Systems Research Institute
for Chemical and Biological Processes
The Systems Research Institute (SRI) for Chemical and Biological Processes, established in 2004, aims to foster interdepartmental research among researchers in engineering, the sciences and the medical community. Dr. Christos Georgakis, is the Director of the Institute. The Institute's mission includes:
- Fostering research projects in the modeling, optimization, monitoring and control of chemical and biological processes; either batch or continuous;
- Cultivating inter-disciplinary academic collaborations with other Tufts departments and schools, as well as with other Universities;
- Develop strong industrial interactions.
Industrial Consortium on Data-Driven Modeling (DDM)
The Institute's present initiative, the industrial consortium on Data Driven Modeling (DDM) of batch and continuous processes, is currently on a fast start-up mode. The DDM consortium is a industry/University cooperative effort for new methodologies and tools that exploit the large availability of process data for the development of models that easily enable the optimization, control and monitoring of industrial batch or continuous processes.More information
Besides the collaborative availability of several faculty members at Tufts, the SRI has secured the collaboration of three very prominent academics outside Tufts: Professors Bakshi, Bonvin and Ydstie at Ohio State, EPFL in Lausanne (Switzerland), and CMU, respectively. More information
Recent Research Breakthroughs
The initial research areas of the planned DDM consortium are motivated from three resent research advances in the area of data-driven models or meta-models for batch and continuous processes.
- Design of Dynamic Experiments: In a recently published paper the classical Design of Experiments (DoE) methodology has been generalized to the Design of Dynamic Experiments (DoDE). This enables the quick and efficient optimization of batch and semi-batch (or fed-batch) processes with respect to a time varying input variable.
- Simple and Accurate Meta-Motels: In a second paper, it was demonstrated that simple metamodels could accurately approximate complex process models of continuous process. This can substantial facilitate complex and time-consuming calculations related to plant-wide optimization, operability studies and plant-wide control.
- Non-Linear Sensitivity Methodology: In a third advance at a recently defended MS thesis, a nonlinear sensitivity methodology has been developed and applied to a metabolic model of a CHO cell to identify which of the 51 model parameters are the most important to estimate from experimental data and which model parameter are inconsequential.
- SRI Director
- Dr. Christos Georgakis
- Professor, Department of Chemical and Biological Engineering, Gordon Senior Faculty Fellow of Systems Engineering
- Affiliated Faculty at Tufts
- Dr. Kyongbum Lee
- Associate Professor and Chair, Department of Chemical and Biological Engineering
- Dr. Nikhil Nair
- Asistant Professor, Department of Chemical and Biological Engineering
- Affiliated Faculty Outside Tufts
- Dr. Bhavik Bakshi
- Professor , Departmants of Chemical & Biomolecular Engineering, Ohio State University
- Dr. Dominique Bonvin
- Professor, Automatic Control Laboratory École Polytechnique Fédéral de Lausanne (EPFL) Switzerland
- Dr. B. Erik Ydstie
- Professor, Departmant of Chemical Engineering, Carnegie Mellon University, Professor of Electrical Engineering by Courtesy, CMU , and Professor II of Electrical Engineering at NUST, Trondheim, Norway
- Contact SRI
- Tufts University
- Science and Technology Center
- 4 Colby Strett, Room 273
- Medford, MA 02155
(617) 627-3991 [FAX]