For example,Бобцов

METHODS FOR SOLVING LINEAR PROGRAMMING PROBLEMS WITH ADDITIONAL RESTRICTIONS TO THE PARTICULAR VARIABLES

Annotation

The paper describes the solution of the problem related to the specific admissible sets of variables in linear programming. We are discussing the feasible set which is the union of segments with multiplier parameter for some variable. The solution of this problem is performed in two stages: at the beginning the relaxed problem of linear optimization is solved (without additional restrictions to the variables), and then auxiliary nonlinear optimization problem is constructed on the basis of the obtained solution. Solution of the mentioned auxiliary problem is based on a specialized method of nonlinear optimization - Box method. The result is the algorithm proposed by the author for solving linear programming problems with additional restrictions to the variables with indication of the accuracy estimates. The solution of this problem has a high practical importance. Such restrictions to the variables in the linear programming problems occur often enough for production problems. Method application is shown on the example of an optimal plan finding for pattern cutting in the paper industry, when the task arises associated with the rounding of reels number for paper machines in terms of the found optimal paper cutting plan.

Keywords

Articles in current issue