Mixed Integer Nonlinear Programs Theory, Algorithms & Applications
|
|
- Eleanore Reynolds
- 5 years ago
- Views:
Transcription
1 Mixed Integer Nonlinear Programs Theory, Algorithms & Applications Mohit Tawarmalani Department of Mechanical and Industrial Engineering Nick Sahinidis Department of Chemical Engineering University Of Illinois Urbana-Champaign Acknowledgment: NSF CAREER award DMII
2 Problem Formulation (P) min f(x, y) Objective Function s.t. g(x, y) 0 Constraints x Z p Integrality Restrictions y R n Continuous Variables Challenges: Multimodal Objective Objective Integrality f(x) Feasible Space Nonconvex Constraints f(x) Convex Objective Projected Objective Nonconvex Feasible Space
3 The Pooling Problem f 11 f 21 3% S $6 1% S $16 Pool x 11 x 21 Blend X 2.5% S $9 X 100 f 12 2% S $10 x 22 x 12 Blend Y 1.5% S $15 Y 200 min cost {}}{ 6f f f 12 X-revenue {}}{ 9(x 11 + x 21 ) Y -revenue {}}{ 15(x 12 + x 22 ) s.t. q = 3f 11 + f 21 x 11 + x 12 Concentration Balance f 11 + f 21 = x 11 + x 12 f 12 = x 21 + x 22 Mass balance qx 11 +2x 21 x 11 + x qx 12 +2x 22 x 12 + x Quality Requirements x 11 + x x 12 + x Demands
4 Molecular Design Problems (Example: Refrigerant Design) UNIVERSE C Cl CH 3 N Cl CH CH 2 2 F CH N Cl CH 2 Cl F F Combinatorial Choice (Source of Discreteness) Freon (CCl 2 F 2 ) F F C Cl Cl Property Prediction (Source of Nonlinearity) Satisfies Thermodynamic requirements?
5 Discrete Location Problems Restaurants in Edmonton p-choice Utility = w j location: j customer: i U ij x j k U d i ikx k }{{} Utility Ratio U ij x j d i w j Utility of location j for customer i decision variable for locating facility j demand by customer i preferential weight of site j
6 Motivation for an Algorithm Applications: Combinatorial Chemistry and Molecular Biology Utility Maximization (Risk Minimization) Parameter Estimation (Design of Experiments) Engineering Design (Process Synthesis, CAD) Layout Design (Area Restrictions) Theoretical Challenges Important Subclasses: Mixed Integer Linear Programming Continuous Nonlinear Programming Nonlinear 0-1 Programming Search for Globality (Deterministic): Either optimality assured or a gap estimate Rigorous termination criteria Verifying local optimality is NP-hard (Murty and Kabadi 1987)
7 Deterministic Approaches Branch and Bound Bound the problem over successively refined partitions Convexification Outer-approximate with increasingly tighter convex programs Decomposition Project out some variables (move them to a subproblem) Our approach Branch and Bound + Convexification
8 Branch and Bound Algorithm Objective P Objective P U R R L Variable L Variable a. Lower Bounding b. Upper Bounding Objective P U L R1 R2 R R1 R R2 Fathom Variable Subdivide c. Domain Subdivision d. Search Tree
9 Illustrative Example 4 f = 5 x 2 x 1 x 1 x 2 f = Separable Reformulation : x 1 x 2 = 1 2 min x 1 x 2 s.t. x 1 x x x 2 4 { (x1 + x 2 ) 2 x 2 1 2} x2 Reformulation min x 1 x 2 s.t. (x 1 + x 2 ) 2 x 2 1 x x x 2 4 Relaxation (not best) 0 x 1 6 6x 1 36 x 2 1 min x 1 x 2 s.t. (x 1 + x 2 ) 2 6x 1 4x x x 2 4 Relaxation Solution: x 1 =6,x 2 =0.89, L = 6.89 Local Search (MINOS): x 1 =6,x 2 =0.67, L = 6.67 Relaxation Solution provides good starting point for local search
10 Challenges Standard Global Optimization Converges Slowly Algorithmic Issues: Tight Relaxations Domain Reduction Finiteness Issues Implementation Issues: Existing NLP solvers often unreliable General purpose (automatic) software Enable solution of industrially relevant problems
11 Factorable Functions (McCormick, 1976) Definition: The factorable functions are recursive compositions of sums and products of functions of single variables. Example: f(x, y) = exp(xy + z ln w)z 3 f { }} { x 5 {}}{ x 3 x 6 ( {}}{{}}{ exp( xy + z ln w }{{}}{{} ) z 3 ) 0.5 x 1 } {{ x 2 } } x 4 {{ } x 7 x 1 = xy x 2 =ln(w) x 3 = zx 2 x 4 = x 1 + x 3 x 5 =exp(x 4 ) x 6 = z 3 x 7 = x 5 x 6 f = x 7
12 Challenges Via Example: y log(x) z 1 x 8 1 y 4 z y log(x) 1 x 8 1 y 4 z yl x l x log(x) 1 x 8 1 y 4 Convex Relaxation (McCormick style) z log(8)y +4 l r 4log(8) z 1 x 8 1 y 4 z l r l r log(8) (x 1) 7 Q: In general do we get the convex envelope? NO Q: In this case? YES (Convex Extensions Wed-9:00am) Q: Simplification? (log 8/7) max {7y +4x 32,x 1} - Non-differentiability -log(x) occurs elsewhere? (common subexpressions) Q: What if y = x? (Note:x log x is convex)
13 Ratio: Convex Extensions x 4 2 y 4 z x/y 2 x 4 2 y 4 Ratio: x/y Convex Extensions yield Convex Envelope y 4 2 x x/y -Envelope Factorable Relaxation is Weak 4 y p z p x L ( x U x x U x L ) 2 ) 2 ( x x y p (z z p ) y(z z p ) x U L x U x { L y p max y L xu x,y x } xl x U yu xl x U x { L } y p min y U xu x x xl,y x U yl xl x U x L z z p,z p y x 4 2 x/y -Factorable 4 z 0.5 +x/4 +y/8 z 2+x/2 y 2 x 4 2 y 4
14 RANGE REDUCTION Relaxed Value Function z U L x L x U x If a variable goes to its upper bound at the relaxed problem solution, this variable s lower bound can be improved
15 PROBING Q. A. What if a variable does not go to a bound? Use probing: temporarily fix variable at a bound. Relaxed Value Function z L U z U x L x* x U x
16 ILLUSTRATIVE EXAMPLE x 2 min f = x 1 x 2 s.t. x 1 x x x 2 4 (P) Reformulation: min f = x 1 x 2 s.t. x 2 3 x 2 1 x 2 8 x 3 = x 1 + x 2 0 x x x f = Relaxation: f = 6.67 min x 1 x 2 s.t. x 2 3 6x 1 4x 2 8 x 3 = x 1 + x 2 0 x x x 3 10 x 1 (R) Solution of R : x 1 = 6, x 2 = 0.89,L = 6.89,λ 1 = 0.2 Local Solution of P with MINOS : U = 6.67 Range reduction : x L 1 x U 1 (U L)/ λ 1 = 4.86 Probing (Solve R with x 2 0) : L = 6,λ 2 = 1 x L 2 = 0.67 Update R with x , x Solution is : L = 6.67 Proof of globality with NO Branching!!
