Cloud Branching. Timo Berthold. joint work with Domenico Salvagnin (Università degli Studi di Padova)


 Hester Gillian Lynch
 1 years ago
 Views:
Transcription
1 Cloud Branching Timo Berthold Zuse Institute Berlin joint work with Domenico Salvagnin (Università degli Studi di Padova) DFG Research Center MATHEON Mathematics for key technologies 21/May/13, CPAIOR 2013, IBM T. J. Watson Research Center
2 MIP & Branching min s.t. c T x Ax b x Z I 0 R C 0 Mixed Integer Program: linear objective & constraints integer variables continuous variables Branching for MIP: based on LP relaxation fractional variables tries to improve dual bound Timo Berthold: Cloud Branching 2 / 17
3 MIP & Branching min s.t. c T x Ax b x Z I 0 R C 0 Mixed Integer Program: linear objective & constraints integer variables continuous variables Branching for MIP: based on LP relaxation fractional variables tries to improve dual bound x i xi x i xi for i I and xi / Z Timo Berthold: Cloud Branching 2 / 17
4 Branching rules in MIP Most infeasible branching often referred to as a simple, standard rule computationally as bad as random branching! Strong branching [ApplegateEtAl1995] solve LP relaxations for some candidates, choose best effective w.r.t. number of nodes, expensive w.r.t. time Pseudocost branching [BenichouEtAl1971] try to estimate LP values, based on history information effective, cheap, but weak in the beginning can be combined with strong branching Timo Berthold: Cloud Branching 3 / 17
5 Branching rules in MIP Most infeasible branching often referred to as a simple, standard rule computationally as bad as random branching! Strong branching [ApplegateEtAl1995] solve LP relaxations for some candidates, choose best effective w.r.t. number of nodes, expensive w.r.t. time Pseudocost branching [BenichouEtAl1971] try to estimate LP values, based on history information effective, cheap, but weak in the beginning can be combined with strong branching Timo Berthold: Cloud Branching 3 / 17
6 Degeneracy naïvely: 1 optimal solution, determined by n constraints Timo Berthold: Cloud Branching 4 / 17
7 Degeneracy naïvely: 1 optimal solution, determined by n constraints at a second thought: 1 optimal solution, k > n tight constraints Timo Berthold: Cloud Branching 4 / 17
8 Degeneracy naïvely: 1 optimal solution, determined by n constraints at a second thought: 1 optimal solution, k > n tight constraints but really: an optimal polyhedron Timo Berthold: Cloud Branching 4 / 17
9 Degeneracy naïvely: 1 optimal solution, determined by n constraints at a second thought: 1 optimal solution, k > n tight constraints but really: an optimal polyhedron Timo Berthold: Cloud Branching 4 / 17
10 Degeneracy naïvely: 1 optimal solution, determined by n constraints at a second thought: 1 optimal solution, k > n tight constraints but really: an optimal polyhedron Goal of this talk: branch on a set (a cloud) of solutions Timo Berthold: Cloud Branching 4 / 17
11 Two questions Goal of this talk: branch on a cloud of solutions 1. How do we get extra optimal solutions? 2. Why should that be a good idea anyway? Timo Berthold: Cloud Branching 5 / 17
12 Generating cloud points 1. How do we get extra optimal solutions? restrict LP to optimal face min/max each variable (OBBT) or feasibility pump objective (pump&reduce [Achterberg2010]) Timo Berthold: Cloud Branching 6 / 17
13 Generating cloud points 1. How do we get extra optimal solutions? restrict LP to optimal face min/max each variable (OBBT) or feasibility pump objective (pump&reduce [Achterberg2010]) (x, x) = x j x j x 1 = 0.4 x 2 = 0.55 Timo Berthold: Cloud Branching 6 / 17
14 Generating cloud points 1. How do we get extra optimal solutions? restrict LP to optimal face min/max each variable (OBBT) or feasibility pump objective (pump&reduce [Achterberg2010]) (x, x) = x j x j = x 1 + x 2 x 1 = 0.4 x 2 = 0.55 Timo Berthold: Cloud Branching 6 / 17
15 Generating cloud points 1. How do we get extra optimal solutions? restrict LP to optimal face min/max each variable (OBBT) or feasibility pump objective (pump&reduce [Achterberg2010]) (x, x) = x j x j = x 1 + x 2 x 1 [0.4, 0.55] x 2 [0.55, 0.9] Timo Berthold: Cloud Branching 6 / 17
16 Generating cloud points 1. How do we get extra optimal solutions? restrict LP to optimal face min/max each variable (OBBT) or feasibility pump objective (pump&reduce [Achterberg2010]) (x, x) = x j x j = +x 1 + x 2 x 1 [0.4, 0.55] x 2 [0.55, 0.9] Timo Berthold: Cloud Branching 6 / 17
17 Generating cloud points 1. How do we get extra optimal solutions? restrict LP to optimal face min/max each variable (OBBT) or feasibility pump objective (pump&reduce [Achterberg2010]) (x, x) = x j x j = +x 1 + x 2 x 1 [0.4, 0.9] x 2 [0.55, 1.0] Timo Berthold: Cloud Branching 6 / 17
