This network provided the reference value for a CUSUM chart on the basis of current process data, but the discriminating criterion was still constructed by a general CUSUM.

• 为改善此一缺点，本研究提出以类神经网路为基础之参考值设定方法，利用制程之现有数据以类神经网路提供及时之参考值大小，再由累积和管制法判断制程是否异常。