Integration of AI and OR Techniques in Constraint Programming for Combinatorial Optimization Problems: 4th International Conference, CPAIOR 2007, Brussels, Belgium, May 23-26, 2007. Proceedings PDF
Preview Integration of AI and OR Techniques in Constraint Programming for Combinatorial Optimization Problems: 4th International Conference, CPAIOR 2007, Brussels, Belgium, May 23-26, 2007. Proceedings
Lecture Notes in Computer Science 4510 CommencedPublicationin1973 FoundingandFormerSeriesEditors: GerhardGoos,JurisHartmanis,andJanvanLeeuwen EditorialBoard DavidHutchison LancasterUniversity,UK TakeoKanade CarnegieMellonUniversity,Pittsburgh,PA,USA JosefKittler UniversityofSurrey,Guildford,UK JonM.Kleinberg CornellUniversity,Ithaca,NY,USA FriedemannMattern ETHZurich,Switzerland JohnC.Mitchell StanfordUniversity,CA,USA MoniNaor WeizmannInstituteofScience,Rehovot,Israel OscarNierstrasz UniversityofBern,Switzerland C.PanduRangan IndianInstituteofTechnology,Madras,India BernhardSteffen UniversityofDortmund,Germany MadhuSudan MassachusettsInstituteofTechnology,MA,USA DemetriTerzopoulos UniversityofCalifornia,LosAngeles,CA,USA DougTygar UniversityofCalifornia,Berkeley,CA,USA MosheY.Vardi RiceUniversity,Houston,TX,USA GerhardWeikum Max-PlanckInstituteofComputerScience,Saarbruecken,Germany PascalVan Hentenryck LaurenceWolsey (Eds.) Integration of AI and OR Techniques in Constraint Programming for Combinatorial Optimization Problems 4th International Conference, CPAIOR 2007 Brussels, Belgium, May 23-26, 2007 Proceedings 1 3 VolumeEditors PascalVanHentenryck BrownUniversity Box1910,Providence,RI02912,USA E-mail:[email protected] LaurenceWolsey UniversitécatholiquedeLouvain 34,VoieduRomanPays,1348Louvain-la-Neuve,Belgium E-mail:[email protected] LibraryofCongressControlNumber:2007926618 CRSubjectClassification(1998):G.1.6,G.1,G.2.1,F.2.2,I.2,J.1 LNCSSublibrary:SL1–TheoreticalComputerScienceandGeneralIssues ISSN 0302-9743 ISBN-10 3-540-72396-XSpringerBerlinHeidelbergNewYork ISBN-13 978-3-540-72396-7SpringerBerlinHeidelbergNewYork Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerialis concerned,specificallytherightsoftranslation,reprinting,re-useofillustrations,recitation,broadcasting, reproductiononmicrofilmsorinanyotherway,andstorageindatabanks.Duplicationofthispublication orpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLawofSeptember9,1965, initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer.Violationsareliable toprosecutionundertheGermanCopyrightLaw. SpringerisapartofSpringerScience+BusinessMedia springer.com ©Springer-VerlagBerlinHeidelberg2007 PrintedinGermany Typesetting:Camera-readybyauthor,dataconversionbyScientificPublishingServices,Chennai,India Printedonacid-freepaper SPIN:12061068 06/3142 543210 Preface This volume contains the papers presented at CP-AI-OR 2007: The Fourth In- ternational Conference on Integration of Artificial Intelligence, Constraint Pro- gramming,andOperationsResearchTechniquesforCombinatorialOptimization Problems held May 23–26, 2007 in Brussels, Belgium. More information about the conference can be found at the Web site: http://www.cs.brown.edu/sites/cpaior07/Welcome.html. There were 80 submissions and each submission was reviewed by at least three Program Committee members. After careful consideration and discussion, the committee decided to accept 28 papers (one of which was withdrawn just be- fore the publication of the proceedings). The papers submitted this year were of high quality and representative of a vibrant, multi-disciplinary community at the intersection of artificial intelligence, constraint programming, and opera- tions research.The programalso included three invited talks and two tutorials. We were extremely pleased that Claude