Department of Animal Science, Cornell University, Ithaca, NY 14853.
L. Moraes Consultoria, Piracicaba, SP 13400-290, Brazil.
J Dairy Sci. 2024 Oct;107(10):7822-7841. doi: 10.3168/jds.2023-24387. Epub 2024 May 31.
Variation in forage composition decreases the accuracy of diets delivered to dairy cows. However, variability of forages can be managed using a renewal reward model (RRM) and genetic algorithm (GA) to optimize sampling and monitoring practices for farm conditions. Specifically, use of quality control charts to monitor forage composition can identify changes in composition for which adjustment in the formulated diet will result in a better match of the nutrients delivered to cows. The objectives of this study were (1) to assess the use of a clustering algorithm to estimate the mean time (d) the process is stable or in control (T) and the magnitude of the change in forage composition between stable periods (Δ) for corn silage and alfalfa-grass silage, which are input parameters for the RRM; (2) to compare optimized farm-specific sampling practices (number of samples, sampling interval, and control limits [Δ]) using previously proposed defaults and our estimates for the T and Δ input parameters; and (3) to conduct a simulation study to compare the number of recommended diet changes under the proposed sampling and monitoring protocols. We estimated the T and Δ parameters for corn silage NDF and starch and alfalfa-grass silage NDF and CP using a k-means clustering approach applied to forage samples collected from 8 farms, 3×/wk during a 16-wk period. We compared 4 sampling and monitoring protocols that resulted from the 2 methods for estimating T and Δ (default values and our proposed method) and either optimizing only the control limit or optimizing the control limits, the number of samples, and the number of days between sampling. We simulated the outcomes of implementing the optimized monitoring protocols using a quality control chart for corn silage and alfalfa-grass silage of each farm. Estimates of T^ and Δ^ from the k-means clustering analysis were, respectively, shorter and larger than previously proposed default values. In the simulated quality control monitoring, larger Δ^ estimates increased the optimized Δ, resulting in fewer detected shifts in composition of forages, a lower frequency of false alarms, and a lower quality control cost ($/d). Recommended diet reformulation intervals from the simulated quality control analysis were specific for the type of forage and farm management practices. The median of the diet reformulation intervals for all farms using our optimal protocols was 14 d (quartile [Q] = 8, Q = 26) for corn silage and 16 d (Q = 8, Q = 26) for alfalfa-grass silage.
饲草料组成的变化会降低奶牛日粮的准确性。然而，饲草料的可变性可以通过更新奖励模型（RRM）和遗传算法（GA）来管理，以优化采样和监测实践，适应农场条件。具体来说，使用质量控制图来监测饲草料组成，可以识别组成的变化，从而对配方日粮进行调整，使提供给奶牛的营养物质更好地匹配。本研究的目的是：（1）评估聚类算法在估计过程稳定或处于控制状态的平均时间（d）（T）以及稳定期之间饲草料组成变化幅度（Δ）方面的应用，这两个参数是 RRM 的输入参数；（2）比较优化的特定于农场的采样实践（样本数量、采样间隔和控制限[Δ]），使用先前提出的默认值和我们对 T 和 Δ输入参数的估计值；（3）进行模拟研究，比较根据提出的采样和监测方案下推荐的日粮变化次数。我们使用 k-均值聚类方法对从 8 个农场采集的饲草料样本进行分析，在 16 周内每周采集 3 次，以估计玉米青贮 NDF 和淀粉以及苜蓿-草青贮 NDF 和 CP 的 T 和 Δ 参数。我们比较了 4 种采样和监测方案，这些方案是由 2 种方法估计 T 和 Δ（默认值和我们提出的方法）以及仅优化控制限或优化控制限、样本数量和采样间隔天数产生的。我们使用每个农场的玉米青贮和苜蓿-草青贮的质量控制图模拟了优化监测方案的实施结果。k-均值聚类分析中 T^和 Δ^的估计值分别比先前提出的默认值短且大。在模拟的质量控制监测中，较大的 Δ^估计值增加了优化后的 Δ，导致检测到的饲草料组成变化较少、假警报频率较低、质量控制成本（美元/天）较低。模拟质量控制分析的推荐日粮重新配方间隔因饲草料类型和农场管理实践而异。使用我们的最佳方案，所有农场的日粮重新配方间隔中位数为 14d（四分位[Q] = 8，Q = 26），用于玉米青贮，16d（Q = 8，Q = 26）用于苜蓿-草青贮。
