computer science publication server: No conditions. Results ordered -Date Deposited. 2021-04-22T23:48:09ZEPrintshttp://e-archive.informatik.uni-koeln.de/images/sitelogo.pnghttp://e-archive.informatik.uni-koeln.de/2021-03-04T14:01:01Z2021-03-05T11:17:58Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/951This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/9512021-03-04T14:01:01ZAn Inexact Quadratic Simplex Algorithm Tailored to Zero-One PolyhedraA simplex algorithm is presented that is tailored to quadratic optimization over the vertices of zero-one polyhedra. While
the algorithm can guarantee only local optima in general, computational experiments show a good performance on
standard benchmark libraries.Sven Mallach2020-10-21T11:52:44Z2021-03-04T14:02:57Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/950This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/9502020-10-21T11:52:44ZExact Facetial Odd-Cycle Separation for Maximum Cut and Binary Quadratic OptimizationThe exact solution of the NP-hard Maximum Cut Problem is important in many applications across, e.g., Physics, Chemistry, Neuroscience, and Circuit Layout – which is also due to its equivalence to the unconstrained Binary Quadratic Optimization Problem. Leading solution methods are based on linear or semidefinite programming, and require the separation of the so-called odd-cycle inequalities. In their groundbreaking work, F. Barahona and A.R. Mahjoub have given an informal description of a polynomial-time algorithm for
this problem. As pointed out recently, however, additional effort is necessary to guarantee that the obtained
inequalities correspond to facets of the cut polytope. In this paper, we shed more light on a so enhanced
separation procedure and investigate experimentally how it performs in comparison to an ideal setting where one could even employ the sparsest, most violated, or geometrically most promising facet-defining odd-cycle inequalities.Michael JüngerJünger, MichaelSven MallachMallach, Sven2019-08-29T08:45:38Z2019-08-29T08:45:38Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/943This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/9432019-08-29T08:45:38ZOdd-Cycle Separation for Maximum Cut and Binary Quadratic OptimizationSolving the NP-hard Maximum Cut or Binary Quadratic Optimization Problem to optimality is important in many applications including Physics, Chemistry, Neuroscience, and Circuit Layout. The leading approaches based on linear/semidefinite programming require the separation of so-called odd-cycle inequalities for solving relaxations within their associated branch-and-cut frameworks. In
their groundbreaking work, F. Barahona and A.R. Mahjoub have given an informal description of a polynomial-time separation procedure for the odd-cycle inequalities. Since then, the odd-cycle separation problem has broadly been considered solved. However, as we reveal, a straightforward
implementation is likely to generate inequalities that are not facet-defining and have further undesired properties. Here, we present a more detailed analysis, along with enhancements to overcome the associated issues efficiently. In a corresponding experimental study, it turns out that these are worthwhile, and may speed up the solution process significantly.Michael JüngerJünger, MichaelSven MallachMallach, Sven2014-07-04T09:03:15Z2014-07-04T09:03:15Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/712This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/7122014-07-04T09:03:15ZInput-Output-Analyse bei unternehmensgrößenspezifischen FragestellungenAchim BachemBachem, AchimLothar ButzButz, LotharMartin GrötschelGrötschel, MartinRainer SchraderSchrader, Rainer2011-03-14T00:00:00Z2012-01-09T11:59:51Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/602This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/6022011-03-14T00:00:00ZSolving Two-Stage Stochastic Steiner Tree Problems by Two-Stage Branch-and-CutWe consider the Steiner tree problem under a two-stage stochastic model with recourse and finitely many scenarios. In this prob- lem, edges are purchased in the first stage when only probabilistic infor- mation on the set of terminals and the future edge costs is known. In the second stage, one of the given scenarios is realized and additional edges are puchased in order to interconnect the set of (now known) ter- minals. The goal is to decide on the set of edges to be purchased in the first stage while minimizing the overall expected cost of the solution. We provide a new semi-directed cut-set based integer programming formula- tion, which is stronger than the previously known undirected model. We suggest a two-stage branch-and-cut (B&C) approach in which L-shaped and integer-L-shaped cuts are generated. In our computational study we compare the performance of two variants of our algorithm with that of a B&C algorithm for the extensive form of the deterministic equiva- lent (EF). We show that, as the number of scenarios increases, the new approach significantly outperforms the (EF) approach.Immanuel BomzeBomze, ImmanuelMarkus ChimaniChimani, MarkusMichael JüngerJünger, MichaelIvana LjubicLjubic, IvanaPetra MutzelMutzel, PetraBernd ZeyZey, Bernd2001-04-02T00:00:00Z2011-12-19T09:45:04Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/198This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/1982001-04-02T00:00:00ZWarehouse Optimization: Packing-algorithms for goods with complex geometriesIn the organization of a warehouse one has to optimize different work routines such as shortest-path-problems, three-dimensional packing problems and assignment/placement problems. Within the framework of the implementation of an integrated warehouse optimization system we started to deal with the packing problems and their integration into algorithms for vehicle routing problems. In this talk we present a practical packing problem arising from the packing of bathtubs and other goods with complex geometry on pallets. We model this problem both as a packing problem and as a vehicle routing problem and discuss