运筹与管理 ›› 2023, Vol. 32 ›› Issue (1): 134-140.DOI: 10.12005/orms.2023.0022

• 应用研究 • 上一篇    下一篇

基于评价信息随机化的群体评价方法与应用

易平涛, 王士烨, 李伟伟, 董乾坤   

  1. 东北大学 工商管理学院,辽宁 沈阳 110167
  • 收稿日期:2020-12-21 出版日期:2023-01-25 发布日期:2023-03-01
  • 作者简介:易平涛(1981-),男,湖南永州人,教授,博士,研究方向:评价理论与方法,信息挖掘;王士烨(1995-),男,山东东营人,博士研究生,研究方向:综合评价。
  • 基金资助:
    国家自然科学基金资助项目(72171040,72171041)

Group Evaluation Method Based on Randomization of Evaluation Information and Its Application

YI Pingtao, WANG Shiye, LI Weiwei, DONG Qiankun   

  1. School of Business Administration, Northeastern University, Shenyang 110167, China
  • Received:2020-12-21 Online:2023-01-25 Published:2023-03-01

摘要: 针对群体评价中共识集结的相关问题,从仿真的视角讨论了评价信息随机化的群体共识聚合求解方法。首先,面向实数类型的评价信息,将精确性的数据给予一定的宽松性处理,进一步结合正态分布的3σ原则,利用随机模拟的方式集结出带有概率特征的可能性排序;其次,面向区间数类型的评价信息,整合出各子区间发生概率不同的区间数评价信息,在充分随机模拟的情况下,给出了带有优胜概率特征的可能性排序。最后,通过相应的算例进行求解分析,说明了该方法的可行性和有效性。基于群体共识视角,针对实数和区间数两种类型的评价信息,分别进行相应的随机化处理,并为进一步探索区间数的分布形式提供了一种新的研究思路。

关键词: 综合评价, 群体评价, 群体共识, 随机模拟, 可能性排序

Abstract: For the problems related to consensus aggregation in group evaluation, the group consensus aggregation solution method for randomization of evaluation information is discussed from the perspective of simulation. Firstly, based on the evaluation information of real number type, the accurate data are treated with a certain degree of looseness, and then combined with the three sigma principle of normal distribution, the possibility ranking with probability characteristics is aggregated by means of random simulation. Secondly, based on the evaluation information of interval number type, the evaluation information of interval number with different occurrence probability of each subinterval is integrated, and the possibility ranking with winning probability is given under the condition of sufficient stochastic simulation. Finally, an example is given to illustrate the feasibility and effectiveness of the proposed method.Based on the perspective of group consensus, the evaluation information of real number and interval number is respectively randomized, which provides a new research idea for further exploration of the distribution form of interval number.

Key words: comprehensive evaluation, group evaluation, group consensus, stochastic simulation, probability ordering

中图分类号: