ChE SEMINAR: Big Data and Digitization: How Process Systems Engineering can contribute

Details

Title: Big Data and Digitization: How Process Systems Engineering can contribute

Biosketch:
Marianthi Ierapetritou is the Bob and Jane Gore Centennial Chair Professor in the Department of Chemical and Biomolecular Engineering at University of Delaware. Prior to that she has been a Distinguished Professor in the Department of Chemical and Biochemical Engineering at Rutgers University. During the last year at Rutgers University, she led the efforts of the university advancing the careers in STEM for women at Rutgers as an Associate Vice President of the University.
Dr. Ierapetritou’s research focuses on the following areas: 1) process operations; (2) design and synthesis of flexible production systems with emphasis on pharmaceutical manufacturing; 3) energy and sustainability process modeling and operations; and 4) modeling of biopharmaceutical production. Her research is supported by several federal (FDA, NIH, NSF, ONR, NASA, DOE) and industrial (BMS, J&J, GSK, PSE, Bosch, Eli Lilly) grants.
Among her accomplishments are the 2022 AICHE Excellence in Process Development Research Award, the appointment as the Gore Centennial Chair Professor in 2019, the promotion to distinguished professor at Rutgers University in 2017, the 2016 Computing and Systems Technology (CAST) division Award in Computing in Chemical Engineering which is the highest distinction in the Systems area of the American Institute of Chemical Engineers (AIChE), the Award of Division of Particulate Preparations and Design (PPD) of The Society of Powder Technology, Japan; the Outstanding Faculty Award at Rutgers; the Rutgers Board of Trustees Research Award for Scholarly Excellence; and the prestigious NSF CAREER award. She has served as a Consultant to the FDA under the Advisory Committee for Pharmaceutical Science and Clinical Pharmacology, elected as a fellow of AICHE and as a Director in the board of AIChE. She has more than 290 publications and has been an invited speaker to numerous national and international conferences.
Dr. Ierapetritou obtained her BS from The National Technical University in Athens, Greece, her PhD from Imperial College (London, UK) in 1995 and subsequently completed her post-doctoral research at Princeton University (Princeton, NJ).

Abstract
Manufacturing industry is quickly adapting tools and methodologies driven by digitization transformation. Data explosion, machine learning, and artificial intelligence are enablers to this revolution.
Process Systems Engineering (PSE) community needs to adapt the methodologies to fully leverage the additional resources for accurate representation of processes and detailed analyses. The approaches can facilitate process development, system analysis, and optimization, supporting goals in areas like sustainability, circular economy, and public health.
Process modeling continues to be an essential tool in PSE, serving to depict complex physical processes and their interconnections. Integrating information across diverse scales poses challenges, as does determining the level of data inclusion. To acquire comprehensive process understanding, efficient tools such as sensitivity analysis, feasibility analysis, life cycle assessment (LCA), and technoeconomic analysis (TEA) can be applied for analysis with data from experiments, pilot plants, databases, and/or models.
Using the developed models, optimization can be performed to identify optimal conditions of the most important variables identified using sensitivity analysis, while satisfying important operability and product quality constraints. Such in silico optimization results can provide insights to the experimental work, but as model complexity increases, the optimization task becomes computationally demanding. When implementing these tools, addressing data uncertainty emerges as a crucial concern. Employing uncertainty quantification (UQ) approaches becomes essential to tackle this issue.
In this talk, we will discuss our group’s work towards developing these tools and highlight their application in pharmaceutical advanced manufacturing and towards sustainable chemical production.