Le Pape, George Nemhauser, and Bart Selman accepted our invitation to be invited speakers at the conference. Simi- larly, we were very fortunate to attract Thierry Benoist and John Chinneck to givetutorialsatCP-AI-OR2007.TheconferencewasalsoprecededbyaMaster Class on constraint-basedscheduling organized by Amedeo Cesta. We would like to thank the ProgramCommittee members who workedhard to produce high-quality reviews for the papers under tight deadlines, as well as the reviewers involved in the paper selection. We also would like to acknowl- edge the contributions of Laurent Michel (Publicity Chair), Barry O’Sullivan (Sponsorship Chair), Susanne Heipcke and Michael Juenger (Academic and In- dustrialLiaisonChairs),andEtienneLoute(ConferenceCo-chair)tothesuccess of CP-AI-OR 2007. It is also a great pleasure to thank Fabienne Henry for her tremendoushelp inorganizingthe conference.The submissions,reviews,discus- sions,andthe preparationofthe proceedingswereallhandledbytheEasyChair system. Finally, we would also like to thank the sponsors of the conference: TheAssociationforConstraintProgramming,theCorkConstraintComputation Centre(Ireland),ILOGS.A.(France),theIntelligentInformationSystemsInsti- tuteatCornellUniversity(USA),theEuropeanCommission(throughtheMarie CurieResearchTrainingNetwork:ADONET),theFondsNationaldeRecherche Scientifique in Belgium, the Belgian Science Policy, ORBEL (Master Class), as wellasBrownUniversity(USA), the Facult´esSt. Louis(Brussels,Belgium)and the Universit´e catholique de Louvain (Louvain-la-Neuve, Belgium). March 2007 Pascal Van Hentenryck Laurence Wolsey Conference Organization Program Chairs PascalVan Hentenryck Laurence Wolsey Local Organization Etienne Loute, Facult´es Universitaires de Saint-Louis, Brussels, Belgium LaurenceWolsey,Universit´ecatholiquedeLouvain,Louvain-La-Neuve,Belgium Program Committee Philippe Baptiste Chris Beck Frederic Benhamou Mats Carlsson Amedeo Cesta John Chinneck Andrew Davenport Rina Dechter Yves Deville Hani El Sakkout Bernard Gendron Carla Gomes John Hooker Stefan Karish Andrea Lodi Laurent Michel Michela Milano Yehuda Naveh Barry O’Sullivan Jean-Francois Puget Jean-Charles R´egin Andrea Roli Louis-Martin Rousseau Michel Rueher Martin Savelsbergh Meinolf Sellmann David Shmoys Helmut Simonis VIII Organization Barbara Smith Stephen Smith Peter Stuckey Michael Trick Mark Wallace Willem-Jan van Hoeve External Reviewers Ionut Aron Jean-No¨el Monette Christian Artigues Bertrand Neveu Gregory Barlow Angelo Oddi Marco Benedetti Gilles Pesant John Betts Federico Pecora Eyal Bin Lam Phuong Pierre Bonami Nicola Policella Sebastian Brand Jakob Puchinger Marie-Claude Cˆot´e Luis Queseda Emilie Danna Claude-Guy Quimper Sophie Demassey Yossi Richter Bistra Dilkina Ashish Sabharwal Nizar El Hachemi Frederic Saubion Dominique Feillet Pierre Schaus Felix Geller Stefano Smriglio Carmen Gervet Christine Solnon Vibhav Gogate Peter Stuckey Tarik Hadzic Daria Terekhov Jin-Kao Hao Andrea Tramontani Shlomo Hoory Gilles Trombettoni Ayoub Insea Correa Charlotte Truchet Manuel Iori Jeremie Vautard Kalev Kask Petr Vilim Serge Kruk Toby Walsh Yahia Lebbah St´ephane Zampelli Bernd Meyer Terry Zimmerman Claude Michel Table of Contents Minimum Cardinality Matrix Decomposition into Consecutive-Ones Matrices: CP and IP Approaches................................... 1 Davaatseren Baatar, Natashia Boland, Sebastian Brand, and Peter J. Stuckey Connections in Networks: Hardness of Feasibility Versus Optimality .... 16 Jon Conrad, Carla P. Gomes, Willem-Jan van Hoeve, Ashish Sabharwal, and Jordan Suter Modeling the Regular Constraint with Integer Programming........... 