Abstract
The fermentation medium is of particular importance for productivity of fermentation and economic success of a bioprocess. Medium formulation addresses improvement of the nutritional and chemical environment of cells in a bioreactor and represents a significant cost and time factor in bioprocess development. This article surveys several techniques for fermentation medium development and some of their recent applications. Classical methods such as factorial design and response surface methodology are compared to modern stochastic search strategies and machine-based learning techniques (genetic algorithms and artificial neural networks). Multiobjective problems are introduced, considering that an economic fermentation process requires high productivity with minimal investment of cost and time. Further, the pros and cons of experimental platforms such as shake flasks, microtiter plates, and miniaturized bioreactors are discussed.
Original language | English |
---|---|
Title of host publication | Comprehensive Biotechnology |
Publisher | Elsevier |
Pages | 199-213 |
Number of pages | 15 |
ISBN (Electronic) | 9780444640475 |
ISBN (Print) | 9780444640468 |
DOIs | |
State | Published - 1 Jan 2019 |
Keywords
- Fermentation
- Genetic algorithm
- Miniaturization
- Multi-objective
- Neural networks
- Optimization
- Statistical design
- Stirred-tank bioreactor