On the stability of the feasible set in linear optimization (Q5945277)

The paper deals with the stability of the feasible set of two classes of linear optimization problems, each family containing the dual problems corresponding to the members of the other family. The authors characterize the problems of these families that are stable in terms of the lower semicontinuity property of the feasible set mapping and the boundedness of the optimal set of the corresponding coupled problem. The results of this paper extend some well-known theorems of \textit{A. C. Williams} [J. Soc. Ind. Appl. Math. 11, 82--94 (1963; Zbl 0115.38102)] and \textit{S. M. Robinson} [Oper. Res. 25, 435--447 (1977; Zbl 0373.90045)] on the stability of ordinary linear programming problems to linear optimization problems with infinitely many variable or constraints.