29 Marie-Claude Coˆt´e, Bernard Gendron, and Louis-Martin Rousseau Hybrid Local Search for Constrained Financial Portfolio Selection Problems ....................................................... 44 Luca Di Gaspero, Giacomo di Tollo, Andrea Roli, and Andrea Schaerf The “Not-Too-HeavySpanning Tree” Constraint..................... 59 Gr´egoire Dooms and Irit Katriel Eliminating Redundant Clauses in SAT Instances .................... 71 Olivier Fourdrinoy, E´ric Gr´egoire, Bertrand Mazure, and Lakhdar Sa¨ıs Cost-Bounded Binary Decision Diagrams for 0–1 Programming ........ 84 Tarik Hadˇzi´c and J.N. Hooker YIELDS: A Yet Improved Limited Discrepancy Search for CSPs ....... 99 Wafa Karoui, Marie-Jos´e Huguet, Pierre Lopez, and Wady Naanaa A Global Constraint for Total Weighted Completion Time ............ 112 Andr´as Kov´acs and J. Christopher Beck Computing Tight Time Windows for RCPSPWET with the Primal-Dual Method ............................................. 127 Andr´as K´eri and Tama´s Kis Necessary Condition for Path Partitioning Constraints................ 141 Nicolas Beldiceanu and Xavier Lorca A Constraint Programming Approach to the Hospitals / Residents Problem ........................................................ 155 David F. Manlove, Gregg O’Malley, Patrick Prosser, and Chris Unsworth Best-First AND/OR Search for 0/1 Integer Programming ............. 171 Radu Marinescu and Rina Dechter A Position-BasedPropagatorfor the Open-Shop Problem ............. 186 Jean-No¨el Monette, Yves Deville, and Pierre Dupont X Table of Contents Directional Interchangeability for Enhancing CSP Solving............. 200 Wady Naanaa A Continuous Multi-resources cumulative Constraint with Positive-NegativeResource Consumption-Production ................. 214 Nicolas Beldiceanu and Emmanuel Poder Replenishment Planning for Stochastic Inventory Systems with Shortage Cost ................................................... 229 Roberto Rossi, S. Armagan Tarim, Brahim Hnich, and Steven Prestwich Preprocessing Expression-Based Constraint Satisfaction Problems for Stochastic Local Search........................................... 244 Sivan Sabato and Yehuda Naveh The Deviation Constraint ......................................... 260 Pierre Schaus, Yves Deville, Pierre Dupont, and Jean-Charles R´egin The LinearProgrammingPolytopeofBinaryConstraintProblemswith Bounded Tree-Width ............................................. 275 Meinolf Sellmann, Luc Mercier, and Daniel H. Leventhal On Boolean Functions Encodable as a Single Linear Pseudo-Boolean Constraint ...................................................... 288 Jan-Georg Smaus Solving a Stochastic Queueing Control Problem with Constraint Programming.................................................... 303 Daria Terekhov and J. Christopher Beck Constrained Clustering Via Concavity Cuts ......................... 318 Yu Xia Bender’s Cuts Guided Large Neighborhood Search for the Traveling Umpire Problem ................................................. 332 Michael A. Trick and Hakan Yildiz A Large Neighborhood Search Heuristic for Graph Coloring ........... 346 Michael A. Trick and Hakan Yildiz Generalizations of the Global Cardinality Constraint for Hierarchical Resources....................................................... 361 Alessandro Zanarini and Gilles Pesant A Column Generation Based Destructive Lower Bound for Resource Constrained Project Scheduling Problems ........................... 376 J. Marjan van den Akker, Guido Diepen, and J.A. Hoogeveen Author Index.................................................. 