摘要
随着需求和退货波动的增大,如何有效控制库存,成为当前零售企业管理者亟待解决的难题.在库存水平动态变化表示为Markov调制Levy过程条件下,利用水平穿越和多维鞅理论,分别确定了库存水平分布及循环期望费用和时间函数,在此基础上,构建了系统长程平均费用率模型.最后,通过仿真试验,对比了3种订货策略下需求和环境状态变化系统的性能,并分析了退货对最优控制策略的影响.进而,得到一些重要的库存管理启示.
With the increase of fluctuations in demands and returns, it is now a very difficult problem for retail managers to control their inventory effectively. Under the condition that the inventory level process is expressed as a Markov- modulated Levy process, the stationary distribution of the inventory level, as well as the expected cycle cost and time functions are derived by utilizing level crossing and multi-dimensional martingale theories. Subsequently, the functions are employed to develop the long-run average cost rate model. Finally, we compare the system performance for three kinds of order policies when the demand rates and the status parameters are varying, and investigate the impact of the returns on the optimal control policy. Consequently, some important management insights are obtained.
出处
《控制理论与应用》
EI
CAS
CSCD
北大核心
2015年第7期934-941,共8页
Control Theory & Applications
基金
国家科技重大专项项目(2010ZX04001-161)
国家高技术研究发展计划("863"计划)项目(2012AA041307)
"泰山学者"建设工程专项经费资助~~
作者简介
娄山佐（1964-），男，副教授，主要研究方向为复杂系统建模与仿真、库存控制、优化算法和随机网络分析等，E-mail： lshanzuo@163.com; Tel.： ＋86 13295319058.
田新诚（1965-），男，教授，主要研究方向为运动控制、机电一体化以及智能控制和优化，E-mail：txch@sdu．edu．cn．
