The project aims to develop a machine learning (ML) based approach for efficient corrosion engineering and predictive maintenance, targeting continuous monitoring as well as accelerated testing protocols for material development in the aerospace industry. To this end, ultrasonic sensing, corrosion analytics, and simulation will be synergistically combined to classify corrosive processes. ML based algorithms will thus be trained to predict corrosion, as well as the type of corrosion, with high reliability. In addition, corrosion testing and monitoring can greatly benefit from early detection to accelerate material development and reduce material consumption through timely detection and minimal repair. Predictive monitoring of corrosion, as well as accelerated development of corrosion-resistant materials based on ML, offer a promising way to advance the aerospace industry toward sustainable material use. Based on this project, the continuous stream of data will be used to classify corrosion that can be intuitively understood through a human-machine interface, including qualified corrosion predictions through ML models generated from test campaigns. This will enable a significant reduction in material development time and open new market opportunities for ultrasound-based sensing and machine learning applications in aerospace.