灰色多指标风险型决策方法研究

Research on grey multi-criteria risk decision-making method

针对方案指标评估值为区间灰数的风险决策问题,提出了灰色多指标风险型决策的概念。将灰色系统理论的思想和方法与经典风险决策方法相融合,对风险型决策问题指标权重完全未知的且指标值为区间灰数的情况进行了探讨。利用分析技巧,建立了灰色模糊关系法及双基点法两种决策方法。在灰色模糊关系算法中,利用信息熵确定的指标权重使决策方法更符合客观要求。双基点算法在一定程度上解决了单方面基于理想点或负理想点进行决策时,未能充分利用已知信息所产生的偏差,决策更贴近于实际,应用说明了所提出的两种决策方法的合理性和算法的有效性。

This paper presents a new conception grey multi-criteria risk decision-making,to study risk multi-criteria decision-making problems in which criteria values are interval gray numbers. Combining thoughts and methods of grey system theory with classical risk decision-making method, it studies risk decision-making problem in which criteria weights are completely unknown and criteria values are interval grey numbers, and sets up two algorithms, grey fuzzy relationship method and two-basic-point method. In the former, criteria weights are determined by information entropy, which ensures the algorithms to be fit for objective desire better. In some extent, the later gets rid of the deviation caused by deficient use of known information in decision-making process only based on ideal point or negative ideal point. Thus, decision result will keep closer to practice. An example illustrates the validity and efficiency of two algorithms given.