Model robustness
Develop robust models to guarantee performance
The robustness of a prediction model is its ability to remain stable against external disturbances. This issue is fundamental in industrial conditions where many factors can vary.
Improving the model robustness involves many chemometric methods, applied during various model development state: when deriving the model, during its update, maintenance or transfer…
The techniques able to improve model robustness will:
IMPROVE drastically the reliability of the model predictions, with a better precision and stability
MAKE OPERATIONAL a model developed under laboratory conditions
SIGNIFICANTLY REDUCE the effect of influential factors
CORRECT drifts
PERFORM maintenance and updating of operational models
INCREASE product/process knowledge and understanding models
For more information on chemometric methods of optimization of robustness, see our R&D pages.
To see for yourself their effects, do not hesitate to contact us!
