Multivalued Decision Diagrams for Sequencing Problems

作     者：Cire, Andre A. van Hoeve, Willem-Jan 

作者机构：Carnegie Mellon Univ Tepper Sch Business Pittsburgh PA 15213 USA 

出 版 物：《OPERATIONS RESEARCH》 (运筹学)

年 卷 期：2013年第61卷第6期

页      面：1411-1428页

核心收录：

学科分类：1201[管理学-管理科学与工程(可授管理学、工学学位)] 07[理学] 070104[理学-应用数学] 0701[理学-数学] 

基　　金：National Science Foundation [CMMI 1130012] Google Research Award Div Of Civil, Mechanical, & Manufact Inn Directorate For Engineering Funding Source: National Science Foundation 

主　　题：sequencing single machine scheduling networks/graphs constraint programming decision diagrams 

摘      要：Sequencing problems are among the most prominent problems studied in operations research, with primary application in, e.g., scheduling and routing. We propose a novel approach to solving generic sequencing problems using multivalued decision diagrams (MDDs). Because an MDD representation may grow exponentially large, we apply MDDs of limited size as a discrete relaxation to the problem. We show that MDDs can be used to represent a wide range of sequencing problems with various side constraints and objective functions, and we demonstrate how MDDs can be added to existing constraint-based scheduling systems. Our computational results indicate that the additional inference obtained by our MDDs can speed up a state-of-the art solver by several orders of magnitude, for a range of different problem classes.