18 Generating cloud points 1. How do we get extra optimal solutions? restrict LP to optimal face min/max each variable (OBBT) or feasibility pump objective (pump&reduce [Achterberg2010]) (x, x) = x j x j = +x 1 + x 2 x 1 [0.4, 0.9] x 2 [0.55, 1.0] intervals instead of values Timo Berthold: Cloud Branching 6 / 17
19 Generating cloud points 1. How do we get extra optimal solutions? restrict LP to optimal face min/max each variable (OBBT) or feasibility pump objective (pump&reduce [Achterberg2010]) (x, x) = x j x j = +x 1 + x 2 x 1 [0.4, 0.9] x 2 [0.55, 1.0] intervals instead of values stalling is cheap! Timo Berthold: Cloud Branching 6 / 17
20 Exploiting cloud points (pseudocost) 2. Why should that be a good idea anyway? z = c T x z = c T x j + j x j x j x j x j x j x j pseudocost update ς + j = and ς xj x j = j xj x j pseudocostbased estimation + j = Ψ + j ( x j x j ) and j = Ψ j (x j x j ) Timo Berthold: Cloud Branching 7 / 17
21 Exploiting cloud points (pseudocost) 2. Why should that be a good idea anyway? z = c T x z = c T x j + j xj l x u j xj j j x j x j x j pseudocost update ς + j =... better: ς + xj x j = j xj u j pseudocostbased estimation + j = Ψ + j ( x j x j ) and j = Ψ j (x j x j ) Timo Berthold: Cloud Branching 7 / 17
22 Exploiting cloud points (pseudocost) 2. Why should that be a good idea anyway? z = c T x z = c T x j j + j + j xj l x u j xj j j xj l x u j xj j j pseudocost update ς + j =... better: ς + xj x j = j xj u j pseudocostbased estimation + j = Ψ + j ( x j x j )... better: + j = Ψ + j ( x j u j) Timo Berthold: Cloud Branching 7 / 17
23 Observation Lemma Let x be an optimal solution of the LP relaxation at a given branchandbound node and xj l j xj u j xj. Then 1. for fixed and, it holds that ς + j ς + j and ς j ς j, respectively; 2. for fixed Ψ + j and Ψ j, it holds that + j + j and j j, respectively. Timo Berthold: Cloud Branching 8 / 17
24 Exploiting cloud points (strong branch) 2. Why should that be a good idea anyway? Full strong branching: solves 2 #frac. var s many LPs uses product of improvement values as branching score improvement on both sides better than on one Timo Berthold: Cloud Branching 9 / 17
25 Exploiting cloud points (strong branch) 2. Why should that be a good idea anyway? Full strong branching: solves 2 #frac. var s many LPs uses product of improvement values as branching score improvement on both sides better than on one Benefit of cloud intervals: frac. var. gets integral in cloud point one LP spared cloud branching acts as a filter new frac. var. s new candidates (one side known) Timo Berthold: Cloud Branching 9 / 17
26 Cloud strong branching algorithm Idea: Use 3partition F 2, F 1, F 0 of branching candidates if strict improvements in both directions for F 2, disregard F 1 F 0 if strict improvement in one direction for F 2 F 1, disregard F 0 Timo Berthold: Cloud Branching 10 / 17
27 Cloud strong branching algorithm Idea: Use 3partition F 2, F 1, F 0 of branching candidates if strict improvements in both directions for F 2, disregard F 1 F 0 if strict improvement in one direction for F 2 F 1, disregard F 0 alternatively: Only use F 2 Timo Berthold: Cloud Branching 10 / 17
28 Cloud strong branching algorithm Idea: Use 3partition F 2, F 1, F 0 of branching candidates if strict improvements in both directions for F 2, disregard F 1 F 0 if strict improvement in one direction for F 2 F 1, disregard F 0 alternatively: Only use F 2 stop pump&reduce procedure when new cloud point does not imply new integral bound Timo Berthold: Cloud Branching 10 / 17
29 Cloud strong branching algorithm Idea: Use 3partition F 2, F 1, F 0 of branching candidates if strict improvements in both directions for F 2, disregard F 1 F 0 if strict improvement in one direction for F 2 F 1, disregard F 0 alternatively: Only use F 2 stop pump&reduce procedure when new cloud point does not imply new integral bound Note: In our experiments, we do not use cloud points for anything else (heuristics, cuts) Timo Berthold: Cloud Branching 10 / 17
30 Implementation into SCIP SCIP: Solving Constraint Integer Programs standalone solver / branchcutandpriceframework modular structure via plugins free for academic use: very fast noncommercial MIP and MINLP solver time in seconds x 3.94x noncommercial commercial GLPK 4.47 lpsolve CBC SCIP CLP x SCIP SoPlex SCIP Cplex x x 0.77x Xpress x Cplex x 0.17x 0 Gurobi solved (of 87 instances) results by H. Mittelmann (10/Jan/2013) Timo Berthold: Cloud Branching 11 / 17
31 SCIP 3.0 SCIP: Solving Constraint Integer Programs better support of MINLP new presolvers and propagators AMPL and MATLAB interface (beta) first releases of GCG and UG time in seconds solved (of 87) 1.6x 5.2x 4.0x 3.0x 3.0x 2.5x 2.0x 1.3x 1.1x 1.0x sip 1.2 SoPlex 1.2.1a SCIP 0.7 SoPlex SCIP 0.80 SoPlex SCIP 0.90 SoPlex SCIP 1.00 SoPlex SCIP 1.1 SoPlex SCIP 1.2 SoPlex SCIP 2.0 SoPlex SCIP 2.1 SoPlex SCIP 3.0 SoPlex Timo Berthold: Cloud Branching 12 / 17
