Optimization of frying process for maintaining nutritional quality to satisfy consumers' sensory attributes: A novel application of multi-criteria decision-making approach

Abstract

The objective of the study was to assess the optimum frying condition of fish considering the multiple perspectives (retention of nutritional quality indices [NQI], reduction of preparation time, and improvement of health benefit) to satisfy consumer-preferred sensory attributes by controlling the most impactful process variables (temperature, time, and oil amount). The multi-criteria decision-making (MCDM) approach is appropriate to handle the numerous conflicting criteria and numerous multiple objectives. First, an artificial neural network (ANN) model was developed to build a non-linear correlation between the cooking process parameters and NQI by an automatic exhaustive search of all available algorithms and activation functions. All the NQI are conflicting in nature. Therefore, the ANN-based multi-objective genetic algorithm was implemented to obtain the Pareto optimal solutions to improve all NQI simultaneously. Five optimised conditions were selected amongst the Pareto optimal solutions, satisfying the above-mentioned multiple criteria. Finally, a well-known MCDM approach, the analytical hierarchy process (AHP), was applied for sensory analysis to evaluate the overall acceptance of the optimised conditions based on the relative importance of consumers' general sensory modalities (flavour, colour & appearance, and taste). Furthermore, the following condition (140.01°C, 7.62 min, 47.87 ml oil/kg of fish) was selected as the most accepted in terms of all quality attributes that may be implemented as the standard condition in domestic and industrial purposes.