The 8th International Conference on System Reliability and Safety
Sicily, Italy - November 20-22, 2024

Mohammad Pourgol, University of Maryland, USA

 

Dr. Mohammad Pourgol is a safety/reliability analyst in multidisciplinary systems analysis and an Associate Professor (adj) of Mechanical Engineering at the University of Maryland. Previously, he served as an Associate Professor of Reliability Engineering at Sahand University of Technology (SUT). He earned his Ph.D. in Reliability Engineering from the University of Maryland (UMD). Dr. Pourgol possesses over 20 years of work experience, encompassing industrial application, research, and teaching in safety applications and reliability engineering. He has worked at various institutions, including Teradyne Semiconductor, Johnson Controls, Sahand University of Technology, UMD, Massachusetts Institute of Technology (MIT). He is an elected ASQ Fellow (Currently member of Executive Committee for its Riska and Reliability Division, ASME Fellow (ASME Safety Engineering and Risk/Reliability Analysis Division (SERAD) Past-Chair), IEEE Senior Member, ANS member, and serves on several technical committees. Dr. Pourgol is also a registered Professional Engineer (PE) in the State of Massachusetts. Additionally, he holds certifications as a Certified Reliability Engineer (ASQ CRE), Certified Six Sigma Black Belt (CSSBB), and Manager of Quality/Organization Excellence (ASQ CMQ/OE). He has authored over 160 papers in archival journals and peer-reviewed conferences on his research and has been invited/keynote speaker at several conferences/webinars. Furthermore, he has filed one US patent. His efforts have been recognized with several awards.

Speech Title: Augmented Physics of Failure (Pof)/Deep Learning Platform for Enhanced Prognostics and Health Management (PHM)
This speech aims to overview a proposed augmented data-driven/Physics of Failure-based approach for monitoring the structural integrity of critical mission systems under dynamic loading. In such systems, various undesirable event scenarios can occur, ranging from abrupt to gradual changes, necessitating a comprehensive assessment of failure probabilities for structures, systems, and components (SSCs). By integrating Physics of Failure (PoF) models or leveraging Deep Learning networks, we seek to improve the accuracy and reliability of failure probability assessments, where dynamic loading conditions introduce complexities and uncertainties. Current Prognostics and Health Management (PHM) techniques typically rely on either PoF models or Data Analytics approaches. While PoF models simulate component behavior and predict remaining useful life (RUL), they may encounter challenges related to complex relationships and data availability. On the other hand, Data Analytics techniques analyze large datasets but often lack physical insights. The proposed augmented approach offers significant advantages by providing a comprehensive understanding through the incorporation of both techniques. This approach aims to bridge the gap between physical insights and data-driven analyses, thereby enhancing the reliability estimation of structural integrity monitoring in critical mission systems subjected to dynamic loading conditions. We will discuss the challenges faced, current limitations in PHM techniques, and how integrating the proposed framework can enhance accuracy and reliability in failure probability assessments. Real-world applications and case studies will be presented to demonstrate the practical implementation, effectiveness and challenges of this approach in ensuring the resilience and durability of vital infrastructures.