Scientific Paper Análisis prospectivo de la base alimentaria en una lechería tropical con programación lineal Prospective analysis of the feeding basis in a tropical dairy farm with linear programming

The objective of the study was to analyze prospectively the feeding basis in a tropical dairy farm, through an optimization model. The variables of this model were: star grass, leucaena associ- ated with common Guinea grass, sugarcane and CT-115. The requirements were estimated for cows that weighed 425 kg of LW, with a production of 4 kg/animal/day, and eight constraints were established. The model was solved with the program WinQSB version 2.0. The objective function was to maximize the DM production per hectare. The optimal solution indicated that the highest part of the area should correspond to star grass, with 45 % of the total. The sensitiv- ity analysis showed that the DM production per hectare of sugarcane can decrease to 7,64 t and increase to an infinite value, without affecting the optimal solution obtained. In the case of star grass, the permissible decrease exceeded the zero value, and the increase should not be higher than 6,30 t of DM/ha. According to the wide ranges that were achieved with the analyses, it can be stated that the model is not sensitive to great changes in the coefficients of the objective function or in the constraints. The mathematical technique of linear programming turned out to be a useful tool for the prospective analysis of the feeding basis in a tropical dairy farm, with an optimization approach.