suitable algorithms and present some computational results.Jörg SchepersSchepers, JörgMichael WottawaWottawa, Michael2001-04-02T00:00:00Z2011-10-24T08:51:58Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/67This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/672001-04-02T00:00:00ZNote on the Convergence of Simulated Annealing AlgorithmsGeneralizing the results of the first author and Rainer Schrader [Inf. Process. Lett. 27, 189-194 (1988)] a short inductive proof is given that shows that the stationary distributions of a simulated annealing algorithm converge to a distribution where nonoptimal elements are generated with probability zero, provided that the ''weak reversibility condition'' of B. Hajek [Math. Oper. Res. 13, No. 2, 311-329 (1988)] holds.Ulrich FaigleFaigle, UlrichWalter KernKern, Walter2001-04-02T00:00:00Z2011-12-19T09:45:30Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/119This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/1192001-04-02T00:00:00ZParallelisierung von Heuristiken für große Traveling Salesman ProblemeWir werden parallele Versionen von TSP-Heuristiken vorstellen, die in der Praxis sowohl durch ihr Laufzeitverhalten als auch durch die Qualität der von ihnen generierten Lösungen überzeugen: eine parallele Insert-Heuristik, Lin-Kernighan-Verbesserungsheuristik sowie ein Clusterungsverfahren. Die beschriebenen Techniken der Parallelisierung (Data-Parallelism, Job-Parallelism, geometrische Aufteilung) lassen sich analog auch auf andere Algorithmen anwenden.Achim BachemBachem, AchimMichael WottawaWottawa, Michael2001-04-02T00:00:00Z2012-01-19T11:19:51Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/150This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/1502001-04-02T00:00:00ZAn efficient parallel cluster-heuristic for large TSPsWe describe an improved clustering heuristic for the Eucledian Traveling Salesman Problem and its parallelization for a distributed memory machine. A geometric decomposition is used for the clustering-stage and special emphasis has been put on the computation of the global tour through the clusters. The heuristic solves problem instances up to 33,000 cities in a few minutes on the parallel machine, while the obtained tour is only a few percent longer than a tour generated by the sequential Lin-Kernighan-algorithm.Achim BachemBachem, AchimBarthel SteckemetzSteckemetz, BarthelMichael WottawaWottawa, Michael2001-04-02T00:00:00Z2011-10-21T13:22:46Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/83This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/832001-04-02T00:00:00ZSome Convergence Results for Probabilistic Tabu SearchDuring recent years, much work has gone into the exploration of general fundamental principles underlying local search strategies for combinatorial optimization. Many of these strategies can be subsumed under the general framework of tabu search, which introduces mechanisms of guidance and control based on flexible memory processes, broadening the range of strategic possibilities beyond those incorporated in memoryless search heuristics such as simulated annealing. We consider some examples of such memory based strategies for modifying both the generation and acceptance probabilities and investigate their impact on convergence results. It turns out that several tabu search ideas can be subjected to mathematical analysis similar to those applied to simulated annealing, making it possible to establish corresponding convergence properties based on a broader foundation.Ulrich FaigleFaigle, UlrichWalter KernKern, Walter2001-04-02T00:00:00Z2011-12-19T09:46:38Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/216This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/2162001-04-02T00:00:00ZPACKLIB: Ein ASCII-Datenformat für PackungsproblemeÄhnlich der TSPLIB von Gerhard Reinelt für Traveling-Salesman-Probleme definieren wir ein Datenformat für Packungsprobleme und stellen mit diesem eine Sammlung von Problemen aus der Literatur vor. Das Format benutzt ausschließlich ASCII-Zeichen und ist somit auf beliebige Computersysteme übertragbar. Die aus der Literatur bekannten Probleminstanzen werden in diesem neuen PACKLIB-Format breitgestellt.Michael WottawaWottawa, Michael2001-04-02T00:00:00Z2012-01-19T13:46:50Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/134This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/1342001-04-02T00:00:00ZComputational Experience with General Equilibrium ProblemsWe report on computational experience with an implementation of three algorithms for the general economic equilibrium problem. As a result we get that the projection algorithm for variational inequalities increases the size of solvable models by a factor of 5-10 in comparison with the classical homotopy method. As a third approach we implemented a simulated annealing heuristic which might be suitable to estimate equilibria for very large models.Achim BachemBachem, AchimWinfried HochstättlerHochstättler, WinfriedBarthel SteckemetzSteckemetz, BarthelAndreas VolmerVolmer, Andreas2001-04-02T00:00:00Z2012-01-19T11:32:42Zhttp://e-archive.informatik.uni-koeln.de/id/eprint/131This item is in the repository with the URL: http://e-archive.informatik.uni-koeln.de/id/eprint/1312001-04-02T00:00:00ZProgramming, Porting and Performance Tests on a 1024-processor TransputerclusterThis paper describes three different porting activities on three parallel applications performed on the new Parsytec GCel-3/1024 Transputer System in Cologne. The communication features of the machine are reported and a brief summary is given on the performance of the ported programs. The three applications differ extremely in computing requirements and parallelization approach: numerical standard algorithms, a TSP-heuristic and the simulation of traffic congestions are examined. In all cases the porting was done in a few days and the gain in performance was very satisfactory even for thousand processors.Achim BachemBachem, AchimTheodor MeisMeis, TheodorKai NagelNagel, KaiMarisa RiemeyerRiemeyer, MarisaMichael WottawaWottawa, Michael