Scholarlink

<p>The assessment and prediction of dairy cattle longevity are essential for improving herd productivity and sustainability. Hence, modelling milking life (ML), productive life (PL), and culling age (CA) using first lactation traits supports better management of the farm. A dataset comprising 377 Holstein Friesian crossbred cows that had first calving between 2010 and 2020 at the University Livestock Farm and Fodder Research and Development Scheme, Mannuthy of Kerala Veterinary and Animal Sciences University, Kerala, India was analyzed and the obtained mean ML, PL and CA were 1003.13, 1444.96 and 2531.15 days respectively. Milking life and PL were positively correlated with lactation yield and number of completed lactations (NCL), but negatively correlated with age at first calving (AFC). Analysis of the data with accelerated failure time models by an 80:20 train-test split (hold out validation) method identified the log-logistic model with AFC and 305-day first lactation yield (LY1305) as optimal for ML (Root Mean Square Error (RMSE) = 648.34) and PL (RMSE = 889.41) prediction. For CA, the log-normal model using LY1305 alone showed the best fit (RMSE = 846.93). The classification of AFC to assess its impact revealed that cows calving between two and three years had the longest ML and PL. These findings highlight the significance of early AFC and high first lactation performance in extending dairy cow longevity in the tropics. The proposed models offer valuable decision-support tools for optimizing herd management using data-driven approaches.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Survival modelling of Holstein Friesian crossbred cattle longevity traits in the tropics using accelerated failure time approaches

  • 期刊论文

  • 发表日期：

    2026年4月22日

查看全文
  • T. B. Maneesha,
  • V. L. Gleeja,
  • Surej Joseph Bunglavan,
  • P. M. Rojan

摘要

The assessment and prediction of dairy cattle longevity are essential for improving herd productivity and sustainability. Hence, modelling milking life (ML), productive life (PL), and culling age (CA) using first lactation traits supports better management of the farm. A dataset comprising 377 Holstein Friesian crossbred cows that had first calving between 2010 and 2020 at the University Livestock Farm and Fodder Research and Development Scheme, Mannuthy of Kerala Veterinary and Animal Sciences University, Kerala, India was analyzed and the obtained mean ML, PL and CA were 1003.13, 1444.96 and 2531.15 days respectively. Milking life and PL were positively correlated with lactation yield and number of completed lactations (NCL), but negatively correlated with age at first calving (AFC). Analysis of the data with accelerated failure time models by an 80:20 train-test split (hold out validation) method identified the log-logistic model with AFC and 305-day first lactation yield (LY1305) as optimal for ML (Root Mean Square Error (RMSE) = 648.34) and PL (RMSE = 889.41) prediction. For CA, the log-normal model using LY1305 alone showed the best fit (RMSE = 846.93). The classification of AFC to assess its impact revealed that cows calving between two and three years had the longest ML and PL. These findings highlight the significance of early AFC and high first lactation performance in extending dairy cow longevity in the tropics. The proposed models offer valuable decision-support tools for optimizing herd management using data-driven approaches.