17 Branch And Reduce Optimization Navigator BARON Preprocessor Data Organizer Range Reduction Solver Links I/O Handler Debugging Facilities USER MODULE Relaxation Local Search Range Reduction Heuristics Sparse Matrix Utilities Heuristics Core Module: expandable, application-independent Application Modules: factorable, linear multiplicative, bilinear, concave, fixed-charge, integer, fractional programming solve convex relaxations using CPLEX, OSL, MINOS, SNOPT, SDPA
18 Factorable NLP Module Formulation: min f(x, y) a g i (x, y) b i =1,...,m x L j x j x U j j =1,...,n y L j y j y U j j =1,...,p x j real y j integer Factorable Functions: f and g are recursive sums, products and ratios of the following terms: exponential, exp(x) logarithmic, log(x) monomial, x a Example: (Factorable Constraint) ( ) 1 x2 y 0.3 z x 2 p +exp xy 100 p 2 y
19 Pooling Problems in Literature Algorithm Foulds 92 Ben-Tal 94 GOP 96 BARON Computer CDC 4340 HP9000/730 RS6000/43P Linpack > Tolerance ** 10 6 Problem N tot T tot N tot N tot T tot N tot T tot Haverly Haverly Haverly Foulds Foulds Foulds Foulds Ben-Tal Ben-Tal Example Example Example Example * Blank indicates problem not reported or not solved ** 0.05% for Haverly 1, 2, 3, 0.05% for Ben-Tal 4 and 1% for Ben-Tal 5
20 Automotive Refrigerant Design (Joback and Stephanopoulos, 1984) Condenser T cnd Expansion Valve Tavg = (T cnd + Tevp)/2 Compressor Tevp Evaporator H ve max C pla s.t. H ve 18.4 C pla 32.2 P vpe 1.4 Property Prediction Constraints Structure Feasibility Constraints Nonnegativity Constraints Integrality Requirements High enthalpy of vaporization H ve reduces the amount of refrigerant Lower liquid heat capacity C pla reduces amount of vapor generated in expansion valve
21 Functional Groups Considered Acyclic Groups Cyclic Groups Halogen Groups Oxygen Groups Nitrogen Groups Sulfur Groups CH 3 r CH 2 r F OH NH 2 SH CH 2 r r > CH r Cl O > NH S > CH r > CH r Br r O r r r > NH r S r > C < r r > C < r r I > CO > N =CH 2 r > C < r r r r > CO =N =CH > C < r r CHO r =N r =C< r =CH r COOH CN =C= r =C< r r COO NO 2 CH r =C< r =O C =C< r r Number of Groups = 44 Maximum Selection Size = 15 Candidates = 39, 895, 566, 894, 524
22 Property Prediction Constraints T b = N n i T bi i=1 β = T br ln(T br ) T 6 br T b T c = N i=1 n it ci ( N i=1 n it ci ) 2 1 P c = ( N i=1 n ia i N i=1 n ip ci ) ( 2 N N ) C p0a = n i C p0ai n i C p0bi i=1 i=1 ( N ) + n i C p0ci T 2 avg T br = T b T c T avgr = T avg T c T cndr = T cnd T c T evpr = T evp T c i=1 ( N ) + n i C p0di T 3 avg i=1 α = ln T 6 br ( Pc ) T avg T br ln(T br ) ω = α β C pla = 1 { [ C p0a ω T avgr ( (1 T avgr) 1/ )]} T avgr 1 T avgr N H vb = n i H vbi i=1 ( 1 Tevp /T c H ve = H vb 1 T b /T c h = T br ln(p c /1.013) 1 T br G = h k = h/g (1 + T br) (3 + T br )(1 T br ) 2 ) 0.38 ln P vpcr = G T cndr [ 1 T 2 cndr + k(3 + T cndr)(1 T cndr ) 3] ln P vper = G T evpr [ 1 T 2 evpr + k(3 + T evpr)(1 T evpr ) 3] n i integer
23 Structure Feasibility Constraints N n i 2 i=1 Y A n i N max Y A A i A Y C n i N max Y C C i C Y M n i N max Y M M i M Y A + Y C 1 Y M Y A + Y C 3Y R n i N max Y R R i R N N n i b i 2( n i 1); i=1 i=1 ( N N )( N ) n i b i n i n i 1 i=1 n i 1if i=1 i=1 n i 1and i SD i S/D i D/S n i 1 n i 1if n i 1and n i 1 i ST i S/T i T /S n i 1if n i 1and n i 1 i SSR i S/SR i SR/S n i 1if n i 1and n i 1 i SDR i S/DR i DR/S n i 1if n i 1and n i 1 i DSR i D/SR i SR/D n i =2Z S ; n i =2Z SR n i =2Z B ; i B i S n i =2Z D i D n i =2Z DR i DR n i =2Z T i T n i (2 b i )=2m i N n i n j (b j 1) + 2 i=1 i SR j =1,...,N n i n i S ai if n i 0 i O, i H i H single-bonded n i n i D ai if n i 0 i O, i H i H double-bonded n i n i T ai if n i 0 i O, i H i H triple-bonded i H n i (S ai + D ai + T ai ) i O n i 2( i H n i 1)
24 Molecular Structures Molecular Structure H ve C pla Molecular Structure H ve C pla ClNO (Cl )( N =)(= O) IF (I )( F) O 3 ( r O r ) FNO (F )( N =)(= O) CHClO (Cl )( CH =)(= O) FSH (F )( SH) CH 3 Cl (CH 3 )( Cl) C 2 HClO 2 (= C <)(= CH )( Cl)(= O) All 44 feasible molecules identified CH 2 FN (F )( N =)(= CH 2 ) ClF (Cl )( F) CClFO (O =)(= C <)( Cl)( F) ClFO (Cl )( O )( F) C 3 H 4 (CH 3 )( C )( CH) C 2 F 2 ( C ) 2 ( F) C 3 H 6 O ( CH 3 ) 2 (= C <)(= O) C 3 H 3 FO (O =)(= CH ) 3 ( F) CF 2 NHO (O =)(= C <)( F ) 2 (> NH) C 2 H 6 ( CH 3 ) CHFO 3 (O =)(= CH )( O ) 2 ( F) C 3 H 3 F 3 ( r > CH r ) 3 ( F) For CCl 2 F 2, H ve /C pla 0.57
