Özet:

The rigorous formal algorithm for formulating a dual problem for different forms (general, basic, standard, and canonical) of a primal linear programming problem is proposed. First, definitions of a pair of dual problems for standard form of primal linear programming are given. This approach is based on the fact that such a pair was noted first, since it had substantial interpretation. The economic interpretation of the standard problem is profit maximization in the production and sale of some types of products. Such an approach substantially indicates the existence of the primal problem (I) and the strictly corresponding dual (conjugate) (II). The problem of cost minimization is accompanying to the primal problem. The basic concept of the duality theory in linear programming problems is the fact that a pair of problems are mutually conjugate — obtaining dual of dual leads to a primal problem. The rigorous approach to obtaining an algorithm for formulating a dual problem is based on the statement that the dual problem of dual is a primal (original) problem. This approach is used in the paper. For different pairs of dual problems, this statement is rigorously proved. The existing schemes of primal to dual conversion are substantial. Given this, the algorithm of the general approach to formulating pairs of conjugate problems is proposed and rigorously proved. Formalization of the developed scheme makes it easy to get pairs of known dual problems. This allowed for the first time to propose and validate the algorithm for constructing a dual problem for an arbitrary form of the primal problem. Author Biographies Lyudmila Chernova, Admiral Makarov National University of Shipbuilding Heroiv Ukrainy ave., 9, Mykolayiv, Ukraine, 54025 PhD, Associate Professor Department of Information Control Systems and Technologies

Anahtar Kelimeler: