makes it possible to cover proximate composition controls for each batch of incoming raw materials, instead of a weekly or a monthly analysis on an average sample. The first advantage is to check the quality of incoming products to be sure they meet the supplier specifications and detect possible adultera- tion. Afterwards, if the product is compliant with quality ex- pectations, its composition will be stored in the formulation software. All data compiled between one week to one month are statistically analysed; this provides insight into the real variability of each raw material.
Crude protein variability Figure 1 shows the variability of the crude protein (CP) con- tent of a Brazilian soybean meal (SBM). The variability ob- served is around ± 2% CP. The feed mill carries out regular analyses of its SBM, but uses a single CP value for formula- tion: 46.8%. However, the actual analytical average is 46.4%. When the CP content of SBM is above 46.4%, there is an ex- cess of protein in the feed that may have an impact on feed cost but also generates counter-performance. When the CP content of the SBM is below 46.4%, there’s a lack of protein in the feed, causing a reduction in animal performance. The economic analysis in this study shows that in a feed mill that produces 100,000 tons of broiler feed per year, (with 25% SBM in the formula and with a cost per percent of CP at 5$ in the feed), the loss of earnings approximates $ 50,000 if you overestimate the protein level (+0.4%). This example illus- trates how powerful a good quality control plan with NIR can
be, if it’s applied to all the products used in the feed mill plant. Most of them can be handled through NIR technology (mainly all organic products), if the NIR calibrations are as ro- bust, accurate and efficient as possible. They should be vali- dated first with the environment of the feed mill (local raw materials, specific weather or industrial factors) and moni- tored frequently to ensure good performance. As an accuracy and robustness key indicator, the “ratio of performance to de- viation” (RPD), (which is the ratio of the standard error of pre- diction (SEP) to the standard deviation (SD) of the samples), represents the capability of the calibration to measure the nutrients with its natural range of variability. Thus, for a range of CP between 40% and 50% with a SD of 1.6%, the SEP must be below 0.8%. The higher the RPD is, the better the calibration will be.
Matrix calculation Wisium, the global premix and services brand of ADM ani- mal nutrition, has developed Equadvice, a dedicated matrix calculation software with more than 300 prediction models available for all species. Matrix calculation is a complemen- tary service to NIR analysis, which contributes to get the best raw material valorisation possible to optimise the nutrition approach and feed costs. Matrix calculation software provides relevant valorisation of all the types of raw materials used in the formulation software. This is compulsory for feed millers that are required to enhance the quality, efficiency and profit- ability of their products.
▶ ALL ABOUT FEED | Volume 29, No. 2, 2021 15
To gain some precious time and become more reactive, feed mills use NIR technology to empower the optimisation of their control plan.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36
