Impact of Cow Parity on the Accuracy of Near-Infrared Spectroscopy for Sustainable Milk Quality Monitoring during Milking

Abstract Accurate real-time monitoring of milk quality during milking is essential for sustainable dairy farming, yet factors such as cow parity may affect the performance of near-infrared (NIR) spectroscopic sensing systems. This study investigated how cow parity (the number of calvings) impacts the reliability and accuracy of NIR spectroscopy in assessing key milk quality indicators: fat content, lactose, and somatic cell count (SCC). Experiments were conducted at Hokkaido University with two cows in their second parity. Milk spectra were recorded across 700–1050 nm using the NIR system, while fat and lactose were measured with a MilkoScan device and SCC with a Fossomatic device. Calibration models were developed using first parity, second parity, and combined datasets through partial least squares regression. Model performance was evaluated via coefficients of determination and standard errors of prediction. Results showed comparable accuracy for milk fat and SCC across parities, whereas lactose measurements were more affected. Cross-validation between first and second parity datasets further confirmed parity-dependent variations, particularly for lactose. These findings suggest that cow parity should be considered when implementing NIR-based milk quality monitoring, supporting more precise, resource-efficient, and sustainable dairy management practices. Competing Interest Statement The authors have declared no competing interest.