Duality theory in interval-valued linear programming problems (Q639940)

Similar to the traditional concepts of linear programming (LP) problems, this paper discusses with the strong and weak duality theorems, duality gaps and solvability for interval-valued linear programming (I-VLP) problems. The proposed problems in this paper are closely related to the inexact LP problems; but the proposed solution concept follows the similar concept of non-dominated solution in multi objective programming problems by defining a partial ordering on the set of all closed intervals. After introducing some solution concepts for set optimization problems in Section 2 (Preliminaries), some necessary notations and operations on intervals are illustrated in the next section. Then the partial orders are defined and based on them, three pairs of primal and dual problems are introduced. In the sequel, the weak and strong duality theorems for each pair of the primal and dual problems and the conditions for solvability of each pairs are presented in three theorems and also the complementary slackness for the pairs ist obtained. By defining the concept of duality gap, the solvability and strong duality theorem of the for the three pairs of problems was investigated again. Based on the primal and dual of I-VLPP, in the last part of the paper (which like Section 4 shows us the similar fuzzy related concepts), the primal and dual lower- and upper-level LP problems are defined as traditional LP problems. Then, the relation between the I-VLPP solution and the set of all finite optimal solution of this new problem is examined.