系统工程与电子技术 ›› 2019, Vol. 41 ›› Issue (5): 1049-1055.doi: 10.3969/j.issn.1001-506X.2019.05.17

• 系统工程 • 上一篇    下一篇

基于云模型和证据理论的装备管理绩效评价

沈延安1, 张君彪2   

  1. 1. 陆军炮兵防空兵学院基础部, 安徽 合肥 230031;
    2. 陆军炮兵防空兵学院研究生大队, 安徽 合肥 230031
  • 出版日期:2019-04-30 发布日期:2019-04-28

Equipment management performance assessment based on cloud model and evidence theory

SHEN Yan’an1, ZHANG Junbiao2   

  1. 1. Basic Department, Army Academy of Artillery and Air Defense, Hefei 230031, China;
    2. The Postgraduate Management Team, Army Academy of Artillery and Air Defense, Hefei 230031, China
  • Online:2019-04-30 Published:2019-04-28

摘要:

针对部队装备管理绩效影响因素复杂,评价过程具有模糊性、不确定性等特点,提出了基于云模型和改进证据理论的装备管理绩效评价方法。首先,构建了装备管理绩效评价指标体系,采用组合赋权法确定指标权重;然后,借助云模型对评语信息进行建模,构造基本概率分配函数,再引入Jousselme距离,通过计算证据可信度和可靠度得到证据的复合权重,以此作为权重系数对证据源进行修正;最后,采用改进的证据融合规则进行各指标基本概率分配融合,得到综合评价结果。通过实例分析验证了指标体系的科学性和算法的有效性,为部队装备管理绩效评价提供了参考。

关键词: 证据理论, 云模型, Jousselme距离, 评价

Abstract:

In view of the complex influence factors of military equipment management performance, the fuzziness and uncertainty of the evaluation process, a method of equipment management performance evaluation based on cloud model and improved evidence theory is proposed. Firstly, the performance evaluation index system of equipment management is established, and the weights of indicators are determined by combination weighting method. Then, a cloud model is used to model the expert comment information, and construct the basic probability assignment function. Then, the Jousselme distance is introduced, and the weight of evidence is obtained by calculating the credibility and reliability of evidence, which is used as a weighting coefficient to adjust original evidence. Finally, the different indexes of basic probability assignment (BPA) are fused to get the evaluation results with the improved evidence fusion rules. The instance analyzing proves the scientificity of the index system and the effectiveness of the algorithm, which can provide a reference for the assessment of military equipment management performance.

Key words: evidence theory, cloud model, Jousselme distance, assessment