The paper presents an automatic method for solving discrete programming problems, where some or all variables can only take discrete values. The method is based on a numerical algorithm that extends the techniques used in ordinary linear programming. It involves systematic parallel shifts in the functional hyperplane to find a solution with integral coordinates. The algorithm is illustrated through two numerical examples and is designed to be compatible with automatic computing. The authors also discuss the limitations and potential improvements of the method, suggesting that it may be particularly useful for testing the validity of ad hoc methods for new problems.The paper presents an automatic method for solving discrete programming problems, where some or all variables can only take discrete values. The method is based on a numerical algorithm that extends the techniques used in ordinary linear programming. It involves systematic parallel shifts in the functional hyperplane to find a solution with integral coordinates. The algorithm is illustrated through two numerical examples and is designed to be compatible with automatic computing. The authors also discuss the limitations and potential improvements of the method, suggesting that it may be particularly useful for testing the validity of ad hoc methods for new problems.