Near-Infrared Reflectance Spectroscopy (NIRS) detection to differentiate morning and afternoon milk based on nutrient contents and fatty acid profiles

Milking time is one of the factors that affect milk quality. The objective of this study was to differentiate morning milk from afternoon based on milk fatty acid profile and create a prediction model using Near-Infrared Reflectance Spectroscopy (NIRS). This study used explorative research and post-observation analysis. Milk sampling was collected from three different dairy farm locations in West Java Provinces (Pangalengan district of Bandung Regency, Cibungbulang District of Bogor Regency, and Tanah Sareal District of Bogor Municipality). Milk quality observed in this study included milk fat, protein, lactose, solid non-fat (SNF), and fatty acid compositions. Milk fat, protein, lactose, and SNF were analyzed using Lactoscan. Fatty acid compositions were identified using gas chromatography (GC). Sample spectrums were collected using NIRSflex 500. The difference between morning and afternoon milking was tested using a t-test carried out by SPSS ver. 25. Qualitative calibration of milk quality was conducted using NIRSCal v5.6 by applying the cluster (CLU) method. The results from lactoscan and GC showed that milk fat, caprylic acid, and myristoleic acid, and total SFA were significantly different (Sig. (2-tailed) < 0.05) in morning and afternoon milk. However, NIRS failed to generate a sophisticated model for the milk quality differentiation, which shows a low Q-value (0.0011231). The quantitative analysis accurately produced milk fat and total SFA predictions but failed to accurately predict caprylic acid and myristoleic acid. This study concluded that morning milk could be differentiated from afternoon milk based on milk fat, caprylic acid, myristoleic acid, and total SFA content. The NIRS technology can differentiate between morning and afternoon milk based on quantitative calibration of total fat and SFA.