IAI DevOps: A Systematic Framework for Prognostic Model Lifecycle Management

This paper proposes IAI DevOps, a systematic framework to address challenges of developing and operationalizing AI models in manufacturing industries, with an emphasis on prognostics and health management (PHM) applications. The paper starts by introducing the growing need of accelerating AI model d...

Full description

Saved in:
Bibliographic Details
Published in2019 Prognostics and System Health Management Conference (PHM-Qingdao) pp. 1 - 6
Main Authors Guo, Ziqi, Bao, Tingwen, Wu, Wenlong, Jin, Chao, Lee, Jay
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2019
Subjects
DevOps
Frequency modulation
Industrial AI
IP networks
Iron
model lifecycle management
PHM
Online AccessGet full text
DOI10.1109/PHM-Qingdao46334.2019.8943069

Cover

More Information
Summary:This paper proposes IAI DevOps, a systematic framework to address challenges of developing and operationalizing AI models in manufacturing industries, with an emphasis on prognostics and health management (PHM) applications. The paper starts by introducing the growing need of accelerating AI model development and enhancing its lifecycle reliability in industrial AI systems. The framework for industrial AI (IAI) DevOps is then proposed, with key components including industrial data management, streamlined model training, risk monitoring and model update, and decision support and feedback. After detailed introduction of the function and principles of each component, an implementation of the supporting platform is illustrated for the reference of the readers. With the help of the IAI DevOps framework, the authors believe that interdisciplinary engineering teams can collaborate better to accelerate the implementation of PHM systems as well as broader industrial AI applications in manufacturing industries.
DOI:10.1109/PHM-Qingdao46334.2019.8943069