GOALI: Real Time Multivariable Statistical Monitoring of Batch Pharmaceutical Processes

Illinois Institute Of Technology, Chicago IL

Investigators

Abstract

Many pharmaceutical products are manufactured by batch and fed-batch fermentation processes with complex reaction mechanisms. Slight changes in operating conditions during critical periods of the batch run may have significant influence on growth and differentiation of organisms, and impact final product quality and yield. Consequently, improvements in real-time process monitoring, fault diagnosis, and process control will enhance product quality and yield, and reduce the number of rejected batches. An objective of this project is to develop batch process monitoring methods based on functional data analysis. The project will entail working with Abbott Labs engineers to develop and test such rigorous statistical methods for looking at the trajectories of process data. When used in a real-time vs. retrospective fashion, the data analysis methods can enable (1) improved end point prediction and (2) the prescription of corrective action to reacquire high yield and desirable product properties when a batch or fed-batch process strays off course.