BIOEPAR, INRA, Oniris, La Chantrerie, 44307 Nantes, France; Laboratory of Molecular Biology, Genomics and Bioinformatics, Department of Biology, Faculty of Nature and Life Sciences, University Hassiba Benbouali of Chlef, 02000 Chlef, Algeria.
BIOEPAR, INRA, Oniris, La Chantrerie, 44307 Nantes, France.
J Dairy Sci. 2019 Oct;102(10):9435-9457. doi: 10.3168/jds.2018-15484. Epub 2019 Aug 14.
The performance of dairy herds is affected mainly by factors related to cows' characteristics and herd management practices. However, these factors are interrelated, and as such, the estimation of their individual effect on the performance of dairy herds remains difficult. The aim of this study was to estimate the weight of these factors as well the interactions between them on the reproductive and economic performance of dairy farms. A stochastic dynamic model was used to simulate most physiological and management processes occurring on a dairy farm. A herd of 60 Holstein cows, with a milk yield of 8,000 L/cow-year, representative of French Holstein dairy herds, was simulated. A total of 216 scenarios were run by combining 2 levels of postpartum cyclicity resumption (average: 45 d, high: 75 d), 3 levels of 21-d conception rate of the herd (i.e., proportion of cows pregnant 21 d after insemination; low: 25%, average: 45%, high: 70%), 3 levels of probability of pregnancy loss until 120 d (low: 3%, average: 15%, high: 43%), 3 levels of sensitivity of estrus detection by the farmer (low: 20%, average: 50%, high: 90%), 2 alternative managerial goals (constant number of cows or constant volume of milk sold), and 2 types of management for the sale and purchase of animals (closed or open herd). The effect of each factor was estimated by sensitivity analysis. The parameter that had the greatest effect on reproductive performance was the sensitivity of estrus detection: a 10-percentage-point increase between the low and average levels and between the average and high levels reduced the calving interval by 16 and 5.7 d, respectively. However, the factor that had the greatest effect on economic performance was the 21-d conception rate: a 10-percentage-point increase between the low and average levels and between the average and high levels increased the gross margin by €62.2 and €22.3/cow-year, respectively. The pregnancy loss until 120 d had an effect on economic performance: an increase of 1 percentage point of this parameter decreased the gross margin by €2/cow-year. The other factors studied, and their interactions, did not have a major effect (low value of sensitivity indices). Closed herds or farms with a constant number of cows had economic losses of €58/cow-year compared with open herds or to farms with constant volume of milk sold. Altogether, our data suggest that, in a typical French dairy farm, farmers' efforts on estrus detection will be more profitable when associated with improvement of the conception rate of the cows.
奶牛群的表现主要受与奶牛特征和牛群管理实践相关的因素影响。然而，这些因素相互关联，因此，估计它们对奶牛群表现的单独影响仍然很困难。本研究的目的是估计这些因素的权重以及它们之间的相互作用对奶牛场繁殖和经济性能的影响。使用随机动态模型模拟奶牛场发生的大多数生理和管理过程。模拟了一个具有代表性的法国荷斯坦奶牛群，由 60 头泌乳奶牛组成，泌乳量为 8000 L/头年。通过组合产后周期性恢复的 2 个水平（平均：45 d，高：75 d）、牛群 21 d 受孕率的 3 个水平（即输精后 21 d 怀孕的奶牛比例；低：25%，平均：45%，高：70%）、妊娠损失概率至 120 d 的 3 个水平（低：3%，平均：15%，高：43%）、农民发情检测灵敏度的 3 个水平（低：20%，平均：50%，高：90%）、2 个替代管理目标（奶牛数量恒定或销售牛奶量恒定）和 2 种动物买卖管理类型（封闭或开放牛群），共运行了 216 种情景。通过敏感性分析来估计每个因素的影响。对繁殖性能影响最大的因素是发情检测灵敏度：低水平和平均水平之间以及平均水平和高水平之间的灵敏度增加 10 个百分点，分别使产犊间隔缩短 16 和 5.7 d。然而，对经济性能影响最大的因素是 21 d 受孕率：低水平和平均水平之间以及平均水平和高水平之间的灵敏度增加 10 个百分点，分别使每头奶牛的毛利润增加 62.2 和 22.3 欧元。至 120 d 的妊娠损失对经济性能有影响：该参数增加 1 个百分点，每头奶牛的毛利润就会减少 2 欧元。研究的其他因素及其相互作用没有产生重大影响（敏感性指数值较低）。与开放牛群或销售牛奶量恒定的农场相比，封闭牛群或奶牛数量恒定的农场每年每头奶牛的经济损失为 58 欧元。总的来说，我们的数据表明，在法国典型的奶牛场中，当农民在发情检测方面的努力与提高奶牛的受孕率相结合时，将会更有利可图。
