任务负载竞争下制造资源配置的非合作博弈方法  

Non-cooperative Game for Manufacturing Resources Configuration Driven by Job Load Competition

作　　者：陈冰[1] 刘凯[1] 杨挺[1] 

机构地区：[1]西北工业大学现代设计与集成制造技术教育部重点实验室,西安710072

出　　处：《中国机械工程》2013年第2期233-239,共7页China Mechanical Engineering

基　　金：国家高技术研究发展计划(863计划)资助项目(2009AA04Z122);国家科技支撑计划资助项目(2012BAF10B09)

摘　　要：在制造资源配置过程中,针对不同生产设备为了追求任务负载而存在的自由竞争关系,引入博弈论描述设备之间在任务负载上的相互制约关系。提出基于非合作博弈的制造资源优化配置方法,把生产设备映射为非合作博弈制造资源优化配置模型的局中人,建立局中人的策略及其收益函数。利用模糊聚类分析的方法对博弈局中人进行聚类,进行Nash均衡的求解。通过实例与遗传算法、禁忌搜索、粒子群算法进行比较,结果表明基于非合作博弈的制造资源优化配置方法获得的任务负载策略能够保证生产设备总体具有更高的利用率,由此,验证了算法的有效性和正确性。Manufacturing resources configuration is to load the manufacturing processes onto relevant equipment,and the equipment to accomplish the manufacturing job within certain working time, human and cost. There are free competition relationships between equipment and manufacturing duties in the configuration processes. The game theory was introduced to describe their restrictive relation. Then an optimization method on manufacturing resources configuration based on non--cooperative game was proposed. The equipment was regarded as player, while player＇s strategy and payoff were built. And by fuzzy cluster analysis the players were clustered to solve Nash equilibrium. Finally, an example was given by compared with GA, PSO and Tabu search. The results show that task load strategy obtained by the proposed algorithm can guarantee the overall equipment utilization is higher. Therefore, the effectiveness and correctness of algorithms herein were proved.

关 键 词：任务负载 制造资源配置 非合作博弈 纳什均衡 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TH166[自动化与计算机技术—计算机科学与技术]