Self-adaptive large neighborhood search algorithm for parallel machine scheduling problems  

作　　者：Pei Wang Gerhard Reinelt Yuejin Tan 

机构地区：[1]College of Information Systems and Management, National University of Defense Technology, Changsha 410073, P. R. China [2]Discrete Optimization Research Group, Ruprecht-Karls Universitaeit Heidelberg, Heidelberg 69120, Germany

出　　处：《Journal of Systems Engineering and Electronics》2012年第2期208-215,共8页系统工程与电子技术（英文版）

基　　金：supported by the National Natural Science Foundation of China (70601035;70801062)

摘　　要：A self-adaptive large neighborhood search method for scheduling n jobs on m non-identical parallel machines with mul- tiple time windows is presented. The problems＇ another feature lies in oversubscription, namely not all jobs can be scheduled within specified scheduling horizons due to the limited machine capacity. The objective is thus to maximize the overall profits of processed jobs while respecting machine constraints. A first-in- first-out heuristic is applied to find an initial solution, and then a large neighborhood search procedure is employed to relax and re- optimize cumbersome solutions. A machine learning mechanism is also introduced to converge on the most efficient neighborhoods for the problem. Extensive computational results are presented based on data from an application involving the daily observation scheduling of a fleet of earth observing satellites. The method rapidly solves most problem instances to optimal or near optimal and shows a robust performance in sensitive analysis.A self-adaptive large neighborhood search method for scheduling n jobs on m non-identical parallel machines with mul- tiple time windows is presented. The problems＇ another feature lies in oversubscription, namely not all jobs can be scheduled within specified scheduling horizons due to the limited machine capacity. The objective is thus to maximize the overall profits of processed jobs while respecting machine constraints. A first-in- first-out heuristic is applied to find an initial solution, and then a large neighborhood search procedure is employed to relax and re- optimize cumbersome solutions. A machine learning mechanism is also introduced to converge on the most efficient neighborhoods for the problem. Extensive computational results are presented based on data from an application involving the daily observation scheduling of a fleet of earth observing satellites. The method rapidly solves most problem instances to optimal or near optimal and shows a robust performance in sensitive analysis.

关 键 词：non-identical parallel machine scheduling problem with multiple time windows （NPMSPMTW） oversubscribed self- adaptive large neighborhood search （SALNS） machine learning. 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] O224[自动化与计算机技术—计算机科学与技术]