论文部分内容阅读
针对不同故障模式与影响分析(FMEA)中专家群决策的不确定性和主观性问题,提出混合概率分布下基于前景理论的三角模糊随机多属性群决策(TFRMAGDM)方法。首先,在前景理论框架下定义基于混合概率分布和三角模糊随机变量(TFRV)的期望-方差决策矩阵,建立基于决策群组意见一致性的专家属性权重极大熵模型。然后,结合期望-方差决策矩阵和属性权重模型,获取综合边际前景期望-方差矩阵,从而根据定义的序关系判别准则确定不同故障模式下风险优先次序排序。最后,将此方法应用于某照明设备的故障模式风险排序。结果表明,用TFRMAGDM方法能够量化分析具有模糊性和随机性的故障模式的风险,明确其优先次序。
Aiming at the uncertainty and subjectivity of expert group decision-making in different failure modes and effects analysis (FMEA), a TFMMA method based on the theory of the future under mixed probability distribution is proposed. Firstly, the expectation-variance decision matrix based on the mixed probability distribution and triangular fuzzy random variables (TFRV) is defined in the framework of the foreground theory, and a maximum entropy model of expert attribute weight based on consensus of decision-making groups is established. Then, combined with the expectation-variance decision matrix and the attribute weight model, the comprehensive marginal foreground expectation-variance matrix is obtained, and then the order of risk prioritization under different failure modes is determined according to the defined ordinal discrimination criterion. Finally, apply this method to a lighting device to troubleshoot risk sequencing. The results show that the TFRMAGDM method can quantitatively analyze the risk of fuzziness and stochastic fault modes and clarify the priorities.