Optimum design of swine feed formulations were carried out using a genetic algorithm in the C language. Twenty different rations of feed ingredients such as corn, grain sorghum, soybean meal, cassava tuber and others were designed under constraint conditions of specified nutrient contents. Examples of the nutrient contents utilized were crude protein, crude fat, crude ash, total digestive nutrition and calcium.
The genetic algorithm was used to minimize the cost of each specific ration. The nutrient content of each ration was based on previous swine breeding trials. Feed formula design simulations were carried out using a range of parameters such as the probability of mutation and population size of the genetic algorithm. Results of the formula optimization by the genetic algorithm were compared with results obtained from a linear programming model.
Based on this study, it was found that both simulations gave nearly equal results for rations of feed ingredients and nutrient contents. About 87% of the optimum designed feed formula corresponded to grain sorgum, corn, soybean meal and cassava tuber. On the other hand, differences of convergence process to optimal solutions due to population size and the probability of mutation were also cleared. This research shows that the genetic algorithm is capable of optimizing feed formulations.
View full abstract