一种基于多子网与线性目标规划的模糊网络分析法模型

A Model of Fuzzy Analytic Network Process Based on Multi-Subnets and Linear Goal Programming

针对经典ANP模型在解决现实复杂评价决策问题中的困难,通过引入"子网"结构,提出了一个多层递阶分解的网络模型,有效降低了建模工作量与复杂度。同时,考虑比较判断的不确定性与多样性,提供了集成多种表达的基于线性目标规划的模糊判断矩阵权向量求解算法。整合多子网与模糊判断,构建了评估流程。以空管体系为例进行能力评估,验证了所提模型算法的有效性。

Aiming at the difficulty of using the classical ANP model to solve complex decision problems, we pro- pose a network model in multiple levels based on hierarchical decomposition, through introducing ＇ subnet＇ struc- ture, which reduces the workload and complexity effectively. At the same time, considering the uncertainty and di- versity of the judgment, we provide a linear goal programming model for fuzzy pairwise comparison matrices with multiple expressions. By adopting the extended ANP model with subnets and fuzzy judgments, we construct the e- valuation process. Finally, an example for the capability evaluation of air traffic management system-of-systems in- dicates that the method is effective.