About the algorithm of dual cuttings for a two-stage stochastic programming problem
From the introduction: Optimization problems of stochastic programming [\textit{D. B. Yudin}, ``Problems and methods of stochastic programming, Moskva: Izdatel'stvo ``Sovetskoe Radio (1979; Zbl 0486.90058)] occur in the mathematical modeling of many economic and technical systems. The reason is that either the exact parameters of these models are not known in advance (this is more typical for economic systems), or some product is designed for functioning in accidental and unpredictable conditions (this is more typical for technical devices). Among various problem definitions, prevailing is the linear two-stage model of perspective solution and the further correction. It implies the minimization of the average loss, taking into account the subsequent optimal modification [\textit{J. R. Birge}, Oper. Res. 33, 989--1007 (1985; Zbl 0581.90065); \textit{E. Fragnière, J. Gondzio} and \textit{J.-P. Vial}, Ann. Oper. Res. 99, 167--187 (2000; Zbl 0990.90083)]. These problems have specific structural singularities in constraints matrices which heed the non-anticipative character of corrections and the independence of alternative scenarios. These singularities enable one to decompose the structured problems on relatively weakly connected blocks of a less dimension and to use parallel algorithms. As the main model of a structured optimization problem, consider the two-block linear programming problem with binding variables [\textit{L. S. Lasdon}, Optimization theory for large systems, Mir, Moscow (1975); \textit{V. I. Tsurkov}, Decomposition in large-scale problems, Moscow: Nauka (1981)], for its solution one uses the forward-dual decomposition algorithm based on the dual cuttings [\textit{A. S. Velichko} and \textit{E. A. Nurminskii}, Decomposition of finite elmeents method using the theory of structured optimization problems, Electronic journal ``Research in Russia, pp. 1237--1256 (2002), \url{http://zhurnal.ape.relarn.ru/articles/2002/113.pdf}; \textit{E. A. Nurminskii}, Numerical methods of convex optimization, Moscow: Nauka (1991; Zbl 0734.90069)].
- Stochastic Decomposition: An Algorithm for Two-Stage Linear Programs with Recourse
- A multicut algorithm for two-stage stochastic linear programs
- Dual decomposition in stochastic integer programming
- scientific article; zbMATH DE number 740448
- A regularized stochastic decomposition algorithm for two-stage stochastic linear programs
This page was built for publication: About the algorithm of dual cuttings for a two-stage stochastic programming problem
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1022201)
