Df. Cook et al., Utilization of neural networks for the recognition of variance shifts in correlated manufacturing process parameters, INT J PROD, 39(17), 2001, pp. 3881-3887
Traditional statistical process control (SPC) charting techniques were deve
loped for use in discrete industries where independence exists between proc
ess parameters over time. Process parameters from many manufacturing indust
ries are not independent, however, but they are serially correlated. Conseq
uently, the power of traditional SPC charts was greatly weakened. The paper
discusses the development of neural network models to identify successfull
y shifts in the variance of correlated process parameters. These neural net
work models can be used to monitor manufacturing process parameters and sig
nal when process adjustments are needed.