Titlebook: Operations Research Proceedings 1994; Selected Papers of t Ulrich Derigs (Lehrstuhl),Achim Bachem,Andreas Dre Conference proceedings 1995 S

清洗

GIST

On the Degree of Decentralization in Large-Scale Linear ProgrammingLarge-scale linear programs with primal blockangular structure are considered. The numerical effort for solving primal-blockangular LPs with fixed problem size depends heavily on how scarce available common resources are. Having this fact in mind we investigate the benefits of . for the common resources and present preliminary results.

Factorable

后退

On Convergence of Proximal Methods to Fixed Points of Extremal MappingsThe problem of finding the equilibrium points of an extremal mapping is considered. The methods controlled by means of feedback are suggested to compute the fixed points of an extremal mapping. Their convergence to the solution of the equilibrium problem is proved.

VAN

A Reduced Directions Method for Solving Stochastic Problems in the Dialog Optimization System ODISFor solving nonlinear stochastic programming problems a reduced generalized gradient method is suggested which is implemented as a special part in the DIALOG OPTIMIZATION SYSTEM ODIS.

榨取

脱水

On Finiteness of Modified Beale’s Algorithm for Solving Nonconvex Quadratic ProgramThe paper deals with Beale’s method, which is generalized for calculating a local minimum of any quadratic objective function subject to linear constraints. A theoretical background of the algorithm is given briefly and the finiteness of the procedure is proved. Solving of degenerate problems is mentioned.

星球的光亮度

Peculate

A Descent Level Bundle Method for Convex Nondifferentiable Optimizationst to the original level methods of Lemarechal, Nemirovskii and Nesterov, our method is globally convergent without any compactness assumptions and requires bounded storage. It does not employ potentially expensive linesearches as does the level method of Brännlund. Preliminary numerical experience is reported.

传染