25 Located Restaurants IBM SP/2 Single Processor Time: 5.5 hours Relaxations take 95% time Iterations: 79 Nodes in Memory: 9
26 Miscellaneous Problems Problem m n N tot N mem T tot Bilinear Problem Design of Water Pump Alkylation Process Design Design of Insulated Tank HENS Chemical Equilibrium Pooling Problem Quadratic Bilinear Nonlinear Equality Economies of Scale Two-Stage Process Systems MINLP, Process Synthesis MINLP, Process Synthesis MINLP, Process Synthesis HENS Reinforced Concrete Beam Quadratic Constraints QCQP Reactor Network Design Two-Stage Process System Problem m n N tot N mem T tot Concave QP Biconvex Program Linearly Constrained QP Linear Multiplicative Concave QP Nonlinear Fixed Charges QCQP Reliability Fixture Design Fixture Design Molecular Design Bilinearly Constrained Bilinearly Constrained HENS MINLP Parameter Estimation Pressure Vessel Design Shekel Function Truss Design Reliability Design of Experiments
27 Convexify Reduce Isolate x* BRANCH-AND-REDUCE Facility Location Molecular design Supply chain operations
GLOBAL OPTIMIZATION WITH GAMS/BARON
GLOBAL OPTIMIZATION WITH GAMS/BARON Nick Sahinidis Chemical and Biomolecular Engineering University of Illinois at Urbana Mohit Tawarmalani Krannert School of Management Purdue University MIXED-INTEGER
More informationGlobal Optimization and Constraint Satisfaction: THE BRANCH-AND- REDUCE APPROACH
Global Optimization and Constraint Satisfaction: THE BRANCH-AND- REDUCE APPROACH Nick Sahinidis University of Illinois at Urbana-Champaign Chemical and Biomolecular Eng. ACKNOWLEDGMENTS Shabbir Ahmed:
More informationSolving Mixed-Integer Nonlinear Programs
Solving Mixed-Integer Nonlinear Programs (with SCIP) Ambros M. Gleixner Zuse Institute Berlin MATHEON Berlin Mathematical School 5th Porto Meeting on Mathematics for Industry, April 10 11, 2014, Porto
More informationMixed-Integer Nonlinear Programming
Mixed-Integer Nonlinear Programming Claudia D Ambrosio CNRS researcher LIX, École Polytechnique, France pictures taken from slides by Leo Liberti MPRO PMA 2016-2017 Motivating Applications Nonlinear Knapsack
More informationComputer Sciences Department
Computer Sciences Department Valid Inequalities for the Pooling Problem with Binary Variables Claudia D Ambrosio Jeff Linderoth James Luedtke Technical Report #682 November 200 Valid Inequalities for the
More informationImplementation of an αbb-type underestimator in the SGO-algorithm
Implementation of an αbb-type underestimator in the SGO-algorithm Process Design & Systems Engineering November 3, 2010 Refining without branching Could the αbb underestimator be used without an explicit
More informationSoftware for Integer and Nonlinear Optimization
Software for Integer and Nonlinear Optimization Sven Leyffer, leyffer@mcs.anl.gov Mathematics & Computer Science Division Argonne National Laboratory Roger Fletcher & Jeff Linderoth Advanced Methods and
More informationCan Li a, Ignacio E. Grossmann a,
A generalized Benders decomposition-based branch and cut algorithm for two-stage stochastic programs with nonconvex constraints and mixed-binary first and second stage variables Can Li a, Ignacio E. Grossmann
More informationCan Li a, Ignacio E. Grossmann a,
A generalized Benders decomposition-based branch and cut algorithm for two-stage stochastic programs with nonconvex constraints and mixed-binary first and second stage variables Can Li a, Ignacio E. Grossmann
More informationNonconvex Generalized Benders Decomposition Paul I. Barton
Nonconvex Generalized Benders Decomposition Paul I. Barton Process Systems Engineering Laboratory Massachusetts Institute of Technology Motivation Two-stage Stochastic MINLPs & MILPs Natural Gas Production
More informationHeuristics and Upper Bounds for a Pooling Problem with Cubic Constraints
Heuristics and Upper Bounds for a Pooling Problem with Cubic Constraints Matthew J. Real, Shabbir Ahmed, Helder Inàcio and Kevin Norwood School of Chemical & Biomolecular Engineering 311 Ferst Drive, N.W.
