传统的过程能力分析基于观测值彼此独立的假设,但是实际生活中的数据常呈现相关的状态。本文对一般的自相关模型进行过程能力分析,从理论和模拟两个角度研究了自相关对传统过程能力分析的影响。提出基于ARMA模型的过程能力分析方法,模拟和数值分析的结果表明对于相关过程而言,该方法较传统方法相比有很大的改进。 Traditional process capability analysis is based on the assumption that observations are independent of each other, but data in real life are often correlated. This paper analyzes the process capability of the general autocorrelation model and studies the influence of autocorrelation on the ability analysis of the traditional process from two aspects of theory and simulation. The method of process capability analysis based on ARMA model is proposed. The results of simulation and numerical analysis show that this method has a great improvement over the traditional method.