Development of a freeze-dried mixture of Nile tilapia (Oreochromis niloticus) croquette using a GA-based multiobjective optimisation.

Background: The mechanically separated meat (MSM) of Nile tilapia is an example of a by-product that can be used in the development of new foods. The aim of this study was to optimise the mixture of different flours in the development of a freeze-dried mixture of fish croquette using Nile tilapia MSM. Flavour, texture and overall acceptance of seven formulations were evaluated by an acceptance test. A genetic algorithm (GA) with desirability functions was combined with a multiobjective optimisation of the response surface models.
Results: The combination of flours was chosen to minimise cost and maximise overall acceptance and fibre content. Overall acceptance showed a statistically significant correlation (P ≤ 0.05) with flavour (r = 0.67) and texture (r = 0.61). The GA-based approach indicated that the highest overall acceptance was obtained when using wheat and rye flours in equal parts. This formulation had an overall acceptance of 7.52, a fibre content of 11.50 g kg⁻¹ and a cost of US$2.21/kg. After 24 h of freeze-drying, the water activity of the mixture was 0.11.
Conclusion: The GA-based approach was able to optimise the croquette formulation. The freeze-drying process contributed to the development of a value-added product with high quality and long shelf-life.
(© 2012 Society of Chemical Industry.)