A hierarchy of relaxations leading to the convex hull representation for general discrete optimization problems

被引:26
|
作者
Adams, WP [1 ]
Sherali, HD
机构
[1] Clemson Univ, Dept Math Sci, Clemson, SC 29634 USA
[2] Virginia Polytech Inst & State Univ, Grado Dept Ind & Syst Engn, Blacksburg, VA 24061 USA
基金
美国国家科学基金会;
关键词
discrete program; binary optimization; reformulation-linearization technique;
D O I
10.1007/s10479-005-3966-4
中图分类号
C93 [管理学]; O22 [运筹学];
学科分类号
070105 ; 12 ; 1201 ; 1202 ; 120202 ;
摘要
We consider linear mixed-integer programs where a subset of the variables are restricted to take on a finite number of general discrete values. For this class of problems, we develop a reformulation-linearization technique (RLT) to generate a hierarchy of linear programming relaxations that spans the spectrum from the continuous relaxation to the convex hull representation. This process involves a reformulation phase in which suitable products using a defined set of Lagrange interpolating polynomials (LIPs) are constructed, accompanied by the application of an identity that generalizes x(1-x) for the special case of a binary variable x. This is followed by a linearization phase that is based on variable substitutions. The constructs and arguments are distinct from those for the mixed 0-1 RLT, yet they encompass these earlier results. We illustrate the approach through some examples, emphasizing the polyhedral structure afforded by the linearized LIPs. We also consider polynomial mixed-integer programs, exploitation of structure, and conditional-logic enhancements, and provide insight into relationships with a special-structure RLT implementation.
引用
收藏
页码:21 / 47
页数:27
相关论文
共 50 条