Computing

, Volume 45, Issue 2, pp 131–144

Computational aspects of a branch and bound algorithm for quadratic zero-one programming

  • P. M. Pardalos
  • G. P. Rodgers
Article

Abstract

In this paper we describe computational experience in solving unconstrained quadratic zero-one problems using a branch and bound algorithm. The algorithm incorporates dynamic preprocessing techniques for forcing variables and heuristics to obtain good starting points. Computational results and comparisons with previous studies on several hundred test problems with dimensions up to 200 demonstrate the efficiency of our algorithm.

AMS Subject Classifications

65K05 90C30 

Key words

Quadratic 0–1 programming branch and bound computation test problems 

Rechnerische Aspekte eines “Branch and Bound”-Algorithmus zur quadratischen Null-Eins-Programmierung

Zusammenfassung

In dieser Arbeit beschreiben wir rechnerische Erfahrungen bei der Lösung von unbeschränkten quadratischen Null-Eins-Problemen mit einem “Branch and Bound”-Algorithmus. Der Algorithmus erlaubt dynamische Vorbereitungs-Techniken zur Erzwingung ausgewählter Variablen und Heuristiken zur Wahl von guten Startpunkten. Resultate von Berechnungen und Vergleiche mit früheren Arbeiten mit mehreren hundert Testproblemen mit Dimensionen bis 200 zeigen die Effizienz unseres Algorithmus.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Barahona, F., A Solvable Case of Quadratic 0–1 Programming, Discrete Applied Mathematics13 (1986), 23–26.Google Scholar
  2. [2]
    Barahona, F., Jünger, M., Reinelt, G., Experiments in Quadratic 0–1 Programming, Mathematical Programming44 (1989), 127–137.Google Scholar
  3. [3]
    Carter, M. W., The Indefinite Zero-One Quadratic Problem, Discrete Applied Mathematics7 (1984), 23–44.Google Scholar
  4. [4]
    Cooper, M., A Survey of Methods for Pure Nonlinear Integer Programming, Management Science3 (1981), 356–361.Google Scholar
  5. [5]
    Gallo, G., Hammer, P. L., Simeone, B., Quadratic Knapsack Problems, Mathematical Programming12 (1980), 132–149.Google Scholar
  6. [6]
    Gulati, V. P., Gupta, S. K., Mittal, A. K., Unconstrained Quadratic Bivalent Programming Problem, European Journal of Operational Research,15 (1984), 121–125.Google Scholar
  7. [7]
    Hansen, P., Hammer, P. L., Logical relations in quadratic 0–1 programming, Revue Roumaine de Mathematiques Pures et Appliques,20 (1984), 418–427.Google Scholar
  8. [8]
    Hammer, P. L., Hansen, P., Simeone, B., Roof Duality, Complementation, and Persistency in Quadratic 0–1 Optimization, Mathematical Programming,28(2), (1984), 121–155.Google Scholar
  9. [9]
    Hansen, P., Methods of Nonlinear 0–1 Programming, Annals of Discrete Mathematics5 (1979), 53–70.Google Scholar
  10. [10]
    Shi Hui, L., An Improved Enumerative Algorithm for Solving Quadratic Zero-One Programming, European Journal of Operational Research15 (1984), 110–120.Google Scholar
  11. [11]
    Jha, S., Pardalos, P. M., Graph Separation Techniques for Quadratic Zero-One Programming, Technical Report CS-87-39, Computer Science Department, The Pennsylvania State University (1987).Google Scholar
  12. [12]
    Körner, F., An Efficient Branch and Bound Algorithm to Solve the Quadratic Integer Programming Problem, Computing30 (1983), 253–260.Google Scholar
  13. [13]
    Krarup, J., Pruza, P. A., Computer Aided Layout Design, Mathematical Programming Study9 (1978), 75–94.Google Scholar
  14. [14]
    Lewis, P., Goodman, A. S., Miller, J. M., Pseudo-Random Number Generator for the System/360, IBM Systems Journal,8(2), (1969), 300–312.Google Scholar
  15. [15]
    McBride, R. D., Yormark, J. S., An Implicit Enumeration Algorithm for Quadratic Integer Programming, Management Science26(3) (1980), 282–296.Google Scholar
  16. [16]
    Pardalos, P. M., Construction of test problems in quadratic bivalent programming, To appear in ACM TOMS (1990).Google Scholar
  17. [17]
    Pardalos, P. M., Rodgers, G. P., Parallel branch and bound algorithms for unconstrained quadratic zero-one programming. In: Impacts of Recent Computer Advances on Operations Research, (eds R. Sharda et al. 1989), North-Holland, 131–143.Google Scholar
  18. [18]
    Pardalos, P. M., Rodgers, G. P., A branch and bound algorithm for the maximum clique problem, To appear in Mathematical Programming.Google Scholar
  19. [19]
    Pardalos, P. M., Rosen, J. B., Global Constrained Optimization: Algorithms and Applications, Lecture Notes in Computer Sciences268 Berlin-Heidelberg-New York: Springer (1987).Google Scholar
  20. [20]
    Picard, J. C., Ratliff, H. D., Minimum Cuts and Related Problems, Networks5 (1974), 357–370.Google Scholar
  21. [21]
    Rodgers, G. P. Algorithms for Unconstrained Quadratic 0–1 Programming and Related Problems on Contemporary Computer Architectures, Ph.D. Dissertation, The Pennsylvania State University, (1989).Google Scholar
  22. [22]
    Williams, A. C., Quadratic 0–1 Programming Using the Roof Dual with Computational Results, RUTCOR Research Report #8-85, The State University of New Jersey, (1985).Google Scholar

Copyright information

© Springer-Verlag 1990

Authors and Affiliations

  • P. M. Pardalos
    • 1
  • G. P. Rodgers
    • 2
  1. 1.Department of Computer ScienceThe Pennsylvania State UniversityUniversity ParkUSA
  2. 2.General Technology DivisionIBM CorporationBurlingtonUSA

Personalised recommendations