基于位置和散度效应因子分类筛选及仿真试验

针对供应链及其它计算机仿真试验中所涉及因子数目众多的情形,序贯分支方法因筛选效率较高被广泛采用.然而,现有的序贯分支方法忽视对散度效应起显著作用的因子.由于相比位置效应,散度效应在产品或供应链系统的质量设计中是同等甚至更为重要的,故提出一种基于位置效应与散度效应的序贯分支方法.首先,将影响位置效应与散度效应的显著因子分类为调节因子,重要因子和稳健因子;然后,结合序贯概率比检验(sequential probability ratio test,SPRT),在控制筛选过程中第一类错误和第二类错误的基础上,采用序贯分支方法分类地筛选出显著因子;最后通过仿真试验说明此方法能够有效且分类地处理位置效应与散度效应下的显著因子筛选问题.

Screening for location and dispersion effects simultaneously by sequential bifurcation

Aiming at the situation that many factors are involved in most simulation experiments, such as supply chain system or other computer simulation experiment, sequential bifurcation (SB) is widely used due to its high efficiency in screening. However, SB usually ignores the factors which are significant to dispersion effects. According to the theory of quality design used in products or supply chain, dispersion effects play an equally important part as location effects, and hence a new sequential bifurcation method considering both location effects and dispersion effects is proposed to screen the significant factors out and classify them simultaneously. Firstly, the significant factors are classified as adjusted factors, important factors and robust factors, according to their significance to the location effects and dispersion effects. Secondly, combined sequential probability ratio test (SPRT) to control the two types of error during the screening procedure, the significant factors are screened out and classified into three types. Finally, the efficiency and robustness of the proposed method which considers location effects and dispersion effects simultaneously is verified through a simulation model.