More informationMultiperiod Blend Scheduling Problem
ExxonMobil Multiperiod Blend Scheduling Problem Juan Pablo Ruiz Ignacio E. Grossmann Department of Chemical Engineering Center for Advanced Process Decision-making University Pittsburgh, PA 15213 1 Motivation
More informationDevelopment of the new MINLP Solver Decogo using SCIP - Status Report
Development of the new MINLP Solver Decogo using SCIP - Status Report Pavlo Muts with Norman Breitfeld, Vitali Gintner, Ivo Nowak SCIP Workshop 2018, Aachen Table of contents 1. Introduction 2. Automatic
More informationApplications and algorithms for mixed integer nonlinear programming
Applications and algorithms for mixed integer nonlinear programming Sven Leyffer 1, Jeff Linderoth 2, James Luedtke 2, Andrew Miller 3, Todd Munson 1 1 Mathematics and Computer Science Division, Argonne
More informationA Branch-and-Refine Method for Nonconvex Mixed-Integer Optimization
A Branch-and-Refine Method for Nonconvex Mixed-Integer Optimization Sven Leyffer 2 Annick Sartenaer 1 Emilie Wanufelle 1 1 University of Namur, Belgium 2 Argonne National Laboratory, USA IMA Workshop,
More informationNetwork Flows. 6. Lagrangian Relaxation. Programming. Fall 2010 Instructor: Dr. Masoud Yaghini
In the name of God Network Flows 6. Lagrangian Relaxation 6.3 Lagrangian Relaxation and Integer Programming Fall 2010 Instructor: Dr. Masoud Yaghini Integer Programming Outline Branch-and-Bound Technique
More informationDisconnecting Networks via Node Deletions
1 / 27 Disconnecting Networks via Node Deletions Exact Interdiction Models and Algorithms Siqian Shen 1 J. Cole Smith 2 R. Goli 2 1 IOE, University of Michigan 2 ISE, University of Florida 2012 INFORMS
More informationMINLP: Theory, Algorithms, Applications: Lecture 3, Basics of Algorothms
MINLP: Theory, Algorithms, Applications: Lecture 3, Basics of Algorothms Jeff Linderoth Industrial and Systems Engineering University of Wisconsin-Madison Jonas Schweiger Friedrich-Alexander-Universität
More informationMixed Integer Non Linear Programming
Mixed Integer Non Linear Programming Claudia D Ambrosio CNRS Research Scientist CNRS & LIX, École Polytechnique MPRO PMA 2016-2017 Outline What is a MINLP? Dealing with nonconvexities Global Optimization
More informationConvex Quadratic Relaxations of Nonconvex Quadratically Constrained Quadratic Progams
Convex Quadratic Relaxations of Nonconvex Quadratically Constrained Quadratic Progams John E. Mitchell, Jong-Shi Pang, and Bin Yu Original: June 10, 2011 Abstract Nonconvex quadratic constraints can be
More information4y Springer NONLINEAR INTEGER PROGRAMMING
NONLINEAR INTEGER PROGRAMMING DUAN LI Department of Systems Engineering and Engineering Management The Chinese University of Hong Kong Shatin, N. T. Hong Kong XIAOLING SUN Department of Mathematics Shanghai
More informationOutline. 1 Introduction. 2 Modeling and Applications. 3 Algorithms Convex. 4 Algorithms Nonconvex. 5 The Future of MINLP? 2 / 126
Outline 1 Introduction 2 3 Algorithms Convex 4 Algorithms Nonconvex 5 The Future of MINLP? 2 / 126 Larry Leading Off 3 / 126 Introduction Outline 1 Introduction Software Tools and Online Resources Basic
More informationGlobal optimization of mixed-integer nonlinear programs: A theoretical and computational study
Math. Program., Ser. A 99: 563 591 (2004) Digital Object Identifier (DOI) 10.1007/s10107-003-0467-6 Mohit Tawarmalani Nikolaos V. Sahinidis Global optimization of mixed-integer nonlinear programs: A theoretical
More informationAn exact reformulation algorithm for large nonconvex NLPs involving bilinear terms
An exact reformulation algorithm for large nonconvex NLPs involving bilinear terms Leo Liberti CNRS LIX, École Polytechnique, F-91128 Palaiseau, France. E-mail: liberti@lix.polytechnique.fr. Constantinos
More information23. Cutting planes and branch & bound
CS/ECE/ISyE 524 Introduction to Optimization Spring 207 8 23. Cutting planes and branch & bound ˆ Algorithms for solving MIPs ˆ Cutting plane methods ˆ Branch and bound methods Laurent Lessard (www.laurentlessard.com)
More informationA Decomposition-based MINLP Solution Method Using Piecewise Linear Relaxations
A Decomposition-based MINLP Solution Method Using Piecewise Linear Relaxations Pradeep K. Polisetty and Edward P. Gatzke 1 Department of Chemical Engineering University of South Carolina Columbia, SC 29208
More informationRelaxations of multilinear convex envelopes: dual is better than primal
of multilinear convex envelopes: dual is better than primal 1 LIX, École Polytechnique, Palaiseau, France June 7th, 2012 11 th International Symposium on Experimental Algorithms (SEA) 2012 - Bordeaux (France)
More informationMath Models of OR: Branch-and-Bound
Math Models of OR: Branch-and-Bound John E. Mitchell Department of Mathematical Sciences RPI, Troy, NY 12180 USA November 2018 Mitchell Branch-and-Bound 1 / 15 Branch-and-Bound Outline 1 Branch-and-Bound
More informationThe Pooling Problem - Applications, Complexity and Solution Methods
The Pooling Problem - Applications, Complexity and Solution Methods Dag Haugland Dept. of Informatics University of Bergen Norway EECS Seminar on Optimization and Computing NTNU, September 30, 2014 Outline
