基于灰靶理论的动态评价方法

发布时间：2018-05-28 03:27

本文选题：灰靶理论 + 动态评价方法　；参考：《系统科学与数学》2017年01期

【摘要】：为了比较多个系统在不同时刻的发展水平和某一段时间的动态累积水平,需要进行动态评价.传统灰靶理论方法无法比较多个系统在某一时间段各个时刻和总体的发展水平,只考虑各评价指标值的差异程度(现状),不考虑指标值增长程度.将二维数据加入时序数据扩展为三维数据,提出基于灰靶理论的动态评价方法,该方法既可以得到各评价对象(方案)好坏程度的评价值及排序结果,也可以得到各评价对象(方案)增长程度的评价值及排序结果,还可以得到同时考虑各评价对象指标值差异程度和增长程度的综合评价值.既可以得到各评价对象(方案)各时刻的综合评价值及排序,也可以得到各评价对象在某个时间段内总体的评价值和排序结果.基于靶心度可以进行模式识别、分级和选优.基于贡献度分析,可以分析各指标对模式靶心度的影响程度,从而找到关键影响因素.最后用一个实例说明本研究实际应用上的有效性.该方法给动态评价提供了思路,是综合评价方法研究的有益补充,可以利用该方法进行应用研究.
[Abstract]:In order to compare the development level of multiple systems at different times and the dynamic accumulation level of a certain period of time, dynamic evaluation is needed. The traditional grey target theory method can not compare the development level of multiple systems at each time and in a certain time, and only consider the difference degree of each evaluation index value (present situation, not the increase degree of index value). A dynamic evaluation method based on grey target theory is proposed by extending two-dimensional data into 3D data. The method can not only get the evaluation value and ranking result of each evaluation object (scheme). The evaluation value and ranking result of the increasing degree of each evaluation object (program) can also be obtained, and the comprehensive evaluation value can also be obtained considering the difference of index value and the growth degree of each evaluation object. Not only can the comprehensive evaluation value and ranking of each evaluation object (scheme) be obtained at each time, but also the overall evaluation value and ranking result of each evaluation object in a certain time period can be obtained. Pattern recognition, classification and optimization can be performed based on target center. Based on the analysis of contribution degree, the influence of each index on the target center of the model can be analyzed, and the key influencing factors can be found. Finally, an example is given to illustrate the effectiveness of this study in practical application. This method provides a train of thought for dynamic evaluation and is a useful supplement to the study of comprehensive evaluation method, which can be used for application research.
【作者单位】：福州大学经济与管理学院;中国科学院科技政策与管理科学研究所;
【基金】：国家自然科学基金项目(71403055) 中国博士后科学基金(2015M570155,2015M571134,2016T90140) 福建省高等学校新世纪优秀人才支持计划资助国家留学基金资助课题
【分类号】：F124.3;N941.5

【相似文献】