直觉模糊环境下考虑有限理性特征的决策方法

摘要

针对决策者具有有限理性的心理特征且属性权重和自然状态发生概率完全未知的直觉模糊多属性决策问题,提出了一种基于前景理论和证据理论的多属性决策方法.首先,利用证据理论得到各自然状态发生的概率,进而确定自然状态的决策权重函数;其次,运用正态分布概率密度函数设计直觉模糊决策参考点,并依据属性值与决策参考点之间的差异计算价值函数矩阵,从而获得前景价值矩阵;以综合前景价值最大化为准则构建最优化模型用以确定属性权重,并依据各方案的综合前景价值进行优劣排序.最后,将所提方法应用于对游戏产品的选择开发实例中.对比实验表明,运用所提方法得到的决策结果合理可靠,且更能体现决策者的原始决策信息.%Considering the intuitionistic fuzzy multi-attribute decision-making problem that the decision makers with bounded rationality psychological characteristics and attribute weights and probability information of situation are completely unknown,a multi-attribute decision-making based on the prospect theory and Dempster-Shafer theory was proposed.Firstly,the probabilites of the states were calculated by Dempster-Shafer theory,and the decision-making weight functions of the states were determined.Then,the normal distribution probability density function was utilized to construct the intuitionistic fuzzy reference point.Based on the difference between the attribute value and the reference point,the value function matrix and the prospect value matrix were obtained.In addition,an optimization model was developed to derive the attribute weights with the principle of maximizing the comprehensive prospect value,and all the alternatives were further ordered.Finally,the proposed approach was applied to a numerical example about the selection of game products.The experimental results show that the decision-making results are reasonable and reliable,and the actual situation can be reflected.