粗糙集理论在装备保障训练效果评估中的应用  

Application in the Evaluation of Equipment Support Training Effect Based on Rough Set

作　　者：崔凯旋[1] 石全[1] 张文宇[1] 胡起伟[1] 吕勇[1] 

机构地区：[1]军械工程学院装备指挥与管理系,河北石家庄050003

出　　处：《指挥控制与仿真》2012年第3期129-134,共6页Command Control & Simulation

摘　　要：装备维修保障训练效果评估中最关键的步骤就是评估指标体系的构建及优化,为了克服传统评价方法中存在指标关联性大和权重确定主观性强的问题,将粗糙集和信息熵理论引入装备维修保障训练效果评估领域。采用信息熵的算法对连续指标离散化,基于粗糙集区分矩阵的方法对指标体系进行属性约简,降低了模型的复杂性;在此基础上提出了粗糙集属性重要度和AHP法相结合确定指标权重的方法,进而通过专家评议应用于训练效果评估实例,得到了较好的评估效果,为装备维修保障训练评估提供了新的参考方法。The key procedure in training effect evaluation of troops is the construction and optimization of indicators system scientifically and reasonably, in order to overcome the subjectivity and relevance between indicators using traditional evaluation method,rough set and information entropy theory were introduced to the evaluation of equipment maintenance support training effect. Successive data was scattered through information entropy algorithm and indicators system was decreased based on differentiate matrix in order to reduced the model＇s complexity; the method rough set attribute significance degree combined with AHP theory was applied to fix the indicators＇ weight, the whole process and method were testified in evaluation example with appraisal data of evaluation experts, verified the method is reasonable and valid, providing a new method for equipment maintenance support training evaluation.

关 键 词：粗糙集 信息熵 属性约简 属性重要度 

分 类 号：E917[军事]