The present study dealt with coupling a newly emerged optimisation algorithm, colonial competitive algorithm (CCA), with neural network models to minimise the moisture and fat content of deep-fat fried mushrooms and to compare its performance with genetic algorithm (GA). The objective function, consisted of neural network-genetic algorithm models, was formulated as a normalised composite function to simultaneously search for optimum decision variables of frying process of mushrooms, i.e., frying temperature, osmotic dehydration, gum coating, and frying time. Initially, to set the control parameters of CCA accurately, fine-tuning was performed for key control parameters based on convergence rate, calculation time, and best cost value. The optimisation results revealed that CCA has a better efficiency in comparison with GA in terms of faster convergence rate and shorter calculation time. The results of optimum values of responses inferred that both CCA and GA has led to similar results but the optimum levels of decision variables were different for the two algorithms. Overall, it was found that CCA could be used as an efficient optimisation algorithm to handle with different optimisation problems in food processing fields.
