Genetically Optimized Neural Network Systems (GONNS) are introduced to find the optimal operating conditions for complex systems. GONNS uses the multiple back-propagation-type neural networks to represent the characteristics of the considered system and estimates the optimal operating conditions by using genetic algorithms. Back-propagation-type neural networks represented the simulated systems of the test cases with an average error of less than 1%. The genetic algorithms converged to the global minimum in less than 350 iterations having an error of less than 2%. The GONNS is an excellent decision-making aid for selection of manufacturing process and preparation of ideal diet.