Modelling Lactation Performance in Frieswal Cows: A Comparative Study of Mathematical Models
Keywords:
Gamma-Type Function, Frieswal Cows, Holstein Friesian, Lactation Curve, Monthly Test Day, R2, RMSE, Non-Linear Models, Peak Yield, Persistency of ProductionAbstract
The present study was undertaken to compare four non-linear lactation curve models (Gamma-Type (GT), Mixed Logarithmic (ML), Wilmink (WL) and Quadratic (QD)) to describe the monthly milk yield patterns in Frieswal cows using the data of five lactations (2019-2024) at JNKVV, Jabalpur. Model performance was evaluated on criteria of peak yield, persistency, time to peak production and goodness of fit. The Gamma-Type model had the best overall accuracy, particularly at the end of lactation, with high and consistent coefficients of determination (R2adj = 0.977–0.988). The best fit for cows in their first lactation was obtained by the Mixed Logarithmic model (R2adj = 0.943), indicating that this model is appropriate for early parity dynamics. The Wilmink model matched the early lactation trends well, but with a relatively lower explanatory power (R2adj = 0.921–0.977). The Quadratic model showed little flexibility to describe the non-linear nature of lactation curves (R2adj = 0.918–0.978). Peak milk yield of Frieswal cows was observed in the second month of lactation. The findings highlight the significance of parity-based model selection to enhance the accuracy of milk yield prediction and enable more effective herd management strategies.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Mujahida Sayyed, Prakash Kumar Kumawat, Tufleuddin Biswas, L.S. Shekhawat, Umesh Singh, Pradeep Mishra, Kuldeep Rajpoot

This work is licensed under a Creative Commons Attribution 4.0 International License.
