基于动态效应的序贯分支验证仿真实验及应用  

Simulation experiments for validating sequential bifurcation:A dynamic-effect method

作　　者：施文[1,2] 周程[2] 刘慧[1] 

机构地区：[1]湖北经济学院物流与工程管理学院,武汉430205 [2]华中科技大学管理学院,武汉430074

出　　处：《系统工程理论与实践》2016年第8期2003-2012,共10页Systems Engineering-Theory & Practice

基　　金：国家自然科学基金(71402048;71372134);博士后面上基金项目(2015M582228);教育部人文社会科学基金(15YJC630074)~~

摘　　要：针对序贯分支仿真筛选实验中静态效应验证方法(V-SE)实验效率低且在仿真因子较多时实验效力难以保证的两方面不足,提出了基于动态效应(V-DE)和基于不重要因子(V-UF)的两种验证方法,V-DE在元模型估计中采用饱和设计并在配对检验中使用动态效应检验统计量,VUF仅对不重要因子的两组极端水平进行实验而不需估计元模型.蒙特卡罗仿真的三种方法对比实验以及基于实际背景仿真案例的应用表明,V-DE和V-UF在获得理想实验效力的同时极大提高了仿真实验的效率,大大降低了仿真工作者的实验成本,具有良好的实际应用价值.To address the deficiencies of static-effect validation method （V-SE）, which exhibits a low efficiency and is difficult to provide the efficacy guaranty for a large-scale simulation system, this paper develops two streamlined methods, called V-DE and V-UF, for the validation of sequential-bifurcation screening procedure. V-DE employs saturated design for the estimation of a metamodel, and exploit a test statistic based on dynamic effect of unimportant inputs. V-UF simply experiments with a simulation model for the two extreme input combinations, without the need for estimating a metamodel. The comparative analysis of Monte Carlo simulations, and the corresponding application results from a real-world case study indicate that the proposed methods indeed greatly reduce replicates while obtaining desired statistical precision. Both methods can handle qualitative factors of a simulation model successfully.

关 键 词：筛选设计 序贯分支 动态效应 不重要因子 显著性分析 

分 类 号：F274[经济管理—企业管理]