
Lee, Jay
Director of Industrial AI Center
Mechanical Engineering
Maryland Robotics Center
Dr. Jay Lee is Clark Distinguished Professor and Director of Industrial AI Center in the Mechanical Engineering Dept. of the Univ. of Maryland College Park.
Previously, he served as an Ohio Eminent Scholar, L.W. Scott Alter Chair and Univ. Distinguished Professor at Univ. of Cincinnati. He was Founding Director of National Science Foundation (NSF) Industry/University Cooperative Research Center (I/UCRC) on Intelligent Maintenance Systems (www.imscenter.net) during 2001-2019. IMS Center has developed research memberships with over 100 global company since 2000 and was selected as the most economically impactful I/UCRC in the NSF Economic Impact Study Report in 2012. He is also the Founding Director of Industrial AI Center (www.iaicenter.com ). He mentored his students and developed a number of start-up companies including Predictronics through NSF iCorp in 2013 and has won 1st Place for PHM Society Data Challenges competition 5 times.
He was on leave from UC to serve as Vice Chairman and Board Member for Foxconn Technology Group (ranked 26th in Global Fortune 500) during 2019-2021 to lead the development of Foxconn Wisconsin Science Park (~$1B investment) in Mt. Pleasant, WI (www.foxconnwiofficial.com). In addition, he advised Foxconn business units to successfully receive five WEF Lighthouse Factory Awards since 2019.
He is a member of Global Future Council on Advanced Manufacturing and Production of the World Economics Council (WEF), a member of Board of Governors of the Manufacturing Executive Leadership Council of National Association of Manufacturers (NAM), Board of Trustees of MTConnect, as well as a senior advisor to McKinsey. Previously, he served as Director for Product Development and Manufacturing at United Technologies Research Center (now Raytheon Technologies Research Center) as well as Program Director for a number of programs at NSF.
He was selected as 30 Visionaries in Smart Manufacturing in by SME in Jan. 2016 and 20 most influential professors in Smart Manufacturing in June 2020, SME Eli Whitney Productivity Award, and SME/NAMRC S.M. Wu Research Implementation Award in 2022. His new book on Industrial AI was published by Springer in 2020.
For publication citations, see
Google Scholar https://scholar.google.com/citations?user=g9GtqgQAAAAJ&hl=en&oi=ao
ResearchID: https://researchid.co/jay.lee
ResearchGate https://www.researchgate.net/profile/Jay_Lee10
2023 Ranking of World Mechanical and Aerospace Professors (based on https://research.com/): 71th in the world and 42nd in the US,
https://research.com/scientists-rankings/mechanical-and-aerospace-engineering
Top 2% of World Scientists (ranked 86th among 83,299 of top 2% world scientists in the field of Industrial Engineering and Automation) by Stanford Univ. 2022. https://elsevier.digitalcommonsdata.com/datasets/btchxktzyw/4
WEF Davos 2023 Featured video on AI for Smart Manufacturing. https://app.frame.io/presentations/c6b734dd-9db2-4900-bab2-903ace86979f
WEF interview article this week https://theinnovator.news/interview-of-the-week-jay-lee-industrial-ai-expert/
Industrial AI, Industrial Big Data, Prognostics and Health Management (PHM), Smart Manufacturing, Digital Twin, etc.
1. Intelligent Metrology Systems of Advanced Semiconductor Manufacturing.
2. Smart Electrification Analytics for EV systems.
3. AI Augmented ICU and Medical Analytics (TBI)
4. Cyber Phyiscal Systems and Digital Twin for Smart Manufacturing.
5. Prognostics of Highly Connected and Complex Systems including Wind Turbine/Wind Farm, High Speed Train, Aircraft Fleet, Factoty Automation and Robot Lines, Semiconductor Fab., etc.
6. Intelligent Maintenance Systems for Mechanical Components (bearing, ball screw, pump, valve, motors, etc) and sensors/sensory systems.
