Data-Centric Optimization: Methods and Applications

2015 AIChE Annual Meeting in Exhibit Hall 1 (Salt Palace Convention Center)

November 08, 2015

As an engineering community we have been working towards developing detailed mathematical simulation models which accurately capture the behavior of complex systems. This trend is reinforced by our drive to couple multiscale information which ranges from the atomistic and molecular level to continuum, process and even plant and enterprise levels. At the same time, abundance of data is transforming traditional process systems engineering operations in this “Big Data” era, complementing existing mechanistic models or even enabling the description and analysis of problems for which a model is not available. However, as we move towards this paradigm we create a gap between the current state-of-the-art high-fidelity models, and the models which are used in design, optimization and control applications. My past and future research interests integrate theory from various fields, such as statistics, machine learning, operations research, mathematics and engineering, to solve significant problems which are characterized by abundance in experimental and/or computational data; lack of closed-form numerically stable models; and integration of data and models across scales. The common denominator in the presented work is development of global optimization methods for systems without mathematical closed-form equations and consequently, without derivative information [1-5]. The main application areas pursued are (a) design and optimization of pharmaceutical manufacturing systems [6-9], (b) design and optimization of adsorption systems for energy applications [5,10], (c) integrated refinery and chemicals planning. References: 1. F. Boukouvala, M.G. Ierapetritou. Feasibility analysis of black-box processes using an adaptive sampling kriging-based method. Computers & Chemical Engineering, 36, 2012. 2. F. Boukouvala, Y. Gao, F.J. Muzzio, M.G. Ierapetritou, Reduced order Discrete Element Method (DEM) modeling, Chem. Eng. Science, 95, 2013. 3. F. Boukouvala, M.G. Ierapetritou, Derivative-free optimization for expensive constrained problems using a novel expected improvement objective function, AICHE Journal, 60(7), 2014. 4. Boukouvala, F. , Misener, R., Floudas, C.A.: Global Optimization Advances in Mixed-Integer Nonlinear Programming, MINLP and Constrained Derivative-free Optimization, CDFO. European Journal of Operational Research. Under Review (2015). 5. Boukouvala, F., Hasan, M.M.F., Floudas, C.A.: Global Optimization of General Constrained Grey-Box Models: New Method and its Application to Constrained PDEs. Submitted to Journal of Global Optimization (2015). 6. F. Boukouvala, F.J. Muzzio, M.G. Ierapetritou. Design Space of pharmaceutical processes using data-driven based methods. Journal of Pharmaceutical Innovation, 5(3), 2010. 7. F. Boukouvala, V. Niotis, R. Ramachandran, A. Vanarase, F.J. Muzzio, M.G. Ierapetritou. Integrated approach for dynamic flowsheet modeling and sensitivity analysis of continuous pharmaceutical manufacturing. Computers and Chemical Engineering, 42, 2012. 8. F. Boukouvala, A. Chaudhuri, R. Zhou, F. Muzzio, M.G. Ierapetritou, R. Ramachandran, Dynamic flowsheet simulation of pharmaceutical tablet production through wet granulation, Journal of Pharmaceutical Innovation, 8(1), 2013. 9. F. Boukouvala, F.J. Muzzio, M.G. Ierapetritou, Surrogate-based optimization of expensive flowsheet models for continuous pharmaceutical manufacturing, J. of Pharm. Innovation, 8(2), 2013. 10. M.M.F. Hasan, F. Boukouvala, E. First, C.A. Floudas, Nationwide, Regional and Statewide CO2 Capture, Utilization, and Sequestration Supply Chain Network Optimization. Ind. & Eng. Chem. Research, 53(18), 2014.