Service life and durability prediction of reinforced concrete, physics-informed machine learning in engineering, self-healing concrete

Dimitrios Iason obtained his Diploma in Mechanical Engineering from the University of West Attica and his Master of Science in Artificial Intelligence from the University of Piraeus, in collaboration with NCSR Demokritos. His undergraduate research focused on remaining useful life prediction (service life) using machine learning techniques applied to real-time data and time–frequency analysis of wind turbine bearings. During his postgraduate studies, his research centered on structural health monitoring, specifically the use of machine learning methods to predict damage severity and defect modes in aeronautical structures, with emphasis on carbon fiber-reinforced composite plates.  

Currently, Dimitrios Iason is a PhD candidate, jointly supervised by the INN/CCM Research Group and IIT/INSANE Research Group, working at the intersection of materials science and machine learning. His research targets machine learning approaches for service life and durability prediction in cementitious materials containing self-healing admixtures. It is anticipated that physics-informed machine learning can develop a robust and generalizable predictive framework that embeds physical laws and governing equations to capture the dynamic interaction between cracking and healing processes across different composite systems, which if adopted in modern applications, may contribute towards enhanced material longevity and sustainability.   

Therefore, his research aims to understand degradation-repair interplay via physics-informed models, tracing predictive techniques' evolution, their performance impacts, and how self-healing admixtures enhance material durability under physical principles.  His primary research focus is the development of machine learning methodologies that integrate physical laws to predict the properties and performance of concrete structures incorporating self-healing agents. At CCM Research Group, he is actively involved in experimental investigations of key phenomena such as corrosion, degradation, and healing in cementitious materials and steel reinforcement. In addition, he contributes to multidisciplinary projects that combine advanced machine learning techniques with classical engineering applications.