

AWX enables the prediction of chemical properties from a sample’s NIR/NIT spectrum.
Developing robust calibration models from spectral data and well-characterized reference samples typically requires significant expertise in chemometrics. Unlike many comparable tools, Application worx G2 is clearly structured and guides users through each step with a strong focus on workflow. Well-organized wizard screens provide comprehensive support, allowing even non-specialists to build calibrations using Multiple Linear Regression (MLR) or Partial Least Squares Regression (PLSR).
During calibration development, users can choose automated optimization options to improve model performance, while experienced analysts can also apply their own expertise at every stage. Processed spectra can be visualized in parallel, enabling obvious outliers or abnormal spectra to be identified and excluded immediately. The selected spectra used for model calculation can be saved as a new batch, supporting ongoing refinement and future model improvements.
Application worx G2 also allows multiple PLSR models—using different combinations of spectral pretreatments—to be calculated in parallel. Users can easily compare model performance and select the most suitable option.
Finally, Application worx G2 compiles applications that bundle calibrations for multiple product properties. These applications can then be used to predict the relevant properties from spectral data collected by NIR spectrometers.
Application Worx G2
| System requirements | MS Windows 10 or 11 |
| Scope of delivery | AWX installation and help files. Registry software online |
| Instrument compatibility | All SpectraAlyzer instruments (GRAIN NEO, BRAUMEISTER, CLINICAL, DAIRY, FLEX, FLOUR, FOOD, MEAT, OLIVE, SPIRITS, SUGAR, TEXTILE, WINE, MIA and etc.), Brabender Kernelyzer and Olivas Gold 2 |
| Administration spectral and chemical reference data, calibration development, calibration adjustment and application composition. | |
| Clearly structured user interface. | |
| Easy spectra import from instrument files. | |
| Easy chemical reference values import from csv file. | |
| Data storage using MySQL database (local pc or intranet/internet). | |
| Model calculation utilising MLR or PLSR or create manually. | |
| Wizard guided calibration set-up. | |
| Different data pretreatments available in PLSR models. | |
| Visualization of the processed spectra. | |
| Parallel calculate multiple PLSR models and directly compare the models. | |
| Best results are automatically selected for each factor. | |
| Statistics support user to choose optimal model. | |
| Utilise outlier statistics to improve the model individually. | |
| Full history of calibration iteration available. | |
| Recalculate existing calibration models by using or by adding new batches. | |
| Easily adjust existing calibration models. | |
| Directly export calibration models for different types of instrument. |