| Summary: | 碩士 === 元智大學 === 工業工程研究所 === 87 === A control chart may indicate an out-of-control condition when some nonrandom patterns occur. Different nonrandom patterns can
be associated with a specific set of assignable causes. Hence, identification of nonrandom patterns can greatly narrow the set of possible causes that must be investigated, and thus the diagnostic search could be reduced in length. In the past, studies on control chart patterns recognition have focused on the recognition of single patterns, little has been done on situations where multiple patterns exist.
In this research, we develop a neural network-based pattern recognizer for the analysis of control chart patterns. This patterns recognizer looks for the following nonrandom patterns: trend, cycle, shift and the multiple combinations of these patterns.
In addition to the raw data, some important features extracted from process data were used as the inputs of neural network.
The pattern recognizer has been evaluated by estimating the average run length and the rate of correct classification. The simulation results show that the pattern recognizer can recognize the control chart patterns with correct classification rate of 92%. The results also show that the features developed in this research can significantly improve the performance of neural network.
|
|---|
