IJESRT INTERNATIONAL JOURNA A study of Livestock Ration Formulation using Goal programming and Random Search Technique for Global

The aim of this paper is to present a non formulation of a daily dairy cow ration. The basis of its construction lies in the models developed by earlier researchers on the Excel Solver platform. It merg goal programming model with linear objective of minimizing the deviations. The single objective linear model helps in making an estimation of least-cost magnitude that might be expected. The programming model to set the targets for each of the goals that should be achieved as closely as possible. In this study, the construction of the non linear model of Goal programming approach is done by formulation of n objective function as square root of the sum of the squares of the deviations in which weight is assigned to each goal according to its priority. This new model was tested at three values of preferential weights for dairy cows with a 10L daily milk yield, using a controlled Random Search Technique for Global optimization. The result obtained depicts the benefit of applied approach & probabilistic solution technique. In contrast to the linear models, which gives only one solution, this model & solution technique provides many possible solution sets to reduce the cost of the diet without compromising its nutritional quality, by allowing for harmless deviations from the goals, using under & over achievements.