Abstract
Screening is arguably the most important step in bioprocess development. The selection of the
best organism and cultivation conditions is crucial for the following steps of the development
process since important decisions are made at very early developmental stages which have a
signi cant impact on the overal performance in manufacturing. In this work we tackle the
challenges related to the conditional screening phase (where typically approx. #strains < 20,
medium, temperature, bioreactor settings, and bioprocess strategy are taken into account) by
designing and operating optimal experiments in robotic experimental facilities. We introduce
an NMPC framework that is able to operate 48 parallel mini-bioreactors aiming to maximize
the probability of selecting the best strain for the industrial process. To achieve this, the
experiment is designed online such that the uncertainty on the parameter estimates allows
the most reliable selection in screening considering the error propagation onto the outputs of
the simulated process.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of NMPC 2018 6th IFAC Conference on Nonlinear Model Predictive Control |
| Number of pages | 2 |
| Publication status | Published - 2018 |
| Event | NMPC 2018 6th IFAC Conference on Nonlinear Model Predictive Control - Duration: 19 Aug 2018 → 22 Aug 2018 |
Conference
| Conference | NMPC 2018 6th IFAC Conference on Nonlinear Model Predictive Control |
|---|---|
| Period | 19/08/18 → 22/08/18 |
Research Field
- Efficiency in Industrial Processes and Systems