摘要
本文提出了一类新的决策问题--有部分信息的不确定性决策问题,使得决策问题的模型更加接近于实际的决策问题。其主要思想是,传统的决策分析往往把决策问题分成三类:确定性、风险性和不确定性。对风险性决策问题,假设决策者已知状态集合的概率分布,而对不确定性决策问题,则假设决策者对状态集合的概率分布是完全未知,而在实际生活中,经常有这样的情形,对状态集合的概率分布,决策者既不是完全知道,又不是一无所知,如何将决策者关于状态集合的概率分布的这个部分信息,应用于决策分析,是本文研究的问题,作者将解决不确定性决策问题的一些方法和准则推广到有部分信息的不确定性决策问题,在一定程度上解决了这个问题。
This paper deals mainly with a new class of decision problems—decision-making under partial knowledge. In clasical decision analysis only decision under certainty, risk and uncertainty is considered. But decision-making under incomplete or partial information is often met in daily life. In such a situa-tion the decision maker is neither completely aware nor unaware of the probabili-ty distribution of the set of states. The aims of this paper are (1) formally putting forward and defining this problem; (2)using the methods and criteria of solving decision-making under uncertainty to solve this problem.
出处
《系统工程学报》
CSCD
1990年第2期63-67,共5页
Journal of Systems Engineering