468324 Data-Driven Optimization Using an Evolutionary Design of Dynamic Experiments for Biopharmaceutical Processes

Friday, November 18, 2016: 5:27 PM
Continental 4 (Hilton San Francisco Union Square)
Zhenyu Wang and Christos Georgakis, Systems Research Institute and Chemical and Biological Engineering, Tufts University, Medford, MA

Biopharmaceutical processes are substantially complex and are not understood well enough to enable the development of an accurate knowledge-driven model for process optimization purposes. In such a case, a data-driven modelling approach provides a very attractive alternative, which, if designed properly, can quickly optimize the process with a small set of experiments. The recently introduced Design of Dynamic Experiments (DoDE) [1], a generalization of the traditional Design of Experiments (DoE) [2, 3], has demonstrated that it is an effective data-driven modeling and optimization approach [1, 4, 5].

Here we apply the DoDE method in an evolutionary manner to optimize biopharmaceutical processes while satisfying budgetary and developmental time constraints on the number of experiments in two representative biopharmaceutical processes; Penicillin fermentation [6] and Hybridoma cell culture [7]. We maximize the amount of product produced by determining the optimal feeding policy and batch duration.

To minimize the number of experiments we initially consider only a few dynamic sub-factors and the simplest Response Surface Model (RSM). After the initial results are analysed and an approximate RSM is estimated, additional experiments are added sequentially to further improve the data-driven model and the process performance. We demonstrate that with a reasonable amount of evolutionary experiments one can come very close to the model-based optimum. The model-based optimization [8], however, requires a very accurate knowledge-driven model that is rarely at hand.

References

1. Georgakis, C., Design of Dynamic Experiments: A Data-Driven Methodology for the Optimization of Time-Varying Processes. Industrial & Engineering Chemistry Research, 2013. 52(35): p. 12369-12382.

2. Box, G.E.P. and N.R. Draper, Response Surfaces, Mixtures, and Ridge Analysis. 2007, Hoboken, NJ: Wiley.

3. Montgomery, D.C., Design and Analysis of Experiments. 8 ed. 2013, New York: Wiley. 729.

4. Fiordalis, A. and C. Georgakis, Data-driven, Using Design of Dynamic Experiments, versus Model-driven Optimization of Batch Crystallization Processes. Journal Of Process Control, 2013. 23(2): p. 179-188.

5. Makrydaki, F., C. Georgakis, and K. Saranteas. Dynamic Optimization of a Batch Pharmaceutical Reaction Using the Design of Dynamic Experiments (DoDE): The Case of an Asymmetric Catalytic Hydrogenation Reaction. in Proceedings of the 9th International Symposium on Dynamics and Control of Process Systems. 2010. Oude Valk College, Belgium.

6. Bajpai, R.K. and M. Reuss, A Mechanistic Model for Penicillin Production. Journal Of Chemical Technology And Biotechnology, 1980. 30(6): p. 332-344.

7. De Tremblay, M., et al., Optimization of Fed-Batch Culture of Hybridoma Cells Using Dynamic Programming Single and Multifeed Cases. Bioprocess Engineering, 1992. 7(5): p. 229-234.

8. Biegler, L.T., An Overview of Simultaneous Strategies for Dynamic Optimization. Chemical Engineering And Processing, 2007. 46(11): p. 1043-1053.


Extended Abstract: File Not Uploaded