基于随机模拟的综合评价方法及应用

针对序关系分析法由专家给出的指标间重要性之比是一个比值区间的问题,提出了一种基于"蒙特卡罗仿真"思想的随机模拟综合评价方法,将点赋值的序关系法推广至区间赋值的情形.基于随机模拟的思想,依据经典的综合评价过程,模拟反映被评价对象之间优劣关系的优胜度矩阵.在此基础上,给出各被评价对象之间的最佳排序.该方法是对经典的序关系分析赋权法的推广与泛化.最后,用一个算例验证了方法的有效性.

Comprehensive Evaluation Based on Stochastic Simulation

In sequential relation analysis, the ratios of weightnesses between indices as given by experts are always an interval of ratios. A comprehensive evaluation method based on Monte Carlo simulation as a stochastic one is therefore presented, where the dot assignment can be extended to the interval assignment. Based on the stochastic simulation and according to the classical comprehensive evaluation process, the superiority matrix reflecting the relation between the superiority and inferiority of the objects being evaluated is simulated. Then, the optimal ranking of all the objects to be evaluated is given. This method is the extension and generalization of the classic sequential relation analysis. A numerical example is given to verify its validity.