On parallelization of a stochastic dynamic programming algorithm for solving large-scale mixed 0- problems under uncertainty

详细信息    查看全文

• 作者：Unai Aldasoro ; Laureano F. Escudero ; María Merino ; Juan F. Monge ; Gloria Pérez
• 关键词：Stochastic dynamic programming ; Inner and outer parallelization ; Multistage stochastic mixed 0- optimization ; Parallel computing ; Message ; passing interface ; 90C15 ; 90C39 ; 90C06 ; 90C11 ; 68W10 ; 68Y05
• 刊名：TOP
• 出版年：2015
• 出版时间：October 2015
• 年：2015
• 卷：23
• 期：3
• 页码：703-742
• 全文大小：1,269 KB
• 参考文献：Al-Khamisl T, M’Hallah R (2011) A two-stage stochastic programming model for the parallel machine scheduling problem with machine capacity. Comput Oper Res 38:1747-759CrossRef
  Aldasoro U, Escudero LF, Merino M, Pérez G (2013) An algorithmic framework for solving large-scale multistage stochastic mixed 0- problems with nonsymmetric scenario trees. Part II: Parallelization. Comput Oper Res 40:2950-960CrossRef
  ARINA (2015) Cluster IZO-SGI, SGIker (UPV/EHU). http://?www.?ehu.?es/?sgi/?recursos/?cluster-arina
  Benders J (1962) Partitioning procedures for solving mixed variables programming problems. Manag Sci 1:238-52
  Beraldi P, Grandinetti L, Musmanno R, Trik C (2000) Parallel algorithms to solve two-stage stochastic linear programs with robustness constraints. Parallel Comput 26:1889-908CrossRef
  Birge JR (1985) Decomposition and partitioning methods for multistage stochastic linear programs. Oper Res 33:989-007CrossRef
  Birge JR (1997) Stochastic programming computation and applications. INFORMS J Comput 9:111-33CrossRef
  Birge JR, Donohue CJ, Holmes DF, Svintsitski O (1996) A parallel implementation of the nested decomposition algorithm for multistage stochastic linear programs. Math Program 75:327-52
  Birge JR, Louveaux FV (2011) Introduction to stochastic programming, 2nd edn. Springer, Berlin
  Blomval J, Lindberg P (2002) A Riccati-based primal interior point solver for multistage stochastic programming. Eur J Oper Res 143:452-61CrossRef
  Conejo AJ, Castillo E, Mínguez R, García-Bertrand R (2006) Decomposition techniques in mathematical programming. Engineering and science applications. Springer, Berlin
  Cristobal MP, Escudero LF, Monge JF (2009) On stochastic dynamic programming for solving large-scale production planning problems under uncertainty. Comput Oper Res 36:2418-428CrossRef
  Culler DE, Gupta A, Singh JP (1997) Parallel computer architecture: a hardware/software approach, 1st edn. Morgan Kaufmann Publishers Inc., San Francisco
  Dempster MAH, Thompson RT (1998) Parallelization and aggregation of nested Benders decomposition. Ann Oper Res 81:163-88CrossRef
  Dias B, Tomin M, Mercato A, Ramos T, Brandi R da, Silva jr IC, Filho JAP (2013) Parallel computing applied to stochastic dynamic programming for long term operation planning of hydrothermal power systems. Eur J Oper Res 229:212-22
  Escudero LF, de la Fuente J, García C, Prieto F (1999) A parallel computation approach for solving multistage stochastic network problems. Ann Oper Res 90:1-1CrossRef
  Escudero LF, Garín MA, Merino M, Pérez G (2010) On an exact algorithm for solving large-scale two-stage stochastic mixed-integer problems: theoretical and computational aspects. Eur J Oper Res 204:105-16CrossRef
  Escudero LF, Garín MA, Merino M, Pérez G (2012) An algorithmic framework for solving large-scale multistage stochastic mixed 0- problems with nonsymmetric scenario trees. Comput Oper Res 39:1133-144CrossRef
  Escudero LF, Garín MA, Merino M, Pérez G (2015) On time stochastic dominance induced by mixed integer-linear recourse in multistage stochastic programs (submitted)
  Escudero LF, Garín MA, Pérez G, Unzueta A (2013) Scenario cluster decomposition of the lagrangian dual in stochastic mixed 0- optimization. Comput Oper Res 40:362-77
  Escudero LF, Monge JF, Morales DR (2015) An sdp approach for multiperiod mixed 0- linear programming models with stochastic dominance constraints for risk management. Comput Oper Res 58:32-0
  Escudero LF, Monge JF, Morales DR, Wang J (2013) Expected future value decomposition based bid price generation for large-scale network revenue management. Transp Sci 47:181-97CrossRef
  Hennessy JL, Patterson DA (2003) Computer architecture: a quantitative approach. Morgan Kaufmann Publishers Inc., San Francisco
  IBM (2015) ILOG CPLEX optimizer. http://?www-01.?ibm.?com/?software/?integration/?optimization/?cplex-optimizer/-/span>
  Latorre JM, Cerisola S, Ramos A, Palacios R (2009) Analysis of stochastic problem decomposition algorithms in computational grids. Ann Oper Res 166:355-79CrossRef
  Li X, Wei J, Li T, Wang G, Yeh WG (2014) A parallel dynamic programming algorithm for multi-reservoir system optimization. Adv Water Resour 67:1-5CrossRef
  Linderoth J, Shapiro A, Wright S (2006) The empirical behavior of sampling methods for stochastic programming. Ann Oper Res 142:215-41CrossRef
  Linderoth JT, Wright S (2008) Decomposition algorithms for stochastic programming on a computational grid. Tech. rep, Mathematics and Computer Science Division, Argonne National Laboratory, Chicago
  Lumbreras S (2014) Decision support methods for large-scale flexible transmission expansion planning. Ph.D. thesis, Institute of Investigación Tecnológica. Universidad Pontificia de Comillas, Madrid
  Mahlke D (2011) A scenario tree-based decomposition for solving multistage stochastic programs with application in energyproduction. Springer, Berli
• 作者单位：Unai Aldasoro (1)
  Laureano F. Escudero (2)
  María Merino (3)
  Juan F. Monge (4)
  Gloria Pérez (3)
  
