Weight is the most basic and important indicator in cattle management, and automation of its measurement serves as a fundamental step toward modern smart livestock farming. Automated weighing systems (AWS) capable of continuously measuring cattle weight, even during movement, have been explored as key monitoring components in smart livestock farming. However, owing to the high measurement variability caused by environmental factors, the accuracy of AWSs has been questioned. These factors include real-time fluctuations due to animal activities (e.g., feeding and locomotion), as well as measurement errors caused by residual feed or excreta within the AWS. Therefore, this study aimed to develop an algorithm to enhance the reliability of steer weight measurements using an AWS, ensuring close alignment with actual cattle body weight. Accordingly, daily weight data from 36 Hanwoo steers were processed using a three-stage approach consisting of outlier detection and removal, weight estimation, and post-processing for weight adjustment. The best-performing algorithm that combined Tukey’s fences for outlier detection, mean-based estimation, and post-processing based on daily weight gain recommended by the National Institute of Animal Science achieved a root mean square error of 12.35 kg, along with an error margin of less than 10% for individual steers. Overall, the study concluded that the AWS measured steer weight with high reliability through the developed algorithm, thereby contributing to data-driven intelligent precision feeding.

Objective: 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 because of 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 prediction.Methods: We used 211 lactation datasets with complete records up to 305 DIM, selected from 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). Because lactation patterns differ between primiparous and multiparous cows, parity-specific models were also developed. Predictive performance was assessed using mean absolute percentage error (MAPE) between predicted and actual yields.Results: The 305-day cumulative milk yield was predicted within an 11% error margin across all base DIMs, with the smallest discrepancy of 0.5 kg observed in the XGBoost model. Machine learning models showed better performance 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).Conclusion: This study provides a lactation stage-dependent and parity-specific prediction framework that bridges early-lactation uncertainty and long-term productivity estimation. By enabling model selection according to data availability and lactation stage, the proposed approach supports the practical application of predictive analytics in dairy farming and provides a decision-support framework for early culling, feeding optimization, and productivity-based herd management.