Statistical evaluation of medium components using plackett-burman experimental design and kinetic modeling of lipase production by bacillus sphaericus

Evaluation of twelve medium components on lipase production by Bacillus sphaericus MTCC511 in submerged batch fermentation was studied using the Plackett-Burman statistical experimental design. The most significant variables affecting lipase production were found to be glucose, olive oil, peptone, NaCl and MnSO₄·H₂O. Maximum lipase activity of 2.82 U ml^-1 was obtained in the batch fermentation. The logistic model for cell growth, the Luedeking-Piret model for lipase production, and the modified Luedeking-Piret model for substrate utilization were capable of predicting the fermentation profile with high determination coefficient (R²) values of 0.9977, 0.9497 and 0.9962 respectively. The results indicate that this statistical method is an efficient tool in detecting the most significant medium components with a minimum number of experimental trials, and the unstructured models were able to describe fermentation kinetics more effectively. The validity of the proposed model enables prediction of the optimal fermentation conditions.