饲料厂权衡敏感性如何影响饲料配方

我作为在爱荷华州的饲料厂工作的商业营养师学到的首批课程之一是，我的计算机生成的配方在生产配方中未100％转移。

这是一个常见的问题，即使是经验丰富的营养学家（通常是作为顾问）的问题，他们忘记考虑了每个饲料厂的能力准确测量饲料优化公式中规定的每个数量。

The case of adding corn

Here is a common example. Most animal feed formulas balance on corn — that is, after all ingredients are accounted for at the required amount, the remainder is rounded up to 100% by adding corn. So one formula may contain, let’s say, 1,005.3 pounds of corn per 2,000-pound ton.

In my experience, very few (large and/or sophisticated) feed mills can accurately weight 5.3 pounds of corn as the equipment used is often within the 5- to 10-pound error range. That means each step in adding more corn can end up 5 to 10 pounds different as automation is not always precise nor accurate, and such large volume ingredients are not added by hand.

Some highly automated systems have the capacity to learn how each raw material behaves so that they keep correcting the final difference (for example, from the time the auger stops until the last kernel of corn hits the scale there is a time-weight gap). However, with each new batch of corn (or any other grain) this learning curve starts anew as dry matter affects the behavior of each raw material.

How to correctly hand-add micro-ingredients

另一个常见的例子，尤其是在小饲料厂中，是手工添加的微型含量。如果营养学家设计了一个要求1.15磅的L-硫代氨酸的配方，但是实际量表的灵敏度为0.1磅，那么几乎不可能准确地考虑到5/100^th’sof a pound difference.

此外，营养学家还应尝试在公式到达生产地板之前最大程度地减少人为错误的因素。例如，如果将数量向上或下四舍五入到合理的数量，即使量表具有如此微小的敏感性，手工添加所有这些微观的工人也会减少错误的错误。

If a formula contains 10 micro-ingredients and the worker needs to prepare 100 such micro-batches per day, then each step towards simplifying the process will also minimize the possibility of a major error. It is not unheard of to misplace the decimal point and instead of 11.11 pounds of L-lysine to add 1.111, and so on.

平衡middle-level raw materials like fish meal

还有一个中间区域，某些成分的数量不足以证明自己的自动垃圾箱合理，但同样，每吨的数量非常重要。

这样的例子就是鱼餐。包含鱼粉的大多数完整配方的包含率在2.5％至5.0％之间。实际上，这将从一个到两个标准袋（每个袋子都有50磅的鱼餐）。

这没什么意义，在饲料厂的水平上造成了很多恼火，必须重22或47磅的鱼粉。而且，从营养的角度来看，我们知道，从动物的角度来看，这种差异在22至25或47磅和50磅之间的差异相对微不足道。因此，知道这些专业成分如何进入饲料厂并相应调整纳入水平是值得的。

Reason behind ingredient discrepancies

要问的一个合理的问题是，为什么营养学家不考虑此类细节，这些细节通常由饲料工厂经理纠正。原因很简单。饲料产品不太可能由计算机软件程序制定以满足特定的营养水平。

对于没有负责任的营养价值的成分，以预定的固定值输入公式（例如，按照其供应商指​​示，以2磅/吨添加的益生元）输入公式有一些例外。但是，大多数原材料都是制定的，以满足确切的养分规格，从而导致奇异的最终纳入水平。

这是另一个示例。说，营养学家将总赖氨酸作为营养素（一个简单的例子）制定了饲料，并将进料级赖氨酸的最大限制设置为原材料（例如L-赖乙烷HCl）。如果赖氨酸的唯一其他主要来源是大豆餐，那么饲料配方计划的最终量为大豆粉，这是不可避免的（几乎在所有情况下）。因此，如果使用6磅/吨的L-赖氨酸HCl，并且仅使用玉米和大豆餐（例如44％粗蛋白），那么玉米和大豆餐的金额分别为1,881和116磅/吨，忽略了在我们的示例中，为了简单起见，通常存在于饲料中的任何其他成分。

Now, in practice, the feed-grade lysine will be added at 6 pounds by hand, but at best we will have 1,880±5 and 110±5 pounds of corn and soybean meal being added by the automated system. Not a huge difference, but not exactly the formula the nutritionist designed to begin with.

Formulate feed for specific feed mill abilities

Modern feed mills can take a feed formulation from the nutritionist and adjust it automatically to specific feed mill abilities. As mentioned before, this was already done and continues to be done in small mills by the feed mill managers.

But this does not need to be the case, as it is best for the nutritionist to have a crystal-clear picture of what is being produced. This might help in case problems arise in the future.

对于穰mple, a small deviation from the prescribed amount of calcium carbonate in a layer formula may end up causing excess numbers of broken eggshells (if less) or liver damage (if more) — and in any other formula calcium carbonate is often being used as a filler with relative impunity!

因此，现代饲料公式程序已经安装了代码来说明饲料厂的能力。这些是输入一次的数字，然后该程序制定了基于原材料预定步骤来满足营养的数字。为此，可能需要一些较小的养分规格灵活性。

In closing, I would like to mention one more cause of distress between nutritionists and feed mills. A formula that calls for 2.5 pounds of wheat/ton (just because wheat is available at the mill) to meet the least-cost criterion is of very little if any significance to cost or animal performance. But it makes a big difference during production of said formula for the feed mill personnel.

—Ioannis Mavromichalis, PhD