New approach to optimize near-infrared spectra with design of experiments and determination of milk compounds as influence factors for changing milk over time

• DoE was used for the first time in order to optimize NIR parameters for milk analysis.
• The optimized method was used to screen changes in fatty acid composition over time.
• The method was applied to distinguish milk by identifying individual spectral markers.

The optimization of near-infrared spectroscopic parameters was realized via design of experiments. With this new approach objectivity can be integrated into conventional, rather subjective approaches. The investigated factors are layer thickness, number of scans and temperature during measurement. Response variables in the full factorial design consisted of absorption intensity, signal-to-noise ratio and reproducibility of the spectra. Optimized factorial combinations have been found to be 0.5 mm layer thickness, 64 scans and 25 °C ambient temperature for liquid milk measurements.Qualitative analysis of milk indicated a strong correlation of environmental factors, as well as the feeding of cattle with respect to the change in milk composition. This was illustrated with the aid of near-infrared spectroscopy and the previously optimized parameters by detection of altered fatty acids in milk, especially by the fatty acid content (number of carboxylic functions) and the fatty acid length.