  1. Dpto. de Matemática Aplicada, Universidad del País Vasco, Leioa, Bizkaia, Spain
  2. Dpto. de Estadística e Investigación Operativa, Universidad Rey Juan Carlos, Madrid, Spain
  3. Dpto. de Matemática Aplicada, Estadística e Investigación Operativa, Universidad del País Vasco, Barrio Sarriena s/n, 48940, Leioa, Bizkaia, Spain
  4. Centro de Investigación Operativa, Universidad Miguel Hernández, Elche, Spain
  
• 刊物主题：Operations Research/Decision Theory; Optimization; Statistics for Business/Economics/Mathematical Finance/Insurance; Industrial and Production Engineering; Game Theory/Mathematical Methods;
• 出版者：Springer Berlin Heidelberg
• ISSN：1863-8279

文摘

A parallel computing implementation of a Serial Stochastic Dynamic Programming approach referred to as the S-SDP algorithm is introduced to solve large-scale multiperiod mixed 0- optimization problems under uncertainty. The paper presents Inner and Outer Parallelization versions of the S-SDP algorithm, referred to as Inner P-SDP and Outer P-SDP, respectively, so that the problem solving elapsed time and gap reduction is analyzed. The basic idea of Inner P-SDP consists of parallelizing the optimization of variations of the MIP subproblems attached to the sets of scenario clusters created by the modeler-defined stages to decompose the original problem. The Outer P-SDP performs simultaneous interconnected executions of the serial algorithm, so that a wider feasibility area is explored using iterative communication to redefine search directions. Strategies are presented to analyze the performance of parallel computation based on Message-Passing Interface threads to solve stage-related subproblems versus the serial version of SDP methodology. The results of using the parallelization are remarkable, as not only faster but also better solutions than the serial version are obtained. In particular, we report up to 10 times speedup for 12 threads on the Inner P-SDP algorithm. The new approach allows problems to be solved using less computing time than a state-of-the-art MIP solver. It can thus solve very large-scale problems that could not otherwise be achieved by plain use of the solver or by the S-SDP algorithm in acceptable elapsed time, if any. Keywords Stochastic dynamic programming Inner and outer parallelization Multistage stochastic mixed 0- optimization Parallel computing Message-passing interface