The batch fermentations were conducted using lactose as the substrate at pH
6.5 and temperature 30 degrees C. Average batch kinetic data was eventuall
y used to develop ail unstructured mathematical model. The kinetic paramete
rs of the model were determined by non-linear regression technique using th
e batch experimental results. Parametric sensitivity analysis showed the ma
ximum specific substrate consumption rate (r(s)(max)) and the maintenance e
nergy constant (m(s)) to be the most sensitive parameters. The experimental
observations in batch fermentation were close to the model predictions. Th
e batch model was extrapolated to identify nutrient feeding strategies, whi
ch were tested successfully for two different fed-batch fermentations. It d
emonstrated enhanced propionic acid productivity. The developed model was f
ound suitable for the design of feeding strategies to increase propionic ac
id production in fed-batch mode of reactor operation. (C) 2000 Elsevier Sci
ence S.A. All rights reserved.