Abstract:The monitoring process for complex infrastructure requires collecting various data sources with varying time scales, resolutions, and levels of abstraction. These data sources include data from human inspections, historical failure records, cost data, high-fidelity physics-based simulations, and online health monitoring. Such heterogeneity presents significant challenges in implementing a diagnostic and prognostic framework for decision-making regarding maintenance (and other life cycle actions). The core challenge lies in the effective integration of physical information and data-driven models, aiming to synergize their strengths to overcome individual limitations. One possible solution is to propose an approach that considers the strengths and limitations of each data source, as well as their compatibility with each other. The flexibility and efficacy of contemporary learning approaches can be used with more systematic and informative physics-based models that draw on domain expertise. This represents an inherent desire to base all inferences on both our engineering knowledge and monitoring data that is at our disposal. In this context, the article reviews recent advances in this field, particularly in physics-based and deep learning techniques. It looks at new theories and models developed in the last five years, especially those used in system health monitoring, predicting damage, and planning maintenance. These new methods are proving to be more accurate and efficient than older, more traditional techniques. However, there are still challenges to be addressed. These include the need for high-quality data, finding the right balance between accuracy and the time it takes to compute, and effectively combining physical models with data-driven models. The paper calls for further research into methods that can handle large amounts of complex data and consider uncertainties in both the models and the data. Finally, it highlights the need to explore how these models can be adapted for different systems and used in real-time applications.

Physics-of-Failure Methods and Prognostic and Health Management of Electronic Components

Computational Analysis of Mechanical and Electromigration Reliability Problems

Sensor-Driven Reliability and Wearout Management

Physics-of-failure-based prognostics for electronic products

Physics of Failure (PoF) Based Lifetime Prediction of Power Electronics at the Printed Circuit Board Level

A Simulation-based strategy used in electrical design for reliability

Probabilistic Physics of Failure Concept and Its Application in Reliability Engineering

A Hybrid Model for Prognostic and Health Management of Electronic Devices

A combined physics of failure and Bayesian network reliability analysis method for complex electronic systems

A review of physics-based learning for system health management

Physics-informed machine learning for reliability and systems safety applications: State of the art and challenges

Adaptive Machine Learning with Physics-based Simulations for Mean Time to Failure Prediction of Engineering Systems

A PoF Based Reliability Assessment Framework for Complex Systems

Degradation Estimation and Prediction of Electronic Packages Using Data-Driven Approach

Prognostics for Semiconductor Sustainability: Tool Failure Behavior Prediction in Fabrication Processes

Integrated Approach for Field Reliability Prediction Based on Accelerated Life Testing

Modeling and Predicting Transistor Aging under Workload Dependency using Machine Learning

Physics-informed deep learning: A promising technique for system reliability assessment

Improving Network Processing Dependability with Heterogeneous Reliability Cores

Stochastic Modeling of Damage Physics for Mechanical Component Prognostics Using Condition Indicators

Failure Analysis and Quantification for Contemporary and Future Supercomputers