Statistical process control (SPC) charts are useful in monitoring a process. However, the typical Shewhart control charts are insensitive in detecting small process shifts. This would require more samples to detect the process disturbances. Consequently, the search for the root causes of the disturbances may need more time, and the process improvement may take longer. One useful solution for this difficulty is to identify the change point of the process in real time. Once this identification is correctly made, the root causes of the disturbances would be easily determined. This study is motivated to integrate SPC charts and neural networks to quickly identify the change point of the process. The concept of the integration mechanism is discussed, and the fruitful results are also demonstrated.
[1]
Ercan Oztemel,et al.
Control chart pattern recognition using neural networks
,
1992
.
[2]
W. Shewhart.
The Economic Control of Quality of Manufactured Product.
,
1932
.
[3]
G. A. Pugh.
Synthetic neural networks for process control
,
1989
.
[4]
Duc Truong Pham,et al.
Control Chart Pattern Recognition Using Spiking Neural Networks
,
2006
.
[5]
N F Hubele,et al.
X¯ CONTROL CHART PATTERN IDENTIFICATION THROUGH EFFICIENT OFF-LINE NEURAL NETWORK TRAINING
,
1993
.
[6]
P. Werbos,et al.
Beyond Regression : "New Tools for Prediction and Analysis in the Behavioral Sciences
,
1974
.