AbstractThe use of linear programming to select diets to meet specific nutritional requirements frequently results in an over-supply of certain nutrients. Nutritional balance is difficult to achieve in diets selected by linear programming owing to the complex inter-relationships of the constraints. Goal programming is presented as a method of achieving nutritional balance in selected diets. An example demonstrating the goal programming approach is followed by a report of an application of the technique to the selection from 150 food raw materials to satisfy the daily nutritional requirements of Thais. The nutritional balance of the raw materials selected by goal programming showed a marked improvement over that selected by linear programming.