Partial List of Current Research Projects:
Winbond Electronics Corporation |
Winbond Electrostatic (ESC) Chuck Remaining Useful Life |
Hitachi High-Technologies Corporation |
Phase V: Chamber Matching based on Etching Digital Twin Model |
Applied Materials, Inc. |
Multivariate Simulation Dataset Generation for Fault Detection with Semi-Automated Feature Extraction and Semi-supervised Limits Setting Applied to Semiconductor Manufacturing Processes |
Mitsubishi Electric Corporation |
Health Assessment & Fault Detection for Industrial Robots |
Hitachi High-Technologies Corporation |
Phase 2021-1: Development of digital twin model for Hitachi plasma etching tool and calibration of tool performance shift |
United Microelectronics Corporation (UMC) |
Applications of Machine Learning Operation Techniques |
National Institute of Standards and Technology |
Industrial Artificial Intelligence Consortium to Advance High-Mix Production Systems |
MxD |
Predictive Maintenance Analytics of Roll-to-Rolll Manufacturing |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1. Industrial AI Course
Introduction to Industrial Artificial Intelligence, Fall 2023
Instructor: Prof. Jay Lee
Pre-requisites: Basic programming skills in Matlab or Python Fundamental knowledge in statistics, linear algebra, calculus and digital signal processing
Homework & Presentation Midterm Test Final Term-Project Report
20% 40% 40%
Course Introduction:
In today’s competitive business environments, companies have urgent needs to use advanced analytical tools to manage their industrial data to gain more insights of their operations. Those insights include machine health condition, system remaining useful life, system performance, and other key performance indicators. With the advent of networked devices, an abundance of data now exists in virtually every level of industry, from the individual manufacturing asset to the manufacturing facility as well as the entire organization. This data is often not used to its greatest potential. It is often acquired, stored and forgotten. Most organizations understand the need for acquiring data, but do not understand how to leverage the data to enhance their system design, condition monitoring and decision-making capabilities. This course will introduce students to advanced technologies—such as prognostics and health management (PHM), intelligent networks, and smart machine systems, and intelligent maintenance systems—that ultimately enable the conversion of industrial big data into actionable information that can be used to improve the design, the productivity and the efficiency of manufacturing operations.
Students will learn to utilize intelligent tools for converting big data to information with a focus on monitoring, assessing, predicting and diagnosing the condition of industrial assets. These tools will be used in conjunction with advanced platforms for intelligent system design and implementation. Students will follow a proven systematic process, which includes instrumentation, experimentation, data acquisition and data analysis. Students will be divided into groups to conduct term-project in applying this process for industrial big data applications.
Textbook:
Jay Lee, Industrial AI, Springer, Industrial AI 2020 https://link.springer.com/book/10.1007/978-981-15-2144-7
References
▪ Bishop, Christopher M. Pattern recognition and machine learning. springer, 2006.
▪ Gelman, Andrew, et al. Bayesian data analysis. Vol. 2. London: Chapman & Hall/CRC, 2014.
▪ Randall, Robert Bond. Vibration-based condition monitoring: industrial, aerospace and automotive applications. John Wiley & Sons, 2011.
Midterm & Final:
One midterm quiz will be given during the semester. No make-up exams will be given unless students have an emergency or made pre-arrangement with instructor. Unless otherwise noted, all tests will be closed book and closed notes. All students need to do a term-project in the e-Manufacturing lab and submit the report for final grade.
Homework:
Homework will be assigned and collected on a regular basis. All homework problems or sets will be graded in detail. Due date will be announced at the time when the homework is assigned. No late homework is accepted except under extenuating circumstances as determined by the instructor.
Course project:
Students will be divided into groups to finished course projects. The course projects are all industrial big data analysis related topics. The course projects requires students to use analytical tools learned from this course to deal with real problems and which is also not limited to only use what you learned here. Innovative solutions and ideas are encouraged by bonus points.
Attendance:
Students are expected to attend all classes (please be on time). Even though most of the course materials are well covered by the posted lecture notes and text, there may be some information that is discussed exclusively in class. If you missed class, you are still responsible for learning the materials taught during your absence.
Disclaimer:
Assignments and course content are subject to modification when circumstances or sound pedagogy dictate and as the course progresses. If changes are made, you will be given due notice.
Academic Integrity:
The University Rules, including the Student Code of Conduct, and other documented policies of the department, college and university related to academic integrity will be enforced. Any violation of these regulations, including acts of plagiarism or cheating, will be dealt with on an individual basis according to the severity of the misconduct.
