Analysis of structure in fuzzy linear programs (Q1814624): Difference between revisions

This paper deals with the study of some structural problems of fuzzy linear programming problems. The general model considered consists of crisp relations with fuzzy number, technical coefficients, right sides and objective function coefficients. Because for each \(\alpha\in [0,1]\) the \(\alpha\)-cut of a fuzzy number is a closed bounded interval, the representation theorem of fuzzy sets allows one to change this fuzzy problem into an equivalent indexed family of linear programming problems. Through using interval arithmetic on each \(\alpha\)-level, the author achieves a useful generalization and extension of the class of linear programming problems. Both crisp and fuzzy problems are within the same theoretical and computational framework. On this basis, in the three main sections of the paper (2,3, and 4), the following problems are dealt with: (a) the uncovering of structure of any fuzzy linear systems; (b) identifying rows that are redundant, infeasible or whose variables have tighter bounds; (c) replacing rows by simple bounds; (d) removing or tightening bounds; (e) duality for interval linear programs; (f) weak dualities; (g) strong dualities; (h) test to uncover the structure of the constraint matrix in linear programming problems; (i) redundancies, infeasibilities and bound structures for dual constraints; (j) forcing variables at their bounds.