摘要
网络制造环境下合作伙伴的选择采用的主要方法是对被选企业的一些宏观信息进行评价,该方法无法保证在形成动态联盟后制造系统的运行效率,并且要求预先确定任务的分配,无法保证各个企业之间任务分配的合理性。针对这样的问题,从更底层的制造资源优化配置入手,研究制造资源如何在网络制造环境下优化配置,从而确定合作伙伴和各合作伙伴所需要承担的生产任务。任务分配涉及企业—资源—任务三者之间的匹配,使得原本就是NP难题的调度问题更加复杂,借鉴传统遗传算法和生物学进化原理,提出一种双链遗传算法来解决这样的规划问题,相对于传统遗传算法这样算法的效率更高,编码解码更容易,可以解决网络制造资源优化配置问题。
The main method of partner-selection in global manufacturing is to evaluate the macro-parameters of candidates. But this kind of method cannot assure the efficiency of the manufacture system in the virtual enterprise, and the .job scheduling must be done before the selection of partners ,that cannot assure the reasonable of the jobs for the partners. To solve the problem ,sourcing oriental configuration is researched in global manufacturing, then the selection of partners is implemented. But thus complicate the .job scheduling problem that is a kind of NP-hard problem. To solve the problem ,a double-link genetic algorithm is presented based on the conventional genetic algorithm and biologic evolution. This method is more effective, and easier to code and decode than the conventional genetic algorithm. The method can easily solve the pro blem of sourcing configuration.
出处
《机械工程学报》
EI
CAS
CSCD
北大核心
2008年第2期189-195,共7页
Journal of Mechanical Engineering
基金
国防重大基础科研资助项目
关键词
遗传算法
虚拟企业
网络化制造
Genetic algorithm Virtual enterprise Global manufacturing
