Bilinear Assignment Problem: Large Neighborhoods and Experimental Analysis of Algorithms

被引:2
|
作者
Sokol, Vladyslav [1 ]
Custic, Ante [2 ]
Punnen, Abraham P. [2 ,3 ]
Bhattacharya, Binay [1 ]
机构
[1] Simon Fraser Univ, Sch Comp Sci, Surrey, BC V3T 0A3, Canada
[2] Simon Fraser Univ, Dept Math, Surrey, BC V3T 0A3, Canada
[3] Northwestern Polytech Univ, Sch Management, Xian 710072, Peoples R China
基金
加拿大自然科学与工程研究理事会;
关键词
nonlinear assignment problems; average solution value; domination analysis; heuristics; local search; exponential neighborhoods; variable neighborhood search; AVERAGE VALUE; SEARCH; COMPLEXITY; LINKAGES;
D O I
10.1287/ijoc.2019.0893
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
The bilinear assignment problem (BAP) is a generalization of the well-known quadratic assignment problem. In this paper, we study the problem from the computational analysis point of view. Several classes of neighborhood structures are introduced for the problem along with some theoretical analysis. These neighborhoods are then explored within a local search and variable neighborhood search frameworks with multistart to generate robust heuristic algorithms. In addition, we present several very fast construction heuristics. Our systematic experimental analysis disclosed some interesting properties of the BAP, different from those of comparable models. We have also introduced benchmark test instances that can be used for future experiments on exact and heuristic algorithms for the problem.
引用
收藏
页码:730 / 746
页数:17
相关论文
共 50 条