﻿ 装备试验数据质量综合评价指标体系与建模方法
 舰船科学技术  2023, Vol. 45 Issue (12): 173-177    DOI: 10.3404/j.issn.1672-7619.2023.12.035 PDF

Comprehensive evaluation index system and modeling method for equipment test data quality
YUE Ming-qiao, MA Yue-fei, ZHAI Yi-chen
No.92493 Unit of PLA, Huludao 125000, China
Abstract: Quality evaluation plays an important role in improving the quality of equipment test data which is very useful for equipment demonstration identification and improvement. Considering the characteristics of the equipment test data, 28 indexes divided into 12 dimensions are presented to assess the data quality. Based on the group decision making and fuzzy decision map methods, the modeling way of equipment test data quality problem is proposed. The approach can select the fit indexes, derive indexes’ priorities, and achieve data quality score. In the end, a numerical example is given to verify the effectiveness of the proposed method.
Key words: equipment test     fuzzy decision map     data quality
0 引　言

1 装备试验数据质量综合评价指标体系

1.1 政策环境层

1.2 固有质量层

 $CDS = \frac{{ConNum}}{{NumT}} 。$ (1)

 $SCD = 1 - \frac{{TSNum}}{{NumT}}。$ (2)

 $CRT = \frac{{NumF}}{{NumD}}。$ (3)

 $VDD = 1 - \frac{{NumN}}{{NumT}} 。$ (4)

 $ECD = 1 - \frac{{ENum}}{{NumT}}，$ (5)
 $VCD = 1 - \frac{{VNum}}{{NumT}} ，$ (6)
 $ECN = 1 - \frac{{NNum}}{{NumT}} 。$ (7)

 $DPT = \frac{{\displaystyle\sum\nolimits_{i = 1}^n {\lg (T{P_i} - T{G_i})} }}{n} ，$ (8)
 $DAT = \frac{{\displaystyle\sum\nolimits_{i = 1}^n {\lg (O{T_i} - U{T_i})} }}{n}。$ (9)

1.3 效用质量层

 $AD = \frac{{AN}}{n} ，$ (10)
 $RD = \frac{{RN}}{n} ，$ (11)
 $PD = \frac{{PN}}{n} 。$ (12)

2 方法步骤

2.1 指标选取

 $N{D_i} = \frac{{\displaystyle\sum\limits_{j = 1}^m {Ju(i,j)} }}{m} \times 100\% 。$ (13)

2.2 基于模糊决策图法确定所选指标的优先级

 $f(x) = \tanh (x) = (1 - {e^{ - x}})/(1 + {e^{ - x}})。$ (14)

 $\left\{ \begin{gathered} C_{(t + 1)}^{} = f(C_{(t)}^{}E)，\\ C_{(0)}^{} = {I_{n \times n}}，\\ {C^*} = {\lim _{x \to + \infty }}\sum\nolimits_{t = 1}^x {C_{(t)}^{}} 。\\ \end{gathered} \right.$ (15)

 $\left\{ \begin{gathered} \overline z = z/\left\{ {\max (z)} \right\} ，\\ {\overline C ^*} = {C^*}/\gamma。\\ \end{gathered} \right.$ (16)

 $\omega = \overline z + {\overline C ^*}\overline z ，$ (17)

$\omega$ 可看作是指标的最终权重向量。

2.3 计算数据质量得分

2.3.1 专家评价法

 $\begin{split} LS =& \{ {l_1}:{\rm{perfect}} \to 9,{l_2}:{\rm{very}}\; {\rm{good}} \to 7,\\ &{l_3}:{\rm{good}} \to 5, {l_4}:{\rm{not}}\; {\rm{ bad}} \to 3,{\kern 1pt} {\kern 1pt} {l_5}:{\rm{poor}} \to 1\}。\end{split}$ (18)

 $s{g_i} = \frac{1}{m}\sum\limits_{j = 1}^m {s_i^j}。$ (19)

 $\overline {s{g_i}} = \frac{1}{9}s{g_i} 。$ (20)

 $\begin{split}&\{ CLR,ALR,ILR,CSS,ASS,ISS,RPS,ROS,\\ &MSS,GSS,DSS,EAS,DM,DC,DRC,DFC\} 。\end{split}$ (21)
2.3.2 抽样统计方法

2.3.3 汇总得到全局得分

 $R = {\omega ^{\rm{T}}}\nu。$ (22)

3 算　例

1）指标选取

 $\{ CSS,MSS,GSS,CDS,VDD,DSS,EAS,PD\}。$ (23)

2） 指标优先级

 图 1 指标间的影响关系 Fig. 1 The influence relationship between the indicators

3）评估结果

4 结　语

 [1] 龚昕, 周大庆, 等. 武器装备试验数据工程理论与实践 [M]. 北京: 国防工业出版社, 2017: 5–6. [2] 陈卫东, 张维明. 数据质量模型及选择运算中的质量传播研究[J]. 计算机工程与应用, 2007(27): 1-3. CHEN W D, ZHANG W M. Reasearch on quality propagation in data quality model and selection operation[J]. Computer engineering and applications, 2007(27): 1-3. DOI:10.3321/j.issn:1002-8331.2007.27.001 [3] PIPINO L L, LEE Y W, WANG R Y. Data Assessment[J]. Communications of the ACM , 2002 , 45 (4): 211–218. [4] 袁满, 张雪. 一种基于规则的数据质量评价模型[J]. 计算机技术与发展, 2013, 23(3): 81-89. YUAN M, ZHANG X. A rule-based data quality evaluation model[J]. Computer Technology and Development, 2013, 23(3): 81-89. [5] YUAN MAN, LIU WEI. A novel data quality controling and assessing model based on rules[C]//ISECS'10 Proceedings of the 2010 Third International Symposium on Electronic Commerce and Security. Guangzhou: Academy Publisher, 2010: 29–32. [6] CHEN Hao, Measurement data quality for ongoing improvement[J]. China Statistics, 2016(6): 18–19. [7] CHEN Chao. Research for electric power big data quality evaluation model and dynamic exploration technology[J]. Modern Electronics Technique. 2014, 37 (4): 153–155. [8] YU R C, TZENG G H. A soft computing method for multi-criteria decision making with dependence and feedback, Applied Mathematics and Computation[J]. 2006, 180(1): 63–75 [9] YUEFEI M A, CHEN F, YAO D, et al. Air traffic management system performance evaluation based on linguistic decision map with multi-subnets[J]. Journal of National University of Defense Technology, 2017, 39(6): 134–142. [10] LI C H, SUN Y H, DU Y W. An improved analytic network process based on fuzzy decision maps[C]// Control and Decision Conference, IEEE, 2008. [11] 孙宏才, 田平等. 网络层次分析法与决策科学[M]. 北京: 国防工业出版社, 2011: 7–8. [12] FAN W, & GEERTS F. Foundations of data quality management[J]. Journal of Synthesis Lectures on Data Management, 2012. [13] 马一鸣. 政府大数据质量评价体系构建研究[D]. 长春: 吉林大学, 2016