Polynomial-size formulations and relaxations for the quadratic multiple knapsack problem

被引:6
|
作者
Galli, Laura [1 ]
Martello, Silvano [2 ]
Rey, Carlos [2 ]
Toth, Paolo [2 ]
机构
[1] Univ Pisa, Dipartimento Informat, Largo B Pontecorvo 3, I-56127 Pisa, Italy
[2] Univ Bologna, DEI Guglielmo Marconi, Viale Risorgimento 2, I-40136 Bologna, Italy
关键词
Combinatorial optimization; Quadratic multiple knapsack; Binary quadratic programming; Lagrangian relaxation; Reformulation linearization technique;
D O I
10.1016/j.ejor.2020.10.047
中图分类号
C93 [管理学];
学科分类号
12 ; 1201 ; 1202 ; 120202 ;
摘要
The Quadratic Multiple Knapsack Problem generalizes, simultaneously, two well-known combinatorial optimization problems that have been intensively studied in the literature: the (single) Quadratic Knapsack Problem and the Multiple Knapsack Problem. The only exact algorithm for its solution uses a formulation based on an exponential-size number of variables, that is solved via a Branch-and-Price algorithm. This work studies polynomial-size formulations and upper bounds. We derive linear models from classical reformulations of 0-1 quadratic programs and analyze theoretical properties and dominances among them. We define surrogate and Lagrangian relaxations, and we compare the effectiveness of the Lagrangian relaxation when applied to a quadratic formulation and to a Level 1 reformulation linearization that leads to a decomposable structure. The proposed methods are evaluated through extensive computational experiments. (C) 2020 Elsevier B.V. All rights reserved.
引用
收藏
页码:871 / 882
页数:12
相关论文
共 50 条