Abstract:

Laboratory and computational experiments are often time- and resource-intensive; thus, maximizing their value and information gain is essential. This seminar explores several paradigms for optimal experiment design. We will start by reviewing classical design of experiments methods and modern approaches such as Bayesian optimization. These approaches do not require mechanistic models and can be effective for relatively few experimental decisions, but high-dimensional decision spaces (e.g., time-varying experimental conditions) pose significant challenges.
The seminar focuses on methods that exploit mathematical models grounded in engineering science to formulate experiment design as a dynamic optimization problem that maximizes expected information. Using membrane separations for critical minerals and materials recovery as a central application, I will show how mechanistic models, parameter estimation, and uncertainty quantification connect laboratory measurements to predictive models and process design. I will highlight recent advances in Pyomo.DoE, an open-source package for model-based design of experiments (MBDoE), developed by my group in collaboration with Sandia National Laboratories and others. Finally, I will introduce DoE-GPT, our emerging collection of agentic workflows that augment large language models to guide scientists and engineers through statistical analysis and experimental planning. Throughout the seminar, I will discuss practical lessons for choosing informative experiments, validating assumptions, and translating data into defensible engineering decisions. I will conclude with opportunities for experiment automation, e.g., self-driving labs, grounded in engineering science and mechanistic models.
Select References
Wang and Dowling. “Bayesian optimization for chemical products and functional materials.” Current Opinion in Chemical Engineering, 2022, 36, 100728. https://doi.org/10.1016/j.coche.2021.100728
Wang and Dowling. “Pyomo.DoE: An Open-Source Package for Model-Based Design of Experiments in Python.” AIChE Journal, 2022, 68(12), e17813. https://doi.org/10.1002/aic.17813
Laky et al. “Optimal Experimental Design Using Eigenvalue-Based Criteria with Pyomo.DoE.” Digital Discovery, 2026. https://doi.org/10.1039/D6DD00160B
Wamble et al. “Optimal Diafiltration Membrane Cascades Enable Green Recycling of Spent Lithium-Ion Batteries.” ACS Sustainable Chemistry & Engineering, 2022, 10, 12207–12225. https://doi.org/10.1021/acssuschemeng.2c02862
Liu et al. “Characterizing Transport Properties of Surface Charged Nanofiltration Membranes via Model-Based Data Analytics.” Industrial & Engineering Chemistry Research, 2025, 64, 12111–12130. https://doi.org/10.1021/acs.iecr.4c04763
Liu et al. “Optimization of Reactive Ink Formulation for Controlled Additive Manufacturing of Copolymer Membrane Functionalization.” ACS Applied Materials & Interfaces, 2024, 16, 59216–59233. https://doi.org/10.1021/acsami.4c11749
Yao et al. “Robust Optimization of Flexible Diafiltration Systems for Critical Mineral Separations.” Chemical Engineering Journal Advances, 2026, 27, 101282. https://doi.org/10.1016/j.ceja.2026.101282


Bio:

Alexander (Alex) Dowling is the Tony and Sarah Earley Collegiate Associate Professor of Energy and the Environment at the University of Notre Dame (Indiana, USA). His research combines chemical engineering, computational optimization, machine learning, and data science, organized in three research themes: (1) molecular-to-systems (multiscale) modeling and optimization, (2) optimal design of experiments and statistical inference, and (3) machine learning for bridging timescales.
Application domains span sustainable energy, environmental management, and systems biology. Prof. Dowling has been recognized with several awards, including the NSF CAREER award, the CAST Outstanding Young Researcher Award, the Junior Sargent Medal, and the James A. Burns, C.S.C., Award. He holds a B.S.E. from the University of Michigan - Ann Arbor and a Ph.D. from Carnegie Mellon University, all in chemical engineering.