A new interior-point approach for large separable convex quadratic two-stage stochastic problems
Visualitza/Obre
10.1080/10556788.2020.1841190
Inclou dades d'ús des de 2022
Cita com:
hdl:2117/345086
Tipus de documentArticle
Data publicació2020-11-03
Condicions d'accésAccés obert
Llevat que s'hi indiqui el contrari, els
continguts d'aquesta obra estan subjectes a la llicència de Creative Commons
:
Reconeixement-NoComercial-SenseObraDerivada 3.0 Espanya
Abstract
Two-stage stochastic models give rise to very large optimization problems. Several approaches havebeen devised for efficiently solving them, including interior-point methods (IPMs). However, usingIPMs, the linking columns associated to first-stage decisions cause excessive fill-in for the solutionof the normal equations. This downside is usually alleviated if variable splitting is applied to first-stage variables. This work presents a specialized IPM that applies variable splitting and exploits thestructure of the deterministic equivalent of the stochastic problem. The specialized IPM combinesCholesky factorizations and preconditioned conjugate gradients for solving the normal equations.This specialized IPM outperforms other approaches when the number of first-stage variables is largeenough. This paper provides computational results for two stochastic problems: (1) a supply chainsystem and (2) capacity expansion in an electric system. Both linear and convex quadratic formu-lations were used, obtaining instances of up to 38 million variables and six million constraints. Thecomputational results show that our procedure is more efficient than alternative state-of-the-art IPMimplementations (e.g., CPLEX) and other specialized solvers for stochastic optimization.
CitacióCastro, J.; Lama, P. A new interior-point approach for large separable convex quadratic two-stage stochastic problems. "Optimization methods software", 3 Novembre 2020, p. 1-29.
ISSN1055-6788
Versió de l'editorhttps://www.tandfonline.com/doi/abs/10.1080/10556788.2020.1841190?journalCode=goms20
Altres identificadorshttp://www-eio.upc.edu/~jcastro/publications/reports/dr2020-01.pdf
Fitxers | Descripció | Mida | Format | Visualitza |
---|---|---|---|---|
dr2020-01.pdf | 1,917Mb | Visualitza/Obre |