Integrated monitoring models for dam health and safety: A review of methods and considerations

Dam safety and health monitoring are critical to ensuring the long-term reliability of water infrastructure, given the complex behavior of dams and the potentially severe consequences of failure. To evaluate current monitoring approaches, this study employs a structured literature review methodology that includes systematic database searching, screening, and qualitative synthesis of 215 relevant publications. The reviewed studies were organized into four major categories comprising artificial intelligence (AI)-based models, numerical models, hybrid models, and modern monitoring technologies. The review critically assesses the strengths, limitations, and application domains of these approaches, highlighting the increasing integration of AI techniques with numerical simulations to improve predictive accuracy while maintaining physical interpretability. Particular attention is given to methodological factors influencing model performance, including input variable selection, data quality, model validation strategies, and uncertainty management. The study also examines the role of explainable artificial intelligence (XAI) in improving transparency, interpretability, and engineering trust in safety-critical decision-making processes. Furthermore, emerging technologies such as digital twins, remote sensing systems, and unmanned aerial vehicle (UAV)-assisted inspections are evaluated as key enablers of real-time monitoring and early warning systems. The review identifies major research challenges, including data scarcity, model generalization, computational demands, and interoperability across monitoring platforms. Future research directions are discussed, emphasizing the potential of Physics-Informed Neural Networks (PINNs), foundation models, federated learning, and cyber-secure monitoring frameworks to support the next generation of intelligent dam monitoring systems. The findings provide practical insights for researchers and practitioners and help guide the development of more reliable and resilient dam monitoring frameworks.