Comparison of RSM, ANN and Fuzzy Logic for extraction of Oleonolic Acid from Ocimum sanctum

Abstract

In Bioprocess Engineering down streaming process is very important step for extraction and purification of secondary metabolites. Already statistical methods i.e. Response Surface Methodology (RSM) utilized to optimized the different types of extraction parameters such as extraction time, particle size, solvent-solid ratio, solvent composition on maximum extraction of bioactive compounds. But these methods do not deal accurately with non-linearity. Regression-based RSM requires the order of the model to be stated (second, third or fourth order). Intelligent computing methods (ICMs) such as: artificial neural network (ANN) and fuzzy logic (FL) are shown to be important tools to deal with these problems. In this work, we have implemented RSM, ANN and FL for the extraction of oleonolic acid from Ocimum sanctum. The results obtained by the RSM, ANN and FL are compared with experimental results. The correlation coefficient and root mean square error (RMSE) of the experimental result with the ICMs is computed and shown in graphical form. It is observed from the result that the correlation coefficient and RMSE for the ANN was highest when compared to FL and RSM. The objective of work is to optimize and compare various extraction parameters using ICMs which deals with non-linear nature of bioactive compounds irrespective of existing RSM techniques.