Stable parametric programming (Q5945488)

The book discusses the notions optimality and stability and their relationship in linear and convex parametric programming. The first two chapters are preliminary. In Chapters 3 to 6 optimality in convex programming is studied. Basic results on convex sets, convex functions and optimality conditions are given. A particular class of convex functions with ''locally flat surfaces'' (LFS functions) is introduced and new results in convex programming using LFS functions for single objective, multiobjective, differentiable and non-smooth programs are given. Chapters 7 to 10 are devoted to the study of stable perturbations in parametric programming. The behaviour of the feasible set, the set of optimal solutions, the Lagrange multipliers and various index sets under perturbations of the parameter is studied using basic tools of point-to-set topology. The last part of the book (Chapters 11 to 15) consists of several topics. Using results obtained for convex parametric models partly convex programs are studied und some classical numerical methods are formulated for these problems. In order to illustrate typical situations using the introduced results a class of navigation problem of Zermelo and an analysis of efficiency testing in data envelopment are discussed. Finally some of the given general results are applied to particular areas of parametric optimization. Semi-abstract and abstract formulations of parametric programming are introduced and several sample extensions from the finite-dimensional case are proposed. The book would be of great interest to both graduate students and researchers who are concerned with optimization problems.