32 Test sets MMM MIPLIB 3.0, MIPLIB 2003, MIPLIB2010 industry and academics 168 instances from diverse applications huge collection of 350 instances many combinatorial ones mainly collected from NEOS server Timo Berthold: Cloud Branching 13 / 17
33 Computational results cloud statistics Test set %Succ Pts LPs %Sav MMM applicable on some MMM, but many instances only few cloud points on average significants amount of LPs saved (if affected) Timo Berthold: Cloud Branching 14 / 17
34 Computational results strong branch cloud branch Test set Nodes Time (s) Nodes Time (s) MMM little less nodes 5.5% faster on MMM (few affected instances) 30 % faster on Timo Berthold: Cloud Branching 15 / 17
35 Wrap up Conclusion: Cloud branching... exploits knowledge of alternative relaxation optima can help to improve pseudocost predictions makes full strong branching up to 30% faster Outlook pseudocost, reliability, hybrid branching cloud points from alternative relaxations (MINLP!) nonchimerical + cloud + propagation =? Timo Berthold: Cloud Branching 16 / 17
36 Cloud Branching Timo Berthold Zuse Institute Berlin joint work with Domenico Salvagnin (Università degli Studi di Padova) DFG Research Center MATHEON Mathematics for key technologies 21/May/13, CPAIOR 2013, IBM T. J. Watson Research Center
Measuring the impact of primal heuristics
KonradZuseZentrum für Informationstechnik Berlin Takustraße 7 D14195 BerlinDahlem Germany TIMO BERTHOLD Measuring the impact of primal heuristics Supported by the DFG Research Center MATHEON Mathematics
More informationUsing diversification, communication and parallelism to solve mixedinteger linear programs
Using diversification, communication and parallelism to solve mixedinteger linear programs R. Carvajal a,, S. Ahmed a, G. Nemhauser a, K. Furman b, V. Goel c, Y. Shao c a Industrial and Systems Engineering,
More informationComputing with the Mixed Integer Rounding Cut.
Computing with the Mixed Integer Rounding Cut. Marcos Goycoolea. Universidad Adolfo Ibañez Santiago, Chile. (joint work with Sanjeeb Dash, IBM Research, NY) A quick history... Gomory s Mixed Integer
More informationSeparation, Propagation, Heuristics
Separation, Propagation, Heuristics Selected tricks from an advanced MINLP solver Ambros Gleixner, Zuse Institute Berlin joint work with T Berthold, B Müller, F Serrano, R Schwarz, and S Weltge MINO /
More informationLarge Neighborhood Search beyond MIP
KonradZuseZentrum für Informationstechnik Berlin Takustraße 7 D14195 BerlinDahlem Germany TIMO BERTHOLD, STEFAN HEINZ, MARC E. PFETSCH 1, STEFAN VIGERSKE 2, Large Neighborhood Search beyond MIP 1 2
More informationReconsidering Mixed Integer Programming and MIPbased Hybrids for Scheduling
KonradZuseZentrum für Informationstechnik Berlin Takustraße 7 D14195 BerlinDahlem Germany STEFAN HEINZ J. CHRISTOPHER BECK Reconsidering Mixed Integer Programming and MIPbased Hybrids for Scheduling
More informationThe Gurobi Optimizer
The Gurobi Optimizer Gurobi History Gurobi Optimization, founded July 2008 Zonghao Gu, Ed Rothberg, Bob Bixby Started code development March 2008 Gurobi Version 1.0 released May 2009 History of rapid,
More informationResource Allocation and Scheduling
Lesson 3: Resource Allocation and Scheduling DEIS, University of Bologna Outline Main Objective: joint resource allocation and scheduling problems In particular, an overview of: Part 1: Introduction and
More informationNoncommercial Software for MixedInteger Linear Programming
Noncommercial Software for MixedInteger Linear Programming J. T. Linderoth T. K. Ralphs December, 2004. Revised: January, 2005. Abstract We present an overview of noncommercial software tools for the
More informationRecent improvements using constraint integer programming for resource allocation and scheduling
Recent improvements using constraint integer programming for resource allocation and scheduling Stefan Heinz 1,, WenYang Ku 2, and J. Christopher Beck 2 1 Zuse Institute Berlin, Takustr. 7, 14195 Berlin,
More information6. Mixed Integer Linear Programming
6. Mixed Integer Linear Programming Javier Larrosa Albert Oliveras Enric RodríguezCarbonell Problem Solving and Constraint Programming (RPAR) Session 6 p.1/40 Mixed Integer Linear Programming A mixed
More informationSBB: A New Solver for Mixed Integer Nonlinear Programming
SBB: A New Solver for Mixed Integer Nonlinear Programming Michael R. Bussieck GAMS Development Corp. Arne S. Drud ARKI Consulting & Development A/S OR2001, Duisburg Overview! SBB = Simple Branch & Bound!
