多产品综合作业调度问题及其求解  

Multi-product complete job-shop scheduling problem and its solution

作　　者：王林平[1,2] 贾振元[1] 王福吉[1] 孟凡斌[1] 

机构地区：[1]大连理工大学精密与特种加工教育部重点实验室,大连116024 [2]烟台大学机电汽车工程学院,烟台264005

出　　处：《系统工程理论与实践》2009年第9期73-77,共5页Systems Engineering-Theory & Practice

基　　金：国家863/CIMS主题(2003AA414022)

摘　　要：描述了同时考虑加工与装配作业的多产品综合作业调度问题,利用经典作业调度问题FT10的数据构造了一个问题例子,设计了基于产品结构信息的字符串直接解码方法.计算结果表明:设计的遗传算法是可行和有效的,得到的装配完工时间接近无限能力概念下得到的最小值,并且优于关键路径调度规则的结果.Assembly scheduling plays a great part in job shops with high-variety low-volume production. We will describe a multi-product complete job shop scheduling problem （CJSSP） with fabrication and assembly operations. The classic job shop scheduling benchmark FT10 was adapted to be our problem instance. A selective decoding string that embodies product structural information was designed to decode chromosome directly. Computation results indicate the feasibility and validity of the proposed approach. Moreover, the achieved assembly finish time is close to the lower limit under the assumption of finite capacity and better than the result of the CP dispatching rule.

关 键 词：多产品 装配约束 遗传算法 解码 综合作业调度问题 

分 类 号：TP301[自动化与计算机技术—计算机系统结构] F406.2[自动化与计算机技术—计算机科学与技术]