The recent developments in the research area highlight a significant shift towards leveraging advanced computational models and artificial intelligence to address complex challenges in infrastructure resilience, energy efficiency, and emergency management. A common theme across the studies is the application of innovative machine learning techniques, such as graph neural networks (GNNs) and transfer learning, to improve the accuracy and efficiency of predictive models. These models are being applied to a wide range of problems, from optimizing evacuation routes during natural disasters to enhancing the resilience of power systems against extreme weather events. Additionally, there is a growing emphasis on integrating expert knowledge with physical principles to develop explainable AI models that can be trusted and effectively utilized in critical decision-making processes. The research also underscores the importance of considering socioeconomic factors and urban development features in assessing and mitigating the impacts of natural disasters on infrastructure systems. Overall, the field is moving towards more sophisticated, data-driven approaches that can provide actionable insights for improving the sustainability and resilience of critical infrastructure systems.

Noteworthy Papers

• Graph Neural Networks for Travel Distance Estimation and Route Recommendation Under Probabilistic Hazards: Introduces a GNN-based framework for efficient evacuation route planning during hurricanes, demonstrating significant improvements in computational efficiency and accuracy.
• Efficient Probabilistic Assessment of Power System Resilience Using the Polynomial Chaos Expansion Method with Enhanced Stability: Proposes an enhanced PCE method for assessing power system resilience, offering improved repeatability and convergence over traditional methods.
• Integrating Expert and Physics Knowledge for Modeling Heat Load in District Heating Systems: Presents HELIOS, an AI model that combines physical principles and expert knowledge for superior heat load modeling in DHS, emphasizing explainability and accountability.
• Anatomy of a Historic Blackout: Decoding Spatiotemporal Dynamics of Power Outages and Disparities During Hurricane Beryl: Provides a comprehensive analysis of power outage dynamics during a historic hurricane, highlighting the role of socioeconomic and urban development factors in recovery disparities.
• ARM-IRL: Adaptive Resilience Metric Quantification Using Inverse Reinforcement Learning: Develops a data-driven method for adaptive resilience metric learning in cyber-physical systems, using inverse reinforcement learning to improve system state estimation.
• European Energy Vision 2060: Charting Diverse Pathways for Europe's Energy Transition: Offers a detailed exploration of potential energy transition pathways for Europe, considering a wide range of social, technological, economic, political, and geopolitical factors.
• Multi-Objective Hyperparameter Selection via Hypothesis Testing on Reliability Graphs: Introduces a novel framework for multi-objective hyperparameter selection, leveraging reliability graphs to ensure robust statistical reliability guarantees.
• Risk and Vulnerability Assessment of Energy-Transportation Infrastructure Systems to Extreme Weather: Proposes a comprehensive assessment framework for evaluating the impact of extreme weather on energy-transportation infrastructure, incorporating a neural network surrogate for vulnerability identification.
• GenTL: A General Transfer Learning Model for Building Thermal Dynamics: Presents GenTL, a general transfer learning model for building thermal dynamics, demonstrating significant improvements in prediction accuracy and efficiency.