast average arrival delays of all the incoming trains at a particular station for a particular time period. Our approach incorporates several architectural innovations and novel feature integrations, including train frequency-aware spatial attention, which significantly enhance predictive performance. To support this effort, we curate and release a comprehensive dataset for the entire Indian Railway Network (IRN), spanning 4,735 stations across 17 zones - the largest and most diverse railway network studied to date. We conduct extensive experiments using multiple state-of-the-art baselines, demonstrating consistent improvements across standard metrics. Specifically, RSTGCN outperforms the best baseline by 18% in MAE, 14% in MAPE, and 1-8% in RMSE on the IRN.