A fast and effective breakpoints heuristic algorithm for the quadratic knapsack problem

被引:0
作者
Hochbaum, D. S. [1 ]
Baumann, P. [2 ]
Goldschmidt, O. [3 ]
Zhang, Y. [1 ]
机构
[1] Univ Calif Berkeley, Ind Engn & Operat Res Dept, Berkeley, CA 94720 USA
[2] Univ Bern, Dept Business Adm, Engehaldenstr 4, CH-3012 Bern, Switzerland
[3] Riverside Cty Off Educ, Riverside, CA 92501 USA
关键词
Combinatorial optimization; Quadratic Knapsack problem; Breakpoints algorithm; Parametric cut; Greedy algorithm; MAXIMUM DIVERSITY; TABU-SEARCH; DISPERSION; GRASP; PSEUDOFLOW; EVOLUTION;
D O I
10.1016/j.ejor.2024.12.019
中图分类号
C93 [管理学];
学科分类号
12 ; 1201 ; 1202 ; 120202 ;
摘要
The Quadratic Knapsack Problem (QKP) involves selecting a subset of elements that maximizes the sum of pairwise and singleton utilities without exceeding a given budget. The pairwise utilities are nonnegative, the singleton utilities maybe positive, negative, or zero, and the node costs are nonnegative. We introduce a Breakpoints Algorithm for QKP, named QKBP, which is based on a technique proposed in Hochbaum (2009) for efficiently generating the concave envelope of the solutions to the relaxation of the problem for all values of the budget. Our approach utilizes the fact that breakpoints in the concave envelopes are optimal solutions for their respective budgets. For budgets between breakpoints, a fast greedy heuristic derives high-quality solutions from the optimal solutions of adjacent breakpoints. The QKBP algorithm is a heuristic which is highly scalable due to an efficient parametric cut procedure used to generate the concave envelope. This efficiency is further improved by a newly developed compact problem formulation. Our extensive computational study on both existing and new benchmark instances, with up to 10,000 elements, shows that while some leading algorithms perform well on a few instances, QKBP consistently delivers high-quality solutions regardless of instance size, density, or budget. Moreover, QKBP achieves these results insignificantly faster running times than all leading algorithms. The source code of the QKBP algorithm, the benchmark instances, and the detailed results are publicly available on GitHub.
引用
收藏
页码:425 / 440
页数:16
相关论文
共 54 条
[1]  
Agca S, 2000, NAV RES LOG, V47, P97, DOI 10.1002/(SICI)1520-6750(200003)47:2<97::AID-NAV2>3.0.CO
[2]  
2-2
[3]   Comparing local search metaheuristics for the maximum diversity problem [J].
Aringhieri, R. ;
Cordone, R. .
JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY, 2011, 62 (02) :266-280
[4]   Tabu Search versus GRASP for the maximum diversity problem [J].
Aringhieri, Roberto ;
Cordone, Roberto ;
Melzani, Yari .
4OR-A QUARTERLY JOURNAL OF OPERATIONS RESEARCH, 2008, 6 (01) :45-60
[5]   SELECTION PROBLEM [J].
BALINSKI, ML .
MANAGEMENT SCIENCE SERIES A-THEORY, 1970, 17 (03) :230-231
[6]   Using a mixed integer programming tool for solving the 0-1 quadratic knapsack problem [J].
Billionnet, A ;
Soutif, E .
INFORMS JOURNAL ON COMPUTING, 2004, 16 (02) :188-197
[7]   Knapsack problems - An overview of recent advances. Part II: Multiple, multidimensional, and quadratic knapsack problems [J].
Cacchiani, Valentina ;
Iori, Manuel ;
Locatelli, Alberto ;
Martello, Silvano .
COMPUTERS & OPERATIONS RESEARCH, 2022, 143
[8]   Exact solution of the Quadratic Knapsack Problem [J].
Caprara, A ;
Pisinger, D ;
Toth, P .
INFORMS JOURNAL ON COMPUTING, 1999, 11 (02) :125-137
[9]   A Computational Study of the Pseudoflow and Push-Relabel Algorithms for the Maximum Flow Problem [J].
Chandran, Bala G. ;
Hochbaum, Dorit S. .
OPERATIONS RESEARCH, 2009, 57 (02) :358-376
[10]   An iterated "hyperplane exploration" approach for the quadratic knapsack problem [J].
Chen, Yuning ;
Hao, Jin-Kao .
COMPUTERS & OPERATIONS RESEARCH, 2017, 77 :226-239