Mixed integer linear programming in process scheduling: Modeling, algorithms, and applications

被引：276

作者：

Floudas, CA ^{[1
]}

Lin, XX ^{[1
]}

机构：

[1] Princeton Univ, Dept Chem Engn, Princeton, NJ 08544 USA

来源：

ANNALS OF OPERATIONS RESEARCH | 2005年 / 139卷 / 01期

基金：

美国国家科学基金会;

关键词：

chemical process scheduling; mixed-integer linear programming (MILP); discrete-time model; continuous-time model; branch and bound;

D O I：

10.1007/s10479-005-3446-x

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：

This paper reviews the advances of mixed-integer linear programming (MILP) based approaches for the scheduling of chemical processing systems. We focus on the short-term scheduling of general network represented processes. First, the various mathematical models that have been proposed in the literature are classified mainly based on the time representation. Discrete-time and continuous-time models are presented along with their strengths and limitations. Several classes of approaches for improving the computational efficiency in the solution of MILP problems are discussed. Furthermore, a summary of computational experiences and applications is provided. The paper concludes with perspectives on future research directions for MILP based process scheduling technologies.

引用

页码：131 / 162

页数：32

共 79 条

[1]

[Anonymous], 1979, Computers and Intractablity: A Guide to the Theoryof NP-Completeness

[2] Issues in the use, design and evolution of process scheduling and planning systems [J].