Facilitating Artificial Intelligence Resilience in Manufacturing Industrial Internet

Artificial intelligence (AI) systems have been increasingly adopted in the Manufacturing Industrial Internet (MII). Investigating and enabling AI resilience is very important to alleviate profound impact of AI system failures in manufacturing and Industrial Internet of Things (IIoT) operations, leading to critical decision making. However, there is a wide knowledge gap in defining the resilience of AI systems and analyzing potential root causes and corresponding mitigation strategies. In the presentation, I will discuss a new framework for investigating the resilience of AI performance over time under hazard factors in data quality, AI pipelines, and the cyber-physical layer. The proposed method can facilitate effective diagnosis and mitigation strategies to recover AI performance based on a multimodal multi-head self-latent attention model. The merits of the proposed method are elaborated using an MII testbed of connected Aerosol Jet® Printing (AJP) machines, fog nodes, and Cloud with inference tasks via AI pipelines. I will also discuss other factors to ensure AI model resilience in MII, such as data quality and computation services. 

Dr. Ran Jin is a Professor and the Director of Laboratory of Data Science and Visualization at the Grado Department of Industrial and Systems Engineering at Virginia Tech. He received his Ph.D. degree in Industrial Engineering from Georgia Tech, Atlanta, his master’s degrees in Industrial Engineering, and in Statistics, both from the University of Michigan, Ann Arbor, and his bachelor’s degree in Electronic Engineering from Tsinghua University, Beijing Dr. Jin’s research focuses on computation services in smart manufacturing and human-centered artificial intelligence (AI). His research includes lightweight middleware design in Industrial Internet, auto-machine learning and deep neural networks in manufacturing, and interface between data quality and data-driven modeling. He published over 60 journal papers in this area. His research has been broadly used by many manufacturing companies in additive manufacturing, thermal spray coating, broaching, semiconductor, printed electronics, optical and continuous fiber manufacturing industries. Dr. Jin served as President of Engineering Faculty Organization at Virginia Tech, Operations Officer of Faculty Senate, and Co-Chair of Commission on Research at Virginia Tech. He also served as Chair of Quality, Statistics, and Reliability subdivision in INFORMS and Associate Editor for IISE Transactions (Design and Manufacturing). Dr. Jin is currently serving as an Associate Editor for ASME Journal of Manufacturing Science and Engineering, Associate Editor for IEEE Transactions on Automation Science and Engineering, and Associate Editor for INFORMS Journal on Data Science.