More informationInteger programming solution methods  introduction
Integer programming solution methods  introduction J E Beasley Capital budgeting There are four possible projects, which each run for 3 years and have the following characteristics. Capital requirements
More informationSolving the ILP using branchandcut
Solving the ILP using branchandcut Solving ILPs is a main topic in combinatorial optimization. We will take a brief look at the branchandcut approach. Branchandcut makes use of two techniques: Cutting
More informationSpecial Session on Integrating Constraint Programming and Operations Research ISAIM 2016
Titles Special Session on Integrating Constraint Programming and Operations Research ISAIM 2016 1. GrammarBased Integer Programming Models and Methods for Employee Scheduling Problems 2. Detecting and
More informationGAMS, Condor and the Grid: Solving Hard Optimization Models in Parallel. Michael C. Ferris University of Wisconsin
GAMS, Condor and the Grid: Solving Hard Optimization Models in Parallel Michael C. Ferris University of Wisconsin Parallel Optimization Aid search for global solutions (typically in nonconvex or discrete)
More informationMixed Integer Linear Programming in R
Mixed Integer Linear Programming in R Stefan Theussl Department of Statistics and Mathematics Wirtschaftsuniversität Wien July 1, 2008 Outline Introduction Linear Programming Quadratic Programming Mixed
More informationSolving convex MINLP problems with AIMMS
Solving convex MINLP problems with AIMMS By Marcel Hunting Paragon Decision Technology BV An AIMMS White Paper August, 2012 Abstract This document describes the Quesada and Grossman algorithm that is implemented
More informationGAMS BranchandCut & Heuristic Facility
GAMS BranchandCut & Heuristic Facility Michael R. Bussieck MBussieck@gams.com GAMS Software GmbH GAMS Development Corp INFORMS Annual Meeting Washington DC 12 October 2008 1 Welcome/Agenda BranchandCut
More informationMaintenance optimization in nuclear power plants modelling, analysis, and experimentation. Karin Kvickström
MASTER STHESIS Maintenance optimization in nuclear power plants modelling, analysis, and experimentation Karin Kvickström DepartmentofMathematicalSciences CHALMERS UNIVERSITY OF TECHNOLOGY UNIVERSITY OF
More informationA branchandbound algorithm for convex multiobjective Mixed Integer NonLinear Programming Problems
A branchandbound algorithm for convex multiobjective Mixed Integer NonLinear Programming Problems Valentina Cacchiani 1 Claudia D Ambrosio 2 1 University of Bologna, Italy 2 École Polytechnique, France
More informationBranchandPrice Approach to the Vehicle Routing Problem with Time Windows
TECHNISCHE UNIVERSITEIT EINDHOVEN BranchandPrice Approach to the Vehicle Routing Problem with Time Windows Lloyd A. Fasting May 2014 Supervisors: dr. M. Firat dr.ir. M.A.A. Boon J. van Twist MSc. Contents
More informationScheduling Home Hospice Care with Logicbased Benders Decomposition
Scheduling Home Hospice Care with Logicbased Benders Decomposition Aliza Heching Compassionate Care Hospice John Hooker Carnegie Mellon University CPAIOR 2016 Banff, Canada Home hospice care The Problem
More informationUsing the analytic center in the feasibility pump
Using the analytic center in the feasibility pump Daniel Baena Jordi Castro Dept. of Stat. and Operations Research Universitat Politècnica de Catalunya daniel.baena@upc.edu jordi.castro@upc.edu Research
More informationBackdoors to Combinatorial Optimization: Feasibility and Optimality
Backdoors to Combinatorial Optimization: Feasibility and Optimality Bistra Dilkina, Carla P. Gomes, Yuri Malitsky 2, Ashish Sabharwal, and Meinolf Sellmann 2 Department of Computer Science, Cornell University,
More informationReconnect 04 Solving Integer Programs with Branch and Bound (and Branch and Cut)
Sandia is a ultiprogra laboratory operated by Sandia Corporation, a Lockheed Martin Copany, Reconnect 04 Solving Integer Progras with Branch and Bound (and Branch and Cut) Cynthia Phillips (Sandia National
More information. P. 4.3 Basic feasible solutions and vertices of polyhedra. x 1. x 2
4. Basic feasible solutions and vertices of polyhedra Due to the fundamental theorem of Linear Programming, to solve any LP it suffices to consider the vertices (finitely many) of the polyhedron P of the
More informationColumn Generation in GAMS Extending the GAMS BranchandCutandHeuristic (BCH) Facility
Column Generation in GAMS Extending the GAMS BranchandCutandHeuristic (BCH) Facility Michael R. Bussieck MBussieck@gams.com GAMS Software GmbH GAMS Development Corp 83rd Working Group Meeting Real
More informationCPLEX Tutorial Handout
CPLEX Tutorial Handout What Is ILOG CPLEX? ILOG CPLEX is a tool for solving linear optimization problems, commonly referred to as Linear Programming (LP) problems, of the form: Maximize (or Minimize) c
More informationTOMLAB  For fast and robust largescale optimization in MATLAB
The TOMLAB Optimization Environment is a powerful optimization and modeling package for solving applied optimization problems in MATLAB. TOMLAB provides a wide range of features, tools and services for
More informationChapter 13: Binary and MixedInteger Programming