More informationA Tighter Piecewise McCormick Relaxation for Bilinear Problems
A Tighter Piecewise McCormick Relaxation for Bilinear Problems Pedro M. Castro Centro de Investigação Operacional Faculdade de Ciências niversidade de isboa Problem definition (NP) Bilinear program min
More informationGlobal optimization of minimum weight truss topology problems with stress, displacement, and local buckling constraints using branch-and-bound
INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING Int. J. Numer. Meth. Engng 2004; 61:1270 1309 (DOI: 10.1002/nme.1112) Global optimization of minimum weight truss topology problems with stress,
More informationMILP reformulation of the multi-echelon stochastic inventory system with uncertain demands
MILP reformulation of the multi-echelon stochastic inventory system with uncertain demands Axel Nyberg Åbo Aademi University Ignacio E. Grossmann Dept. of Chemical Engineering, Carnegie Mellon University,
More informationDevelopment of an algorithm for solving mixed integer and nonconvex problems arising in electrical supply networks
Development of an algorithm for solving mixed integer and nonconvex problems arising in electrical supply networks E. Wanufelle 1 S. Leyffer 2 A. Sartenaer 1 Ph. Toint 1 1 FUNDP, University of Namur 2
More informationStochastic Integer Programming An Algorithmic Perspective
Stochastic Integer Programming An Algorithmic Perspective sahmed@isye.gatech.edu www.isye.gatech.edu/~sahmed School of Industrial & Systems Engineering 2 Outline Two-stage SIP Formulation Challenges Simple
More informationREPORTS IN INFORMATICS
REPORTS IN INFORMATICS ISSN 0333-3590 Sensitivity Analysis Applied to the Pooling Problem Lennart Frimannslund, Geir Gundersen and Dag Haugland REPORT NO 380 December 2008 Department of Informatics UNIVERSITY
More informationRelaxations and Randomized Methods for Nonconvex QCQPs
Relaxations and Randomized Methods for Nonconvex QCQPs Alexandre d Aspremont, Stephen Boyd EE392o, Stanford University Autumn, 2003 Introduction While some special classes of nonconvex problems can be
More informationExact Computation of Global Minima of a Nonconvex Portfolio Optimization Problem
Frontiers In Global Optimization, pp. 1-2 C. A. Floudas and P. M. Pardalos, Editors c 2003 Kluwer Academic Publishers Exact Computation of Global Minima of a Nonconvex Portfolio Optimization Problem Josef
More informationImproved Big-M Reformulation for Generalized Disjunctive Programs
Improved Big-M Reformulation for Generalized Disjunctive Programs Francisco Trespalacios and Ignacio E. Grossmann Department of Chemical Engineering Carnegie Mellon University, Pittsburgh, PA 15213 Author
More informationIntroduction to Integer Linear Programming
Lecture 7/12/2006 p. 1/30 Introduction to Integer Linear Programming Leo Liberti, Ruslan Sadykov LIX, École Polytechnique liberti@lix.polytechnique.fr sadykov@lix.polytechnique.fr Lecture 7/12/2006 p.
More informationLifted Inequalities for 0 1 Mixed-Integer Bilinear Covering Sets
1 2 3 Lifted Inequalities for 0 1 Mixed-Integer Bilinear Covering Sets Kwanghun Chung 1, Jean-Philippe P. Richard 2, Mohit Tawarmalani 3 March 1, 2011 4 5 6 7 8 9 Abstract In this paper, we study 0 1 mixed-integer
More informationSolving nonconvex MINLP by quadratic approximation
Solving nonconvex MINLP by quadratic approximation Stefan Vigerske DFG Research Center MATHEON Mathematics for key technologies 21/11/2008 IMA Hot Topics Workshop: Mixed-Integer Nonlinear Optimization
More informationLecture 8: Column Generation
Lecture 8: Column Generation (3 units) Outline Cutting stock problem Classical IP formulation Set covering formulation Column generation A dual perspective Vehicle routing problem 1 / 33 Cutting stock
More informationStochastic Decision Diagrams
Stochastic Decision Diagrams John Hooker CORS/INFORMS Montréal June 2015 Objective Relaxed decision diagrams provide an generalpurpose method for discrete optimization. When the problem has a dynamic programming
More informationScenario Grouping and Decomposition Algorithms for Chance-constrained Programs
Scenario Grouping and Decomposition Algorithms for Chance-constrained Programs Siqian Shen Dept. of Industrial and Operations Engineering University of Michigan Joint work with Yan Deng (UMich, Google)
More informationOPTIMIZATION. joint course with. Ottimizzazione Discreta and Complementi di R.O. Edoardo Amaldi. DEIB Politecnico di Milano
OPTIMIZATION joint course with Ottimizzazione Discreta and Complementi di R.O. Edoardo Amaldi DEIB Politecnico di Milano edoardo.amaldi@polimi.it Website: http://home.deib.polimi.it/amaldi/opt-15-16.shtml
More informationSome Recent Advances in Mixed-Integer Nonlinear Programming
Some Recent Advances in Mixed-Integer Nonlinear Programming Andreas Wächter IBM T.J. Watson Research Center Yorktown Heights, New York andreasw@us.ibm.com SIAM Conference on Optimization 2008 Boston, MA
More informationIBM Research Report. Branching and Bounds Tightening Techniques for Non-Convex MINLP
RC24620 (W0808-031) August 13, 2008 Mathematics IBM Research Report Branching and Bounds Tightening Techniques for Non-Convex MINLP Pietro Belotti 1, Jon Lee 2, Leo Liberti 3, François Margot 1, Andreas
