摘要
针对广义不确定型决策问题,提出了一种基于案例的决策理论(CBDT)的决策分析方法.首先,分别计算针对"问题"和"方案"的目标案例与历史案例间的相似度,通过事先设置相似度阈值并构建相似历史案例集来确定对应于备选方案的相似历史案例;然后,分别计算针对"问题–方案"对的目标案例与相似历史案例间的相似度和相似历史案例的方案实施效果的效用值,集结相似度和效用值,进而得到每个备选方案的综合效用值,并依据综合效用值的大小来选择最优的备选方案.最后,通过一个算例说明了提出方法的可行性与有效性.
This paper proposes a decision analysis method based on case-based decision theory （CBDT） to solve the general uncertain decision making problem. First, the similarities between target case and historical case with regard to problem and alternative are calculated, respectively. Then the similar historical cases corresponding to each alternative are determined by setting similarity threshold beforehand and building similar historical case set. On the basis of the above, the similarities between the target case and similar historical case on ＂problem-alternative＂ pairs are calculated, and then for each alternative used in the similar historical case, the utility of its implementation effect is calculated. Furthermore, the overall utility of each alternative is calculated by integrating the similarities and utilities, and the desirable alternative can be determined according to the obtained overall utilities. Finally, a numerical example is used to illustrate the feasibility and validity of the proposed method.
出处
《系统工程学报》
CSCD
北大核心
2014年第1期21-29,共9页
Journal of Systems Engineering
基金
国家自然科学基金资助项目(71271051
71021061)
中央高校基本科研业务费资助项目(N110706001
N120606 001)
关键词
广义不确定型决策
基于案例的决策理论
相似度
效用值
方案优选
general uncertain decision making
case-based decision theory
similarity
utility
alternativeselecting