Chapter 3: Binary and MixedInteger Programming The general branch and bound approach described in the previous chapter can be customized for special situations. This chapter addresses two special situations:
More informationJeanFrançois Puget @JFPuget. Analytics Role. 2015 IBM Corporation
JeanFrançois Puget @JFPuget Analytics Role Big Data = All Data Not just about large volume Volume Variety Velocity Veracity Data at Scale Terabytes to petabytes of data Data in Many Forms RDBMs, objects,
More informationRecovery of primal solutions from dual subgradient methods for mixed binary linear programming; a branchandbound approach
MASTER S THESIS Recovery of primal solutions from dual subgradient methods for mixed binary linear programming; a branchandbound approach PAULINE ALDENVIK MIRJAM SCHIERSCHER Department of Mathematical
More informationSuperpowered Optimization in Python With Gurobi and Anaconda. Renan Garcia, Ph.D., Gurobi Op4miza4on Michael Grant, Ph.D., Con4nuum Analy4cs
Superpowered Optimization in Python With Gurobi and Anaconda Renan Garcia, Ph.D., Gurobi Op4miza4on Michael Grant, Ph.D., Con4nuum Analy4cs Renan Garcia, Ph.D. Optimization Support Engineer at Gurobi Optimization
More informationBranch and Bound Methods
Branch and Bound Methods basic ideas and attributes unconstrained nonconvex optimization mixed convexboolean optimization Prof. S. Boyd, EE364b, Stanford University Methods for nonconvex optimization
More informationDiscrete Optimization
Discrete Optimization [Chen, Batson, Dang: Applied integer Programming] Chapter 3 and 4.14.3 by Johan Högdahl and Victoria Svedberg Seminar 2, 20150331 Todays presentation Chapter 3 Transforms using
More informationSolving Mixed Integer Linear Programs Using Branch and Cut Algorithm
1 Solving Mixed Integer Linear Programs Using Branch and Cut Algorithm by Shon Albert A Project Submitted to the Graduate Faculty of North Carolina State University in Partial Fulfillment of the Requirements
More informationSimplified Benders cuts for Facility Location
Simplified Benders cuts for Facility Location Matteo Fischetti, University of Padova based on joint work with Ivana Ljubic (ESSEC, Paris) and Markus Sinnl (ISOR, Vienna) Barcelona, November 2015 1 Apology
More informationLinear Programming: Introduction
Linear Programming: Introduction Frédéric Giroire F. Giroire LP  Introduction 1/28 Course Schedule Session 1: Introduction to optimization. Modelling and Solving simple problems. Modelling combinatorial
More informationA Branch and Bound Algorithm for Solving the Binary Bilevel Linear Programming Problem
A Branch and Bound Algorithm for Solving the Binary Bilevel Linear Programming Problem John Karlof and Peter Hocking Mathematics and Statistics Department University of North Carolina Wilmington Wilmington,
More informationApplied Algorithm Design Lecture 5
Applied Algorithm Design Lecture 5 Pietro Michiardi Eurecom Pietro Michiardi (Eurecom) Applied Algorithm Design Lecture 5 1 / 86 Approximation Algorithms Pietro Michiardi (Eurecom) Applied Algorithm Design
More informationA Constraint Programming based Column Generation Approach to Nurse Rostering Problems
Abstract A Constraint Programming based Column Generation Approach to Nurse Rostering Problems Fang He and Rong Qu The Automated Scheduling, Optimisation and Planning (ASAP) Group School of Computer Science,
More informationLecture 3: Linear Programming Relaxations and Rounding
Lecture 3: Linear Programming Relaxations and Rounding 1 Approximation Algorithms and Linear Relaxations For the time being, suppose we have a minimization problem. Many times, the problem at hand can
More informationExact and Heuristic Methods for Mixed Integer Linear Programs
UNIVERSITÀ DEGLI STUDI DI PADOVA Sede Amministrativa: Università degli Studi di Padova Dipartimento di Matematica Pura ed Applicata Dottorato di Ricerca in Matematica Computazionale XVIII Ciclo Exact and
More informationStrategic planning in LTL logistics increasing the capacity utilization of trucks
Strategic planning in LTL logistics increasing the capacity utilization of trucks J. Fabian Meier 1,2 Institute of Transport Logistics TU Dortmund, Germany Uwe Clausen 3 Fraunhofer Institute for Material
More informationPerformance of Optimization Software  an Update
Performance of Optimization Software  an Update INFORMS Annual 2011 Charlotte, NC 1318 November 2011 H. D. Mittelmann School of Math and Stat Sciences Arizona State University 1 Services we provide Guide
More informationSolving Integer Programming with BranchandBound Technique
Solving Integer Programming with BranchandBound Technique This is the divide and conquer method. We divide a large problem into a few smaller ones. (This is the branch part.) The conquering part is done
More informationPractical Guidelines for Solving Difficult Mixed Integer Linear Programs
1 2 3 Practical Guidelines for Solving Difficult Mixed Integer Linear Programs Ed Klotz Alexandra M. Newman IBM, 926 Incline Way, Suite 100, Incline Village, NV 89451 Division of Economics and Business,
More informationAn Overview Of Software For Convex Optimization. Brian Borchers Department of Mathematics New Mexico Tech Socorro, NM 87801 borchers@nmt.
