Journal Home Page

Cumulative Index

List of all Volumes

Complete Contents
of this Volume

Previous Article

Next Article

Journal of Convex Analysis 09 (2002), No. 2, 665--691
Copyright Heldermann Verlag 2002

On the Topology of Generalized Semi-Infinite Optimization

Gerhard-Wilhelm Weber
Fakultät für Mathematik, Technische Universität, Reichenhainer Strasse, 09126 Chemnitz, Germany

This survey article reflects the topological and inverse behaviour of generalized semi-infinite optimization problems P(f, h, g, u, v) and presents the analytical methods. These differentiable problems admit an infinite set Y(x) of inequality constraints y which depends on the state x. We extend our previous investigations ["Generalized Semi-Infinite Optimization and Related Topics", Research and Expositions in Mathematics 29, Heldermann Verlag (2003)] based on research of Guddat, Jongen, Rueckmann, Twilt and others. Under suitable assumptions on boundedness and qualifying conditions on lower y-stage and upper x-stage, we present manifold, continuity and global stability properties of the feasible set M[h, g, u, v], and corresponding structural stability properties of P(f, h, g, u, v) referring to slight data perturbations. Hereby, the character of our investigation is essentially specialized by the linear independence constraint qualification locally imposed on Y(x). The achieved results are important for algorithm design and convergence. Two extensions refer to unboundedness and nondifferentiable max-min-type objective functions.
In the course of explanation, the perturbational approach gives rise to study inverse problems of reconstruction. We trace them into optimal control of ordinary differential equations, and indicate related investigations in heating processes, continuum mechanics and discrete tomography. Throughout the article, we realize discrete-combinatorial aspects and methods.

Keywords: Generalized semi-infinite optimization, constraint qualification, structural stability, inverse problem, reconstruction, nondifferentiability, optimal control, directed graph, discrete tomography.

MSC: 05C20 49J15 90C34

[ Fulltext-pdf (1660 KB)]