More informationCyclic short-term scheduling of multiproduct batch plants using continuous-time representation
Computers and Chemical Engineering (00) Cyclic short-term scheduling of multiproduct batch plants using continuous-time representation D. Wu, M. Ierapetritou Department of Chemical and Biochemical Engineering,
More informationIBM Research Report. On Convex Relaxations of Quadrilinear Terms. Sonia Cafieri LIX École Polytechnique Palaiseau France
RC24792 (W94-122 April 3, 29 Mathematics IBM Research Report On Convex Relaxations of Quadrilinear Terms Sonia Cafieri LIX École Polytechnique 91128 Palaiseau France Jon Lee IBM Research Division Thomas
More informationSTRONG VALID INEQUALITIES FOR MIXED-INTEGER NONLINEAR PROGRAMS VIA DISJUNCTIVE PROGRAMMING AND LIFTING
STRONG VALID INEQUALITIES FOR MIXED-INTEGER NONLINEAR PROGRAMS VIA DISJUNCTIVE PROGRAMMING AND LIFTING By KWANGHUN CHUNG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN
More informationAnalyzing the computational impact of individual MINLP solver components
Analyzing the computational impact of individual MINLP solver components Ambros M. Gleixner joint work with Stefan Vigerske Zuse Institute Berlin MATHEON Berlin Mathematical School MINLP 2014, June 4,
More informationImproved quadratic cuts for convex mixed-integer nonlinear programs
Improved quadratic cuts for convex mixed-integer nonlinear programs Lijie Su a,b, Lixin Tang a*, David E. Bernal c, Ignacio E. Grossmann c a Institute of Industrial and Systems Engineering, Northeastern
More informationStrengthened Benders Cuts for Stochastic Integer Programs with Continuous Recourse
Strengthened Benders Cuts for Stochastic Integer Programs with Continuous Recourse Merve Bodur 1, Sanjeeb Dash 2, Otay Günlü 2, and James Luedte 3 1 Department of Mechanical and Industrial Engineering,
More informationA Test Set of Semi-Infinite Programs
A Test Set of Semi-Infinite Programs Alexander Mitsos, revised by Hatim Djelassi Process Systems Engineering (AVT.SVT), RWTH Aachen University, Aachen, Germany February 26, 2016 (first published August
More informationValid Inequalities and Convex Hulls for Multilinear Functions
Electronic Notes in Discrete Mathematics 36 (2010) 805 812 www.elsevier.com/locate/endm Valid Inequalities and Convex Hulls for Multilinear Functions Pietro Belotti Department of Industrial and Systems
More informationFrom structures to heuristics to global solvers
From structures to heuristics to global solvers Timo Berthold Zuse Institute Berlin DFG Research Center MATHEON Mathematics for key technologies OR2013, 04/Sep/13, Rotterdam Outline From structures to
More informationOnline generation via offline selection - Low dimensional linear cuts from QP SDP relaxation -
Online generation via offline selection - Low dimensional linear cuts from QP SDP relaxation - Radu Baltean-Lugojan Ruth Misener Computational Optimisation Group Department of Computing Pierre Bonami Andrea
More informationOptimization of a Nonlinear Workload Balancing Problem
Optimization of a Nonlinear Workload Balancing Problem Stefan Emet Department of Mathematics and Statistics University of Turku Finland Outline of the talk Introduction Some notes on Mathematical Programming
More informationAn RLT Approach for Solving Binary-Constrained Mixed Linear Complementarity Problems
An RLT Approach for Solving Binary-Constrained Mixed Linear Complementarity Problems Miguel F. Anjos Professor and Canada Research Chair Director, Trottier Institute for Energy TAI 2015 Washington, DC,
More informationIntroduction to Integer Programming
Lecture 3/3/2006 p. /27 Introduction to Integer Programming Leo Liberti LIX, École Polytechnique liberti@lix.polytechnique.fr Lecture 3/3/2006 p. 2/27 Contents IP formulations and examples Total unimodularity
More informationAn Integrated Approach to Truss Structure Design
Slide 1 An Integrated Approach to Truss Structure Design J. N. Hooker Tallys Yunes CPAIOR Workshop on Hybrid Methods for Nonlinear Combinatorial Problems Bologna, June 2010 How to Solve Nonlinear Combinatorial
More informationOn mathematical programming with indicator constraints
On mathematical programming with indicator constraints Andrea Lodi joint work with P. Bonami & A. Tramontani (IBM), S. Wiese (Unibo) University of Bologna, Italy École Polytechnique de Montréal, Québec,
More informationComparing Convex Relaxations for Quadratically Constrained Quadratic Programming
Comparing Convex Relaxations for Quadratically Constrained Quadratic Programming Kurt M. Anstreicher Dept. of Management Sciences University of Iowa European Workshop on MINLP, Marseille, April 2010 The
More informationN. L. P. NONLINEAR PROGRAMMING (NLP) deals with optimization models with at least one nonlinear function. NLP. Optimization. Models of following form:
0.1 N. L. P. Katta G. Murty, IOE 611 Lecture slides Introductory Lecture NONLINEAR PROGRAMMING (NLP) deals with optimization models with at least one nonlinear function. NLP does not include everything
More informationOutline. Relaxation. Outline DMP204 SCHEDULING, TIMETABLING AND ROUTING. 1. Lagrangian Relaxation. Lecture 12 Single Machine Models, Column Generation