An Overview Of Software For Convex Optimization Brian Borchers Department of Mathematics New Mexico Tech Socorro, NM 87801 borchers@nmt.edu In fact, the great watershed in optimization isn t between linearity
More informationON SOLVING THE PROGRESSIVE PARTY PROBLEM AS A MIP
ON SOLVING THE PROGRESSIVE PARTY PROBLEM AS A MIP ERWIN KALVELAGEN Abstract. The Progressive Party Problem [9] has long been considered a problem intractable for branchandbound mixed integer solvers.
More informationComputational advances in solving Mixed Integer Linear Programming problems
Carnegie Mellon University Research Showcase @ CMU Department of Chemical Engineering Carnegie Institute of Technology 2011 Computational advances in solving Mixed Integer Linear Programming problems Ricardo
More informationScheduling the German Basketball League
Scheduling the German Basketball League Stephan Westphal Institute for Numerical and Applied Mathematics, University of Göttingen, Germany Daniel H. Wagner Prize for Excellence in Operations Research Practice,
More informationHeuristics for Dynamically Adapting Constraint Propagation in Constraint Programming
Heuristics for Dynamically Adapting Constraint Propagation in Constraint Programming Kostas Stergiou AI Lab University of the Aegean Greece CPAIOR 09 Workshop on Bound reduction techniques for CP and MINLP
More informationLocalSolver: blackbox local search for combinatorial optimization
LocalSolver: blackbox local search for combinatorial optimization Frédéric Gardi Bouygues elab, Paris fgardi@bouygues.com Joint work with T. Benoist, J. Darlay, B. Estellon, R. Megel, K. Nouioua Bouygues
More informationPerformance of Commercial and Noncommercial Optimization Software
Performance of Commercial and Noncommercial Optimization Software IINFORMS 2012 Phoenix, AZ October 2012 H. D. Mittelmann School of Math and Stat Sciences Arizona State University 1 Services we provide
More informationScheduling Home Health Care with Separating Benders Cuts in Decision Diagrams
Scheduling Home Health Care with Separating Benders Cuts in Decision Diagrams André Ciré University of Toronto John Hooker Carnegie Mellon University INFORMS 2014 Home Health Care Home health care delivery
More informationINTEGER PROGRAMMING. Integer Programming. Prototype example. BIP model. BIP models
Integer Programming INTEGER PROGRAMMING In many problems the decision variables must have integer values. Example: assign people, machines, and vehicles to activities in integer quantities. If this is
More informationDantzigWolfe bound and DantzigWolfe cookbook
DantzigWolfe bound and DantzigWolfe cookbook thst@man.dtu.dk DTUManagement Technical University of Denmark 1 Outline LP strength of the DantzigWolfe The exercise from last week... The DantzigWolfe
More informationOptimized Scheduling in RealTime Environments with Column Generation
JG U JOHANNES GUTENBERG UNIVERSITAT 1^2 Optimized Scheduling in RealTime Environments with Column Generation Dissertation zur Erlangung des Grades,.Doktor der Naturwissenschaften" am Fachbereich Physik,
More informationParallel Branch and Cut for Dummies
Parallel Branch and Cut for Dummies Ted Ralphs Department of Industrial and Manufacturing Systems Engineering Lehigh University Bethlehem, PA 1 Outline of Talk Introduction to Parallel Systems Introduction
More informationMeasuring Rationality with the Minimum Cost of Revealed Preference Violations. Mark Dean and Daniel Martin. Online Appendices  Not for Publication
Measuring Rationality with the Minimum Cost of Revealed Preference Violations Mark Dean and Daniel Martin Online Appendices  Not for Publication 1 1 Algorithm for Solving the MASP In this online appendix
More informationIntroduction: Models, Model Building and Mathematical Optimization The Importance of Modeling Langauges for Solving Real World Problems
Introduction: Models, Model Building and Mathematical Optimization The Importance of Modeling Langauges for Solving Real World Problems Josef Kallrath Structure of the Lecture: the Modeling Process survey
More informationSection Notes 4. Duality, Sensitivity, Dual Simplex, Complementary Slackness. Applied Math 121. Week of February 28, 2011
Section Notes 4 Duality, Sensitivity, Dual Simplex, Complementary Slackness Applied Math 121 Week of February 28, 2011 Goals for the week understand the relationship between primal and dual programs. know
More informationRouting in Line Planning for Public Transport
KonradZuseZentrum für Informationstechnik Berlin Takustraße 7 D14195 BerlinDahlem Germany MARC E. PFETSCH RALF BORNDÖRFER Routing in Line Planning for Public Transport Supported by the DFG Research
More informationSolving NP Hard problems in practice lessons from Computer Vision and Computational Biology
Solving NP Hard problems in practice lessons from Computer Vision and Computational Biology Yair Weiss School of Computer Science and Engineering The Hebrew University of Jerusalem www.cs.huji.ac.il/ yweiss
More informationApproximation Algorithms: LP Relaxation, Rounding, and Randomized Rounding Techniques. My T. Thai
