A 9-hour blended short course designed to pose the structural health monitoring (SHM) in the context of a statistical pattern recognition paradigm to support the damage identification process and risk-informed integrity management. The remote part introduces the concept of SHM and accelerates the learning process, while the in-person part focuses on bridging the gap between research and practical application. The techniques are shown with hands-on experiences applied to bridges using vibration-based monitoring (e.g., natural frequencies, mode shapes, and damping ratios) along with probabilistic numerical modeling. Unsupervised learning algorithms, such as Gaussian mixture models, and supervised learning algorithms like artificial neural networks or support vector machines are introduced. Practical considerations, limitations, grand challenges, and trends of SHM are all covered.