Outline DMP204 SCHEDULING, TIMETABLING AND ROUTING 1. Lagrangian Relaxation Lecture 12 Single Machine Models, Column Generation 2. Dantzig-Wolfe Decomposition Dantzig-Wolfe Decomposition Delayed Column
More informationNumerical Optimization. Review: Unconstrained Optimization
Numerical Optimization Finding the best feasible solution Edward P. Gatzke Department of Chemical Engineering University of South Carolina Ed Gatzke (USC CHE ) Numerical Optimization ECHE 589, Spring 2011
More informationIndicator Constraints in Mixed-Integer Programming
Indicator Constraints in Mixed-Integer Programming Andrea Lodi University of Bologna, Italy - andrea.lodi@unibo.it Amaya Nogales-Gómez, Universidad de Sevilla, Spain Pietro Belotti, FICO, UK Matteo Fischetti,
More informationthe results from this paper are used in a decomposition scheme for the stochastic service provision problem. Keywords: distributed processing, telecom
Single Node Service Provision with Fixed Charges Shane Dye Department ofmanagement University of Canterbury New Zealand s.dye@mang.canterbury.ac.nz Leen Stougie, Eindhoven University of Technology The
More informationStochastic Dual Dynamic Integer Programming
Stochastic Dual Dynamic Integer Programming Jikai Zou Shabbir Ahmed Xu Andy Sun December 26, 2017 Abstract Multistage stochastic integer programming (MSIP) combines the difficulty of uncertainty, dynamics,
More informationStructured Problems and Algorithms
Integer and quadratic optimization problems Dept. of Engg. and Comp. Sci., Univ. of Cal., Davis Aug. 13, 2010 Table of contents Outline 1 2 3 Benefits of Structured Problems Optimization problems may become
More informationNew Global Algorithms for Quadratic Programming with A Few Negative Eigenvalues Based on Alternative Direction Method and Convex Relaxation
New Global Algorithms for Quadratic Programming with A Few Negative Eigenvalues Based on Alternative Direction Method and Convex Relaxation Hezhi Luo Xiaodi Bai Gino Lim Jiming Peng December 3, 07 Abstract
More informationMachine learning, ALAMO, and constrained regression
Machine learning, ALAMO, and constrained regression Nick Sahinidis Acknowledgments: Alison Cozad, David Miller, Zach Wilson MACHINE LEARNING PROBLEM Build a model of output variables as a function of input
More informationELE539A: Optimization of Communication Systems Lecture 6: Quadratic Programming, Geometric Programming, and Applications
ELE539A: Optimization of Communication Systems Lecture 6: Quadratic Programming, Geometric Programming, and Applications Professor M. Chiang Electrical Engineering Department, Princeton University February
More informationNONLINEAR. (Hillier & Lieberman Introduction to Operations Research, 8 th edition)
NONLINEAR PROGRAMMING (Hillier & Lieberman Introduction to Operations Research, 8 th edition) Nonlinear Programming g Linear programming has a fundamental role in OR. In linear programming all its functions
More informationGeneration and Representation of Piecewise Polyhedral Value Functions
Generation and Representation of Piecewise Polyhedral Value Functions Ted Ralphs 1 Joint work with Menal Güzelsoy 2 and Anahita Hassanzadeh 1 1 COR@L Lab, Department of Industrial and Systems Engineering,
More informationFeasibility Pump for Mixed Integer Nonlinear Programs 1
Feasibility Pump for Mixed Integer Nonlinear Programs 1 Presenter: 1 by Pierre Bonami, Gerard Cornuejols, Andrea Lodi and Francois Margot Mixed Integer Linear or Nonlinear Programs (MILP/MINLP) Optimize
More informationTightening a Discrete Formulation of the Quadratic Assignment Problem
A publication of 1309 CHEMICAL ENGINEERING TRANSACTIONS VOL. 32, 2013 Chief Editors: Sauro Pierucci, Jiří J. Klemeš Copyright 2013, AIDIC Servizi S.r.l., ISBN 978-88-95608-23-5; ISSN 1974-9791 The Italian
More informationCourse Outline. FRTN10 Multivariable Control, Lecture 13. General idea for Lectures Lecture 13 Outline. Example 1 (Doyle Stein, 1979)
Course Outline FRTN Multivariable Control, Lecture Automatic Control LTH, 6 L-L Specifications, models and loop-shaping by hand L6-L8 Limitations on achievable performance L9-L Controller optimization:
More informationSemidefinite Programming Basics and Applications
Semidefinite Programming Basics and Applications Ray Pörn, principal lecturer Åbo Akademi University Novia University of Applied Sciences Content What is semidefinite programming (SDP)? How to represent
More informationNonconvex Quadratic Programming: Return of the Boolean Quadric Polytope
Nonconvex Quadratic Programming: Return of the Boolean Quadric Polytope Kurt M. Anstreicher Dept. of Management Sciences University of Iowa Seminar, Chinese University of Hong Kong, October 2009 We consider
More informationAdvanced linear programming
Advanced linear programming http://www.staff.science.uu.nl/~akker103/alp/ Chapter 10: Integer linear programming models Marjan van den Akker 1 Intro. Marjan van den Akker Master Mathematics TU/e PhD Mathematics
More informationResearch Article A New Global Optimization Algorithm for Solving Generalized Geometric Programming
Mathematical Problems in Engineering Volume 2010, Article ID 346965, 12 pages doi:10.1155/2010/346965 Research Article A New Global Optimization Algorithm for Solving Generalized Geometric Programming
More informationOperations Research Lecture 6: Integer Programming