Approximation Algorithms: LP Relaxation, Rounding, and Randomized Rounding Techniques My T. Thai 1 Overview An overview of LP relaxation and rounding method is as follows: 1. Formulate an optimization
More information5 INTEGER LINEAR PROGRAMMING (ILP) E. Amaldi Fondamenti di R.O. Politecnico di Milano 1
5 INTEGER LINEAR PROGRAMMING (ILP) E. Amaldi Fondamenti di R.O. Politecnico di Milano 1 General Integer Linear Program: (ILP) min c T x Ax b x 0 integer Assumption: A, b integer The integrality condition
More informationHet inplannen van besteld ambulancevervoer (Engelse titel: Scheduling elected ambulance transportation)
Technische Universiteit Delft Faculteit Elektrotechniek, Wiskunde en Informatica Delft Institute of Applied Mathematics Het inplannen van besteld ambulancevervoer (Engelse titel: Scheduling elected ambulance
More informationUsing computing resources with IBM ILOG CPLEX CO@W2015
CPLEX Optimization IBM Germany 20151006 Using computing resources with IBM ILOG CPLEX CO@W2015 Hardware resources Multiple cores/threads Multiple machines No machines Software resources Interfacing with
More informationIntegrating Benders decomposition within Constraint Programming
Integrating Benders decomposition within Constraint Programming Hadrien Cambazard, Narendra Jussien email: {hcambaza,jussien}@emn.fr École des Mines de Nantes, LINA CNRS FRE 2729 4 rue Alfred Kastler BP
More informationScheduling of Mixed BatchContinuous Production Lines
Université Catholique de Louvain Faculté des Sciences Appliquées Scheduling of Mixed BatchContinuous Production Lines Thèse présentée en vue de l obtention du grade de Docteur en Sciences Appliquées par
More informationOPTIMIZED STAFF SCHEDULING AT SWISSPORT
Gurobi User Conference, Frankfurt, 01.02.2016 OPTIMIZED STAFF SCHEDULING AT SWISSPORT Prof. Dr. Andreas Klinkert Dr. Peter Fusek Dipl. Ing. Roman Berner Rita Thalmann Simona Segessenmann Zurich University
More informationScheduling and (Integer) Linear Programming
Scheduling and (Integer) Linear Programming Christian Artigues LAAS  CNRS & Université de Toulouse, France artigues@laas.fr Master Class CPAIOR 2012  Nantes Christian Artigues Scheduling and (Integer)
More informationHighperformance local search for planning maintenance of EDF nuclear park
Highperformance local search for planning maintenance of EDF nuclear park Frédéric Gardi Karim Nouioua Bouygues elab, Paris fgardi@bouygues.com Laboratoire d'informatique Fondamentale  CNRS UMR 6166,
More informationComputer Sciences Department
Computer Sciences Department Algorithms and Software for Convex Mixed Integer Nonlinear Programs Pierre Bonami Mustafa Kilinc Jeff Linderoth Technical Report #1664 October 2009 ALGORITHMS AND SOFTWARE
More informationExact and Approximate Sparse Solutions of Underdetermined Linear Equations
KonradZuseZentrum für Informationstechnik Berlin Takustraße 7 D14195 BerlinDahlem Germany SADEGH JOKAR MARC E. PFETSCH Exact and Approximate Sparse Solutions of Underdetermined Linear Equations Supported
More informationHow to speedup hard problem resolution using GLPK?
How to speedup hard problem resolution using GLPK? Onfroy B. & Cohen N. September 27, 2010 Contents 1 Introduction 2 1.1 What is GLPK?.......................................... 2 1.2 GLPK, the best one?.......................................
More informationResource Allocation Modeling Techniques Applied to Air Force Crew Scheduling Problem
Kepler Research, Inc. Proprietary Resource Allocation Modeling Techniques Applied to Air Force Crew Scheduling Problem Submitted by: Kepler Research, Inc. February 2012 1. Introduction and Background The
More informationSymmetry Handling in Binary Programs via Polyhedral Methods
Symmetry Handling in Binary Programs via Polyhedral Methods Christopher Hojny joint work with Marc Pfetsch Technische Universität Darmstadt Department of Mathematics 26 International Conference on Mathematical
More informationChapter 3 INTEGER PROGRAMMING 3.1 INTRODUCTION. Robert Bosch. Michael Trick
Chapter 3 INTEGER PROGRAMMING Robert Bosch Oberlin College Oberlin OH, USA Michael Trick Carnegie Mellon University Pittsburgh PA, USA 3.1 INTRODUCTION Over the last 20 years, the combination of faster
More informationSome Optimization Fundamentals
ISyE 3133B Engineering Optimization Some Optimization Fundamentals Shabbir Ahmed Email: sahmed@isye.gatech.edu Homepage: www.isye.gatech.edu/~sahmed Basic Building Blocks min or max s.t. objective as
More informationIn a previous paper we (Weiss and Rasmussen 2007) demonstrated lessons that can be learned by formulating
Vol. 7, No. 3, May 2007, pp. 228 237 issn 15320545 07 0703 0228 informs I N F O R M S Transactions on Education doi 10.1287/ited.7.3.228 2007 INFORMS Advanced Lessons on the Craft of Optimization Modeling
More informationA short OPL tutorial
A short OPL tutorial Truls Flatberg January 17, 2009 1 Introduction This note is a short tutorial to the modeling language OPL. The tutorial is by no means complete with regard to all features of OPL.
