基于改进MTGS的资源管控预案评估指标赋权算法  

Evaluation Index Weight Algorithm of Resource Control Plan Based on Improved Mahalanobis-Taguchi Gram-Schmidt

作　　者：段艳红[1,2] 丁建江[1] 许红波[1] 周芬[1] 

机构地区：[1]空军预警学院空天预警实验室 [2]95333部队

出　　处：《装甲兵工程学院学报》2017年第2期101-106,共6页Journal of Academy of Armored Force Engineering

基　　金：国家自然科学基金青年科学基金资助项目(61401503)

摘　　要：针对决策者确定评估指标权重存在主观偏好这一问题,提出了基于灰靶贡献度排序和田口施密特正交(Mahalanobis-Taguchi Gram-Schmidt,MTGS)的指标赋权算法。该算法在MTGS系统方法中结合灰色系统技术,直接从评估指标数据入手,并通过计算指标灰靶贡献度得出了指标重要性排序结果;依据排序结果,构造了新的评估指标矩阵,有效地避免了人为因素对指标权重结果的影响。最后,以某传感器组网系统的体系抗干扰资源管控预案评估为例,通过与熵值客观赋权算法的结果进行对比验证了该算法的可靠性。Aiming at the problem that decision-maker has subjective misjudgment in determining evaluation index weight,a new index weight algorithm based on grey target contribution degree ranking and Mahalanobis-Taguchi Gram-Schmidt（ MTGS） is proposed. The algorithm is combined with grey system technique in MTGS,makes a good use of the evaluation index data directly,and works out the index importance ranking result by calculating grey target contribution degree of each index. Then,according to the results of ranking,a new evaluation index matrix is constructed,and can effectively avoid human influence factors on the result of index weight. In the end,taking the system of systems anti-jamming resource control plan evaluation of a sensor networking system as an example,the reliability of the algorithm is verified by comparing with the entropy objective weight algorithm.

关 键 词：灰靶贡献度 传感器组网 资源管控预案评估 田口施密特正交 指标赋权算法 

分 类 号：O159[理学—数学]