The Department enjoys one of the largest research groups in the world in Process Systems Engineering (PSE), an area central to chemical and biochemical process industries. With the emerging focus on biomolecular engineering, our PSE group has reshaped itself as the Chemical and BioSystems Thrust (ChemBioSys) to provide the much-needed systems perspective in biology and biotechnology. The main objective of the Thrust is the development of efficient methodologies and tools to obtain innovative and non-intuitive solutions for the design and operation of chemical and biomolecular systems. Traditionally, the systems engineering approach has helped a variety of industries such as chemical, food, petroleum, polymer, pharmaceutical, agrochemicals, bio-pharmaceuticals, cosmetics, electronic materials, and semiconductor achieve safe/stable operation and enhanced business profits. Now, the recent developments in life sciences have opened up formidable challenges and unique opportunities in areas such as systems biology and bioinformatics, which require systems approach. Therefore, chemical sciences coupled with sophisticated computational techniques – ranging from statistical data analysis and optimization to artificial intelligence – provide an excellent platform for deriving deep insight into biological systems.

Current research activities in chemical and biosystems engineering in the Department include, but are not limited to,

Advanced Process Control

A significant component of our research is in the area of advanced process control. Activities in this area include closed-loop system identification, control methodologies for decentralized control systems and multivariable processes, process monitoring, and control loop performance assessment. Fuzzy logic control has been applied to non-linear systems. Data-based control strategies and monitoring tools for process fault detection and diagnosis are also in progress to benefit from the increasing number of variables that are measured and stored in a modern plant. Tools and procedures are developed for measuring the performance of control loops and to determine the causes for poor loop performance. Some specific experimental systems of industrial relevance chosen for the above studies are pH control, on-line monitoring and control of crystallization processes in protein and pharmaceutical systems, and control of fermentation reactors.

Artificial intelligence, Supply Chain and Logistics

In recent years, our activities have expanded into exciting contemporary domains such as artificial intelligence applications, supply chain management & logistics, and ab initio kinetic modeling for green chemistry and product design. Work has been initiated on the optimal design of multi-functional and miniaturized process units. The main objective is to develop efficient methodologies and tools to obtain innovative and non-intuitive solutions for the design and operation of these complex systems. The current research encompasses a wide range of length and time scales, ranging from individual molecules to global clusters of multinational enterprises in the context of supply chain management, and from nanoseconds for elementary reactions to months and even years of plant operations. Illustrative examples include computer-aided design of molecules and catalysts, design and evaluation of benign processes, alarm and abnormal situation management, scheduling of non-continuous multi-product batch plants, enterprise-wide modeling, and global supply chain management. The methodologies that are most commonly used for these problems include mixed integer mathematical programming, discrete optimization techniques, specialized heuristic procedures, artificial intelligence, multivariate statistics and signal processing. The fundamental contributions are in the areas of knowledge representation, knowledge extraction, and partially or fully automated decision support for complex, dynamic systems.

Biological Systems Engineering

Recent developments in life sciences have opened up formidable new challenges and opportunities. Chemical sciences coupled with sophisticated computational techniques, ranging from statistical data analysis and optimization to artificial intelligence, provide an excellent platform for deriving deep insight into biological systems and translating biological knowledge to disease treatment and biomarkers. Modeling and regulation of human physiological systems, and development of computational approaches for information modeling and analysis of biological systems are some of the topics currently pursued.

Design & Optimization

Modeling, design, and simulation of reaction, separation and coupled reaction-separation processes from first principles are common interests shared between ChemBioSys and Chemical Engineering Sciences. Artificial neural networks and regression techniques are used for modeling plant data and for predicting plant emissions and pollutants in order to reduce emissions at the source. Identification of process models using multivariate data constitutes another facet of modeling activities. Based on these modeling capabilities, multi-objective optimization of complex industrial processes, such as reformers for hydrogen production, crackers for ethylene, and styrene reactors are carried out using an adaptation of the non-dominating sorting genetic algorithm and simulated annealing. The optimization studies lead to potentially significant cost savings and enhanced productivity. Reliable equation-solving methods for phase equilibrium calculations are developed and applied to simulation and optimization of multi-phase distillation. Stochastic optimization methods are also used for phase equilibrium calculations by global minimization of free energy.

Faculty Members working in this Research Area