391 Minimum Cardinality Matrix Decomposition into Consecutive-Ones Matrices: CP and IP Approaches Davaatseren Baatar1, Natashia Boland1, Sebastian Brand2, and Peter J. Stuckey2 1 Department of Mathematics, Universityof Melbourne, Australia 2 NICTA Victoria Research Lab, Department of Comp. Sci. and Soft. Eng. Universityof Melbourne, Australia Abstract. We consider the problem of decomposing an integer ma- trix into a positively weighted sum of binary matrices that have the consecutive-ones property. This problem is well-known and of practical relevance. It has an important application in cancer radiation therapy treatment planning: the sequencing of multileaf collimators to deliver a given radiation intensity matrix, representing (a component of) the treatment plan. Two criteria characterise the efficacy of a decomposition: the beam- on time (length of time the radiation source is switched on during the treatment),andthecardinality (thenumberofmachineset-upsrequired to deliver theplanned treatment). Minimisingtheformerisknowntobeeasy.Howeverfindingadecom- position of minimal cardinality is NP-hard. Progress so far has largely been restricted to heuristic algorithms, mostly using linear program- ming, integer programming and combinatorial enumerative methods as the solving technologies. We present a novel model, with correspond- ing constraint programming and integerprogramming formulations. We comparethesecomputationallywithpreviousformulations,andweshow that constraint programming performs very well bycomparison. 1 Introduction The problem of decomposing an integer matrix into a weighted sum of binary matrices has received much attention in recent years, largely due to its applica- tion in radiation treatment for cancer. Intensity-modulated radiation therapy (IMRT) has been increasingly used for the treatmentofa varietyofcancers[17]. This treatmentapproachemploys two devices that allow higher doses of radiation to be administered to the tumour, while decreasing the exposure of sensitive organs (Fig. 1). The first is that the sourceof radiationcanbe rotatedabout the body of the patient:by positioning the tumour at a “focal point”, and aiming the radiation beam at this point from various angles, the tumour receives a high dose from all angles, while the surrounding tissue only gets high exposure from some angles. P.VanHentenryckandL.Wolsey(Eds.):CPAIOR2007,LNCS4510,pp.1–15,2007. (cid:2)c Springer-VerlagBerlinHeidelberg2007 2 D. Baatar et al. Target volume 2Dtransaxialsliceofthebody Beamhead Organsatrisk Fig.1. Intensity-modulatedradiotherapy The second is more subtle, and involves repeated exposures from the same angle,wheretheuniform-intensityrectangularfieldofradiationdeliveredbythe radiationsourceis“shaped”inadifferentwayforeachexposure,andeachexpo- surecanbeforadifferentlengthoftime.Thisprocessbuildsupacomplexprofile ofreceivedradiationinthepatient’sbody,effectivelyconvertingtheuniformra- diationfielddeliveredbythemachinetoanintensity-modulatedfield.Thelatter isusuallydescribedbydiscretisingthe2-dimensionalrectangularfield,andspec- ifying a radiation intensity level in each discrete element, representing the total length of time for which that element should be exposed to radiation. A treatment plan for a single IMRT treatment session with a patient thus typicallyconsistsofasetofangles,togetherwithamatrixforeachangle,known as the intensity matrix, which represents the modulated field to be delivered at that angle. Typically the intensity is scaled so that the entries in the intensity matrixareinteger.Indeed,they areusuallyquite smallintegers.Findingagood treatmentplanisachallengingprobleminitsownright,andhasbeenthesubject ofagreatdealofresearch.Werecommendthereaderrefertothepapers[13,9,15] and references therein. In this paper, we assume a treatment plan is given, and focus on the deliveryofthemodulatedfield(intensitymatrix)atagivenangle.IMRT canbe
Description:The list of books you might like