Operations Research Lecture 6: Integer Programming Notes taken by Kaiquan Xu@Business School, Nanjing University May 12th 2016 1 Integer programming (IP) formulations The integer programming (IP) is the
More informationFixed-charge transportation problems on trees
Fixed-charge transportation problems on trees Gustavo Angulo * Mathieu Van Vyve * gustavo.angulo@uclouvain.be mathieu.vanvyve@uclouvain.be November 23, 2015 Abstract We consider a class of fixed-charge
More informationLecture 23 Branch-and-Bound Algorithm. November 3, 2009
Branch-and-Bound Algorithm November 3, 2009 Outline Lecture 23 Modeling aspect: Either-Or requirement Special ILPs: Totally unimodular matrices Branch-and-Bound Algorithm Underlying idea Terminology Formal
More informationInteger and Combinatorial Optimization: Introduction
Integer and Combinatorial Optimization: Introduction John E. Mitchell Department of Mathematical Sciences RPI, Troy, NY 12180 USA November 2018 Mitchell Introduction 1 / 18 Integer and Combinatorial Optimization
More informationLinear and non-linear programming
Linear and non-linear programming Benjamin Recht March 11, 2005 The Gameplan Constrained Optimization Convexity Duality Applications/Taxonomy 1 Constrained Optimization minimize f(x) subject to g j (x)
More informationComputations with Disjunctive Cuts for Two-Stage Stochastic Mixed 0-1 Integer Programs
Computations with Disjunctive Cuts for Two-Stage Stochastic Mixed 0-1 Integer Programs Lewis Ntaimo and Matthew W. Tanner Department of Industrial and Systems Engineering, Texas A&M University, 3131 TAMU,
More informationADVANCES IN GLOBAL OPTIMIZATION FOR STANDARD, GENERALIZED, AND EXTENDED POOLING PROBLEMS WITH THE (EPA) COMPLEX EMISSIONS MODEL CONSTRAINTS
ADVANCES IN GLOBAL OPTIMIZATION FOR STANDARD, GENERALIZED, AND EXTENDED POOLING PROBLEMS WITH THE (EPA) COMPLEX EMISSIONS MODEL CONSTRAINTS Ruth Misener, Chrysanthos E. Gounaris, and Christodoulos A. Floudas
More informationMultiobjective Mixed-Integer Stackelberg Games
Solving the Multiobjective Mixed-Integer SCOTT DENEGRE TED RALPHS ISE Department COR@L Lab Lehigh University tkralphs@lehigh.edu EURO XXI, Reykjavic, Iceland July 3, 2006 Outline Solving the 1 General
More informationMcCormick Relaxations: Convergence Rate and Extension to Multivariate Outer Function
Introduction Relaxations Rate Multivariate Relaxations Conclusions McCormick Relaxations: Convergence Rate and Extension to Multivariate Outer Function Alexander Mitsos Systemverfahrenstecnik Aachener
More informationA note on : A Superior Representation Method for Piecewise Linear Functions
A note on : A Superior Representation Method for Piecewise Linear Functions Juan Pablo Vielma Business Analytics and Mathematical Sciences Department, IBM T. J. Watson Research Center, Yorktown Heights,
More informationOQNLP: a Scatter Search Multistart Approach for Solving Constrained Non- Linear Global Optimization Problems
OQNLP: a Scatter Search Multistart Approach for Solving Constrained Non- Linear Global Optimization Problems Zsolt Ugray, The University of California at Riverside, MIS Dept. Leon Lasdon, The University
More informationSemidefinite Relaxations for Non-Convex Quadratic Mixed-Integer Programming
Semidefinite Relaxations for Non-Convex Quadratic Mixed-Integer Programming Christoph Buchheim 1 and Angelika Wiegele 2 1 Fakultät für Mathematik, Technische Universität Dortmund christoph.buchheim@tu-dortmund.de
More informationSUNS: A NEW CLASS OF FACET DEFINING STRUCTURES FOR THE NODE PACKING POLYHEDRON CHELSEA NICOLE IRVINE. B.S., Kansas State University, 2012
SUNS: A NEW CLASS OF FACET DEFINING STRUCTURES FOR THE NODE PACKING POLYHEDRON by CHELSEA NICOLE IRVINE B.S., Kansas State University, 01 A THESIS Submitted in partial fulfillment of the requirements for
More informationNetwork Flow Interdiction Models and Algorithms with Resource Synergy Considerations
Network Flow Interdiction Models and Algorithms with Resource Synergy Considerations Brian J. Lunday 1 Hanif D. Sherali 2 1 Department of Mathematical Sciences, United States Military Academy at West Point
More informationA Piecewise Linearization Framework for Retail Shelf Space Management Models
A Piecewise Linearization Framework for Retail Shelf Space Management Models Jens Irion Universitaet Karlsruhe (TH Kaiserstr. 12 76131 Karlsruhe, Germany Faiz Al-Khayyal and Jye-Chyi Lu School of Industrial
More informationThe Fixed Charge Transportation Problem: A Strong Formulation Based On Lagrangian Decomposition and Column Generation
The Fixed Charge Transportation Problem: A Strong Formulation Based On Lagrangian Decomposition and Column Generation Yixin Zhao, Torbjörn Larsson and Department of Mathematics, Linköping University, Sweden
More informationUnit 1A: Computational Complexity
Unit 1A: Computational Complexity Course contents: Computational complexity NP-completeness Algorithmic Paradigms Readings Chapters 3, 4, and 5 Unit 1A 1 O: Upper Bounding Function Def: f(n)= O(g(n)) if
More informationis called an integer programming (IP) problem. model is called a mixed integer programming (MIP)
INTEGER PROGRAMMING Integer Programming g 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 information