More informationQuality Assurance For Mathematical Modeling Systems
Quality Assurance For Mathematical Modeling Systems Armin Pruessner, Michael Bussieck, Steven Dirkse, Alex Meeraus GAMS Development Corporation 1217 Potomac Street NW Washington, DC 20007 1 Agenda Motivation
More informationMathematical finance and linear programming (optimization)
Mathematical finance and linear programming (optimization) Geir Dahl September 15, 2009 1 Introduction The purpose of this short note is to explain how linear programming (LP) (=linear optimization) may
More informationModeling and Solving the Capacitated Vehicle Routing Problem on Trees
in The Vehicle Routing Problem: Latest Advances and New Challenges Modeling and Solving the Capacitated Vehicle Routing Problem on Trees Bala Chandran 1 and S. Raghavan 2 1 Department of Industrial Engineering
More informationMinimally Infeasible Set Partitioning Problems with Balanced Constraints
Minimally Infeasible Set Partitioning Problems with alanced Constraints Michele Conforti, Marco Di Summa, Giacomo Zambelli January, 2005 Revised February, 2006 Abstract We study properties of systems of
More informationMATHEMATICAL ENGINEERING TECHNICAL REPORTS
MATHEMATICAL ENGINEERING TECHNICAL REPORTS Truss Topology Optimization with a Limited Number of Different CrossSections: A MixedInteger SecondOrder Cone Programming Approach Yoshihiro KANNO METR 2014
More informationBasic solutions in standard form polyhedra
Recap, and outline of Lecture Previously min s.t. c x Ax = b x Converting any LP into standard form () Interpretation and visualization () Full row rank assumption on A () Basic solutions and bases in
More informationLecture 3. Linear Programming. 3B1B Optimization Michaelmas 2015 A. Zisserman. Extreme solutions. Simplex method. Interior point method
Lecture 3 3B1B Optimization Michaelmas 2015 A. Zisserman Linear Programming Extreme solutions Simplex method Interior point method Integer programming and relaxation The Optimization Tree Linear Programming
More informationENERGY EFFICIENT CONTROL OF VIRTUAL MACHINE CONSOLIDATION UNDER UNCERTAIN INPUT PARAMETERS FOR THE CLOUD
ENERGY EFFICIENT CONTROL OF VIRTUAL MACHINE CONSOLIDATION UNDER UNCERTAIN INPUT PARAMETERS FOR THE CLOUD ENRICA ZOLA, KARLSTAD UNIVERSITY @IEEE.ORG ENGINEERING AND CONTROL FOR RELIABLE CLOUD SERVICES,
More informationBranch, Cut, and Price: Sequential and Parallel
Branch, Cut, and Price: Sequential and Parallel T.K. Ralphs 1, L. Ladányi 2, and L.E. Trotter, Jr. 3 1 Department of Industrial and Manufacturing Systems Engineering, Lehigh University, Bethlehem, PA 18017,
More informationChapter 6: Sensitivity Analysis
Chapter 6: Sensitivity Analysis Suppose that you have just completed a linear programming solution which will have a major impact on your company, such as determining how much to increase the overall production
More informationIntegrated maintenance scheduling for semiconductor manufacturing
Integrated maintenance scheduling for semiconductor manufacturing Andrew Davenport davenport@us.ibm.com Department of Business Analytics and Mathematical Science, IBM T. J. Watson Research Center, P.O.
More informationMinimizing costs for transport buyers using integer programming and column generation. Eser Esirgen
MASTER STHESIS Minimizing costs for transport buyers using integer programming and column generation Eser Esirgen DepartmentofMathematicalSciences CHALMERS UNIVERSITY OF TECHNOLOGY UNIVERSITY OF GOTHENBURG
More information1 Polyhedra and Linear Programming
CS 598CSC: Combinatorial Optimization Lecture date: January 21, 2009 Instructor: Chandra Chekuri Scribe: Sungjin Im 1 Polyhedra and Linear Programming In this lecture, we will cover some basic material
More informationTwo objective functions for a real life Split Delivery Vehicle Routing Problem
International Conference on Industrial Engineering and Systems Management IESM 2011 May 25  May 27 METZ  FRANCE Two objective functions for a real life Split Delivery Vehicle Routing Problem Marc Uldry
More informationInverse Optimization by James Orlin
Inverse Optimization by James Orlin based on research that is joint with Ravi Ahuja Jeopardy 000  the Math Programming Edition The category is linear objective functions The answer: When you maximize
More informationSome representability and duality results for convex mixedinteger programs.
Some representability and duality results for convex mixedinteger programs. Santanu S. Dey Joint work with Diego Morán and Juan Pablo Vielma December 17, 2012. Introduction About Motivation Mixed integer
More informationLECTURE 5: DUALITY AND SENSITIVITY ANALYSIS. 1. Dual linear program 2. Duality theory 3. Sensitivity analysis 4. Dual simplex method
LECTURE 5: DUALITY AND SENSITIVITY ANALYSIS 1. Dual linear program 2. Duality theory 3. Sensitivity analysis 4. Dual simplex method Introduction to dual linear program Given a constraint matrix A, right
More information