Graph-Based Inference and Topology-Aware Multi-Agent Reinforcement Learning for Large-Scale Railway Network Management

2026-09-24

AI Summary

This study addresses the intractability of centralized computation and the failure of distributed coordination in large-scale railway network maintenance by proposing a topology-aware multi-agent reinforcement learning framework. Methodologically, the approach integrates hierarchical Bayesian models, Gaussian process kernels on graphs, graph neural networks, and graph Transformers for effective graph reasoning. Furthermore, it achieves zero-shot transfer, enabling policies trained on small networks to be directly deployed on unseen larger networks without retraining. Experimental results demonstrate that the proposed method significantly outperforms heuristic algorithms and standard baselines, substantially reducing training time while maintaining superior performance. This work provides an efficient and scalable new paradigm for large-scale infrastructure maintenance.

Abstract

Modern infrastructure asset management constitutes a complex sequential decision-making problem, characterized by long planning horizons and system-level interactions, such as spatial deterioration correlations and economies of scale. While deep reinforcement learning has shown promise in optimizing maintenance policies, scaling to real-world networks remains challenging. Centralized approaches become computationally intractable in large-scale systems, whereas decentralized approaches often fail to capture essential coordination mechanisms. To address these challenges, we propose a graph-based framework that integrates accurate environment modeling with scalable decision support. First, we employ a hierarchical Bayesian model leveraging a Gaussian Process on Graph kernel to infer a realistic, spatially correlated networked environment of railway maintenance planning from real-world data provided by the Swiss Federal Railways. Second, we introduce a topology-aware Multi-Agent Reinforcement Learning (MARL) framework by integrating graph neural networks and graph Transformers to optimize network-level policies. A central contribution of this work is the demonstration of scalability through zero-shot transfer learning: graph-based agents, trained only on small network portions, are successfully deployed in a zero-shot manner on large-scale unseen networks without any retraining. Numerical results indicate that the proposed method significantly outperforms optimized heuristics and standard MARL baselines, reducing computational training time while maintaining superior performance on large-scale networks.

Problem

• Infrastructure Asset Management
• Sequential Decision-Making
• Multi-Agent Reinforcement Learning
• Scalability
• Railway Network Maintenance

Innovation

• Graph-Based Inference
• Topology-Aware MARL
• Zero-Shot Transfer Learning
• Graph Neural Networks
• Hierarchical Bayesian Model