Applications of Graph Neural Networks in Civil Infrastructures: A Review on Transportation, Power, Water, and Structural Systems

This paper provides a systematic and structured review of Graph Neural Network (GNN) applications across foundational civil infrastructure domains, including transportation networks, power systems, water distribution systems, and structural systems.

Unlike prior domain-specific surveys, this work unifies cross-sector GNN developments and categorizes recent research (2021–2024) by:

• Infrastructure domain
• Task type (estimation, forecasting, optimization, anomaly detection, planning)
• GNN architecture (GCN, GAT, GraphSAGE, spatio-temporal GNNs, graph transformers, physics-informed GNNs)
• Level of deployment maturity

The review synthesizes over 500 recent studies and evaluates how GNNs are used as:

• Predictive models for forecasting and classification
• Surrogate models for accelerating simulation and optimization
• Decision-support tools for operational control and long-term planning

A major contribution of the paper is its structured gap analysis, identifying limitations that persist across domains:

• Scalability to large-scale infrastructure graphs
• Real-time adaptability and streaming graph learning
• Uncertainty quantification and probabilistic outputs
• Interpretability and regulatory alignment
• Robustness under distribution shifts and extreme events

The paper further proposes research directions including:

• Hierarchical and distributed GNN architectures
• Physics-informed and constraint-aware learning
• Hybrid GNN–optimization and GNN–reinforcement learning frameworks
• Equity-aware and explainable infrastructure AI systems

By bridging machine learning theory and infrastructure practice, this review provides a roadmap for deploying GNNs in safety-critical, large-scale civil systems.

Recommended citation: Anand, H., Khayambashi, K., Zandsalimi, Z., Taghizadeh, M., Hasnat, M. A., & Alemazkoor, N. (2025). Applications of Graph Neural Networks in Civil Infrastructures: A Review on Transportation, Power, Water, and Structural Systems. Engineering Applications of Artificial Intelligence (Under Review).
Download Paper