A decision support framework for reliable early prediction of long-term cumulative milk yield in dairy cows

Early prediction of cumulative milk yield from initial lactation records enables timely culling, selective breeding, and optimized feed allocation, thereby improving herd efficiency and farm profitability. However, predicting long-term production remains challenging due to high variability in early-lactation patterns and limited available data. This study aimed to develop and compare predictive models for estimating long-term cumulative milk yield from daily production records, and to identify the optimal approach and minimum days in milk (DIM) threshold for reliable early-lactation prediction.

We used 211 lactation datasets with complete records of up to 305 DIM, selected from the daily milk yield records of 727 dairy cows. Six models were evaluated: three statistical methods (simple averaging, adjustment factor, and correction coefficient) and three ensemble machine learning algorithms (CatBoost, XGBoost, and Random Forest). In addition, because lactation patterns differ between primiparous and multiparous cows, parity‑specific models were developed, and predictive performance was assessed using mean absolute percentage error (MAPE) between predicted and actual yields.

The 305-day cumulative milk yield was predicted within an error margin of 11% across all base DIMs, with the smallest discrepancy of 0.5 kg observed in the XGBoost model. By algorithms, machine learning models outperformed during early lactation (MAPE 8.878 at base DIM 30 for parity ≥2), whereas statistical models achieved higher accuracy after base DIM 120 (MAPE 4.311 at base DIM 150 for parity 1).

This study provides a novel lactation stage-dependent and parity-specific prediction framework that bridges the gap between early-lactation uncertainty and reliable long-term productivity estimation. By enabling model selection tailored to data availability and lactation stage, the proposed approach enhances the practical applicability of predictive analytics in dairy farming. This framework provides a practical decision-support tool for dairy farm management that facilitates early culling, feeding optimization, and productivity-based herd management.