论文部分内容阅读
在质量管理中,最常用的控制图是休哈特的X-R图。使用这种图有个前提条件,就是工序必须服从正态分布。在这种情况下,X图和R图犯第Ⅰ类错误的概率分别为0.0027和0.0046。如果工序分布不是正态而是偏态,仍用传统的休哈特X-R图,将会增加犯第Ⅰ类错误的概率,导致误发警报,给生产带来不利影响。实际中,偏态分布工序不少,诸如一些化
In quality management, the most commonly used control chart is the Shewhart X-R chart. There is a prerequisite for using this figure, which is that processes must follow a normal distribution. In this case, the probability of Type I error in X-Chart and R-Chart is 0.0027 and 0.0046, respectively. If the process distribution is not normal but skewed, still using the traditional Shewhart X-R graph, the probability of making a Type I error is increased, resulting in a false alarm and adversely affecting production. In practice, there are many decentralized processes, such as some