Manufacturing Data Science: from Predictive to Prescriptive Analytics
Fri, 03 Jul
|Online
What's the point of predicting the future if you don't know what to do about it? Prof. Chia-Yen Lee presents a practical framework for bridging predictive and prescriptive analytics in manufacturing — turning data into decisions that actually drive action.
Registration closes 03 Jul 2026, 2:00 pm GMT+2
Time & Location
03 Jul 2026, 1:00 pm – 2:00 pm GMT+2
Online
Date/Time in Local Time Zones
🕖 Friday, 3 July 2026, 7:00 PM — Taiwan Standard Time (TST, UTC+8)
🕖 Friday, 3 July 2026, 7:00 PM — China Standard Time (CST, UTC+8)
🕗 Friday, 3 July 2026, 8:00 PM — Japan Standard Time (JST, UTC+9)
🕚 Friday, 3 July 2026, 11:00 PM — New Zealand Standard Time (NZST, UTC+12)
🕖 Friday, 3 July 2026, 7:00 AM — Eastern Daylight Time (EDT, UTC−4)
🕓 Friday, 3 July 2026, 4:00 AM — Pacific Daylight Time (PDT, UTC−7)
🕐 Friday, 3 July 2026, 1:00 PM — Central European Summer Time (CEST, UTC+2)
Manufacturing Data Science: from Predictive to Prescriptive Analytics
This talk introduces smart manufacturing as a multi-objective, decision-oriented system that leverages computational intelligence and self-learning capabilities to autonomously optimize manufacturing resources and processes. The talk identifies common pitfalls and protocols of data science in manufacturing practice. In particular, a Five-Phase Analytics framework is proposed to provide a guideline for methodological development. The framework focuses on creating synergy by integrating predictive and prescriptive analytics and emphasizes 'Prediction is Process, Decision is Purpose' to support proactive decision-making. Several empirical studies are presented, including raw material price prediction and procurement, prognostics and health management (PHM), chiller energy-saving optimization, and spatio-temporal anomaly diagnosis. In addition, we identify a Death Valley from predictive to prescriptive analytics, Optimization-Guided Learning (OGL) is developed to address the issue and enhance decision robustness in an analytics loop.
Speaker Biography
Chia-Yen Lee received the Ph.D. degree from the Department of Industrial and Systems Engineering, Texas A&M University, USA, in 2012. He is currently a Professor in the Department of Information Management, Associate Dean in the College of Management, and Director for Entrepreneurship and Innovation MBA (EiMBA), National Taiwan University. He was a Director and Professor in the Institute of Manufacturing Information and Systems, National Cheng Kung University (NCKU), Taiwan. He has temporarily transferred as Deputy Director in Taiwan Semiconductor Manufacturing Company (tsmc). He currently serves as Senior Editor for IEEE Transactions on Automation Science and Engineering, and Associate Editor for IEEE Transactions on Semiconductor Manufacturing. His research interests include manufacturing data science, productivity and efficiency analysis, optimisation-guided learning, and trustworthy AI. He has authored or co-authored over 65 papers in leading journals and received over 50 grants from industry-academia collaborations, including both high-tech and traditional manufacturing industries.
Registration closes 03 Jul 2026, 2:00 pm GMT+2