• Article
  • | |
  • Metrics
  • |
  • Reference [15]
  • |
  • Related [20]
  • |
  • Cited by
  • | |
  • Comments
    Abstract:

    Branch-and-Bound (B&B) algorithm is one of the fundamental methods for combinatorial optimization problems. The running time is dominated by the data structure used to implement B&B algorithm for the given problem and the related branching strategy. In this paper, the data structure called Cubeheap and the related algorithms (INSERT and DELETE) are discussed. The lower bound Ω(m+hlogh) of the running time for general B&B algorithm is proposed by constructing the mapping between the B&B procedure and the sorting procedure, where m is the number of the evaluated nodes and h is the number of the expanded nodes. According to the lower bound, Cubeheap is the near-optimal data structure to implement the general B&B algorithm. The experimental results for a concrete combinatorial optimization problem, Job-assignment, are obtained by running the B&B algorithm with the Cubeheap. The method to improve the balance of the Cubeheap is also proposed.

    Reference
    [1]Kumar V, Kanal L. A general branch and bound formulation for under-standing and synthesizing And/Or tree search procedures. Artificial Intelligence, 1983,21(1-2):179~198
    [2]Nau D S, Kumar V, Kanal L. General branch and bound and it's relation to A* and AO*. Artificial Intelligence, 1984,23(1):29~58
    [3]Nguyen V T. Global optimization techniques for solving the general quadratic integer programming problem. Computational Optimization and Applications, 1998,10(2):149~163
    [4]Washburn A R. Branch and bound methods for a search problem. Naval Research Logostics, 1998,45(3):243~257
    [5]Wah B W, Ma E Y W. Manip——a multicomputer architecture for solving combinatorial extremum search problems. IEEE Transactions on Computer, 1984,C-33(5):377~390
    [6]Yu C F, Wah B W. Efficient branch-and-bound algorithms on a two-level memory system. IEEE Transactions on Software Engineering, 1988,14(9):1342~1356
    [7]Quinn M. Analysis and implementation of branch-and-bound algorithms on hypercube multicomputer. IEEE Transactions on Computer, 1990,39(3):384~387
    [8]Kaklamanis C, Persiano G. Branch-and-Bound and backtrack search on mesh-connected array of processors. Mathematics Systems Theory, 1994,27(5):471~489
    [9]Yang M K, Das C R. Evaluation of a parallel branch-and-bound algorithm on a class of multiprocessors. IEEE Transactions on Parallel and Distributed Systems, 1994,5(1):74~86
    [10]Gendron B, Crainic T G. Parallel branch-and-bound algorithms: survey and synthesis. Operations Research, 1994,42(6):1042~1066
    [11] Wu Ji-gang. General data structure and algorithms for branch and bound search. Information Intelligence and Systems, 1996 IEEE International Conference on Systems, Man and Cybernetics. Beijing: International Academic Publishers, 1996. 1586~1588
    [12]Aho A V, Hopcroft J E, Ullman J D. The Design and Analysis of Computer Algorithm. Reading, MA: Addison-Wesley Publishing Company, 1974. 87~92, 172~176
    [13]Wu Ji-gang, Zhu Hong. The least basic operations on heap and improved heapsort. Journal of Computer Science and Technology, 1994,9(3):261~266
    [14]Chen Guo-liang. Design and Analysis of Parallel Algorithms. Beijing: High Educational Press, 1994 (陈国良.并行算法设计与分析.北京:高等教育出版社,1994)
    [15]Gonnet C H, Munro J I. Heaps on heaps. SIAM Journal of Computing, 1986,15(4):964~971
    Cited by
    Comments
    Comments
    分享到微博
    Submit
Get Citation

武继刚,陈国良,吴明.立体堆与分枝界限算法.软件学报,2000,11(7):984-989

Copy
Share
Article Metrics
  • Abstract:3833
  • PDF: 5035
  • HTML: 0
  • Cited by: 0
History
  • Received:April 01,1999
  • Revised:June 21,1999
You are the first2038616Visitors
Copyright: Institute of Software, Chinese Academy of Sciences Beijing ICP No. 05046678-4
Address:4# South Fourth Street, Zhong Guan Cun, Beijing 100190,Postal Code:100190
Phone:010-62562563 Fax:010-62562533 Email:jos@iscas.ac.cn
Technical Support:Beijing Qinyun Technology Development Co., Ltd.

Beijing Public Network Security No. 11040202500063