-2-
Class Schedule
week |
date |
Topics |
Assignment |
1 |
|
Overview of Industrial Big Data Analytics and Industrial AI (I) |
HW1 |
2 |
|
Overview of Industrial Big Data Analytics and Industrial AI (II) |
|
3 |
|
Data Issues on Industrial Big Data System: Data Source, Data Quality, and Data Context |
|
4 |
|
Fundamentals of Signal Processing and Data Analytics |
HW2 |
5 |
|
Invited Speaker: David Siegel (TBD) Introduction of Open Source Analytics Tools (R, Python, SAS, etc.) and Platforms (Spark, etc.) |
|
6 |
|
Invited Speaker: (TBD) Machine Health Monitoring using Industrial Big Data: Case Study I: Machine Level Heath Monitoring Case Study II: Robot Health & Production Systems |
HW3 |
7 |
|
Invited Speaker: (TBD) Industrial Big Data for Networked and Distributed Systems: Case Study II: Wind Turbine and Wind Farm Case Study III: Smart EV Battery & Mobility |
HW4 |
8 |
|
Homework Review and Midterm Review |
|
9 |
|
Midterm |
|
10 |
|
Introduction to IoT and Cloud Platform (Advantech, TBD) |
|
11 |
|
Introduction of Group Projects and Examples Industrial Big Data Research Special Topic I (Deep Learning, feature engineering, etc.) |
Final Projects |
12 |
|
Course Project Discussions |
|
13 |
|
Course Project Discussions |
|
14 |
|
Course Project Discussions |
|
15 |
|
Final Presentation |
|
16 |
|
Final Project Report Due |
-3-
Assignments:
Assignment #1:
• The important publications, government reports, and white papers are made available to the students.
• Survey the literature to get an understanding of Industrial 4.0, digital transformation in manufacturing, cyber-physical systems.
• Develop a survey report and a 5-minute presentation to share their vision with the class.
• The homework will be finished by individual student.
Assignment #2:
• A bearing vibration dataset and specific homework requirements will be given to the student. The dataset is generated from the vibration test in the IMS center.
• Sample code for Fast Fourier Transform and logistic regression will be provided to the student to finish the homework.
• The student will have to finish the homework through collaboration with other group members.
• Each group must prepare a 5-minute presentation to present their work in the class.
Assignment #3:
• The students are asked to use the Self-Organizing Map (SOM) and SOM-MQE to detect and diagnose the bearing failure. Sample code for SOM and SOM-MQE will be provided to the class to finish the homework.
• The assignment will be finished by the group, and each group is asked to report their work in the class through a 5-minute presentation.
Assignment #4:
• Students are asked to use the Support Vector Machine (SVM) and other algorithms to detect and diagnose the bearing failures. Sample code for SVM will be provided.
• The assignment will be finished by the group, and each group is asked to report their work in the class through a 5-minute presentation.
Invited Speakers:
The class will invite three speakers from the industry to share their strategic vision about industry 4.0, smart manufacturing, and other advanced topics in manufacturing. In the past, the speakers from IMS member companies, e.g. Taiwan Semiconductor Manufacturing Company, Protector & Gamble, Mazak, Hitachi, Predictronics, were invited to speak in the class. These invited speakers will give a one-hour lecture in the class.
Invited Lecture Session:
The class also has lecture sessions about the Internet of Things, Deep Learning and advanced analytics, cybersecurity in manufacturing. These sessions will be delivered either by IMS
-4-
Final Projects:
Project 1: Fault detection in semiconductor Etching
Project 2: Gearbox fault diagnosis and prognosis
Project 3: Bearing fault diagnosis and prognosis.
Project 4: Virtual metrology for critical semiconductor manufacturing processes
Project 5: Machine tool degradation prognosis and remaining useful life prediction.
Project 6: Aero-engine remaining useful life prediction
In the final project, each student group will focus on one of the 6 projects that are listed above. The dataset and detailed requirements will be given to students to start the project. TA will be assigned to each group to provide support and answer questions. Each group will work independently to fulfill the requirements in the final project. In the exam week, each group will make a 15-minute presentation to report their work and document all the details in a report. When promising results are achieved or novel methods are proposed, the instructor will advise the student group to publish papers at conferences or research journals.
The concept of design these projects is to motivate students to learn by address real-world problems independently. The datasets used in these final projects are from IMS past research projects or open data competitions. They are all real-world datasets generated from machines or production lines. Although each student group will focus on one of the 6 projects in this class, the datasets of all these projects will be made accessible to all the students for their future use.
- Researcher ID: https://researchid.co/jay.lee
- Google Scholar: https://scholar.google.com/citations?user=g9GtqgQAAAAJ&hl=en&oi=ao
- ResearchGate: https://www.researchgate.net/profile/Jay-Lee-27
UMD Hosts 3D Maryland Expert Group Meeting
Event Highlights 3D Printing, Additive Manufacturing Technologies at UMDJaydev Desai wins seed grant for diagnostic robotic system
Research will aid in prevention and treatment of atrial fibrillation.S.K. Gupta and teammates win 2007 'Invention of the Year' physical science award
Gupta also finalist in information science category; Gary Rubloff finalist in physical science.- American Society of Mechanical Engineers (ASME), Society of Manufacturing Engineers (SME), Prognostics and Health Management (PHM) Society, International Society of Engineering Asset Management (ISEAM)