Prediction of Mass Transfer during Osmotic Dehydration of Black Fig Fruits (Ficus carica) in Ternary Systems: Comparison of Response Surface Methodology and Artificial Neural Network

Osmotic dehydration of fig fruits (cv. Sabz) in ternary solution of water, sucrose and sodium chloride at different solution concentrations, temperature and process durations were analyzed. A comparative approach was made between artificial neural network (ANN) and response surface methodology (RSM) to predict the mass transfer parameters. Results showed that all independent variables positively decreased the weight meaning that increasing each factor resulted in increasing weight loss and this relationship was linear. Osmo-dehydrated figs had better quality compared to samples without osmosis. All four independent variables explained 94% of the weight loss, 90% moisture content reduction and 89% of the solid gain. The determined optimum processing conditions were temperature of 60°C, sucrose concentration of 70%, sodium chloride concentration of 5% and immersion time of 5h. The results showed that properly trained ANN model is found to be more accurate in prediction as compared to RSM model.