Multivariate statistical process control
Develop a statistical model of an existing process
- Use rapid, low-cost on-line or in-situ spectroscopic measurements.
- Uses multivariate statistics / chemometrics to characterize the processes from relevant, sensitive measurements.
- Control limits derived from sound statistical practice and historical database of measurements.
- Up-front investment in calibration is avoided
- On-going calibration maintenance costs are avoided.