Welcome to the webpage of the Data-Driven Process Systems Engineering Lab! We are a computational research group in the School of Chemical & Biomolecular Engineering at the Georgia Institute of Technology. Our work lies in the Process Systems Engineering field, with applications in energy, process intensification and manufacturing of pharmaceuticals and bioproducts. Our aim is to integrate new developments in data-science with traditional chemical engineering fundamentals, to develop the new generation of modeling and optimization tools for complex multiscale systems.  

Recent News

New publication!

Atlanta, GA

March 25, 2019

Jackie published her first paper on variable selection algorithm for surrogate modeling. The article can be found here. Congratulations!

Journal of Global Optimization Best Paper Award

Atlanta, GA

April 15, 2019

Dr. Boukouvala won the 2017 Best Paper Award in Journal of Global Optimization for her paper on the global optimization of general constrained grey-box models and its application to pressure swing adsorption. The award carries a 1000 USD prize.  The…

Optimization Letters 2017 Best Paper Award

Atlanta, GA

December 05, 2018

Fani won the 2017 Best Paper Award in Optimization Letters for her paper on the ARGONAUT algorithm for global optimization of grey-box problems. The OPTL Best Paper Award carries a 500 USD prize. The co-author of this paper is the…

Latest Publications

Nonlinear Variable Selection Algorithms for Surrogate Modeling

AIChE Journal 2019

Jianyuan Zhai Fani Boukouvala

Having the ability to analyze, simulate and optimize complex systems is becoming more important in all engineering disciplines. Decision-making using complex systems usually leads to nonlinear optimization problems, which rely on computationally expensive simulations. Therefore, it is often challenging to…

Optimization of black-box problems using Smolyak grids and polynomial approximations

Journal of Global Optimization 2018

C.A. Kieslich F. Boukouvala C.A. Floudas

A surrogate-based optimization method is presented, which aims to locate the global optimum of box-constrained problems using input–output data. The method starts with a global search of the n-dimensional space, using a Smolyak (Sparse) grid which is constructed using Chebyshev extrema…