The animal nutrition and production market has been going through an interesting moment. On one hand, a domestic and foreign market demanding more and more Brazilian products, population and income with constant growth expectations, and new technologies that increase productivity and efficiency of costs and resources used. On the other hand, two years of pandemics and the conflict in Ukraine that made companies in the sector struggle to deal with the shortage of raw materials, the high prices of agricultural commodities, logistical barriers, exchange rate variation and high inflation that made the consumer market rethink (or think harder) some of its consumption habits.
All these challenges have been significantly impacting the increase in production costs, pulled mainly by grains such as wheat, soybeans and corn, which are fundamental to compose food and animal feed. In this moment of cost pressure, the competitiveness among the companies of the sector is immense, with increasingly narrow margins – most of them fighting over the reduction of the cents per ton produced.
In this scenario, the search for strategies that bring more efficiency to the production process, improve the use of available resources and can reduce costs (either by increasing animal performance with improved feed efficiency or by reducing raw material production costs) has become even more relevant. As we have seen in previous articles, there are several ways to do this, from the use of premixes, additives, and specific nutritional solutions to enhance performance objectives, to a greater focus on the formulation process and product reformulation.
Today, we’re going to talk a bit about technologies that, while not so new, become more accessible over the years and have substantial impacts on the animal nutrition and production process. In particular, let’s talk about a technology that brings analytical methods that can be used to make faster decisions to decrease variability and deviations that impact product specifications – consequently, a good amount of money can be saved and a consistent quality product can be offered to customers. Let’s talk about NIRS technology, which has enormous potential to improve the monitoring and control of processes involving feed production within factories.
NIRS technology has a wide variety of applications, being used on a large scale to analyze foods of different origins: cereals and their derivatives, seeds, finished products and premixes, animal inputs, liquids, granulates, powders, among many others. Of course, the idea here is not to explain the whole principle of electromagnetic radiation emission, chemometrics and data science – but to explain the main concept, its application in the industry and how formulators can benefit from its data.
What is NIRS technology?
In short, NIR (which stands for Near Infrared Reflectance) is a fast, high-precision technology that uses the natural principle of the emission of electromagnetic radiation to perform analyses to determine the chemical composition of food and other organic (and even some inorganic) samples. It is a vibrational spectroscopy technology that uses the energy of photons in the spectral range of 750 nanometers (nm) to 2500nm to measure the interaction of electromagnetic radiation with the vibrational motions of a molecular system – and thus measure its interaction with matter.
When using NIR, the electromagnetic wave that has been produced by equipment using this technology is partially absorbed and reflected by the sample of material being analyzed. The organic compounds present in the sample absorb this energy, which is transformed into the vibration of molecular bonds. From this moment on, the radiation that is not absorbed, that is, reflected, is analyzed and quantified.
The basic working principle of NIR can be defined as:
- NIR light is directed at the sample
- The light is modified according to the composition of the analyzed item
- The modified light is detected by its transmission and the flux that is reflected (reflectance)
- The modified spectrum, measured by differences between the amount of light emitted by the NIR and the amount of light reflected by the sample, is converted to information that quantifies the elemental composition of the sample being analyzed
This process of converting light to determine the components also integrates one of the key points of the NIR: the statistical and mathematical methods that process the data to obtain the results, starting, including, modeling coming from chemical data. We can translate these models into the famous NIR calibration curves, which are nothing more than mathematical models that generate equations that should represent very well (i.e. be statistically relevant) the prediction of a given chemical element for each sample component.
The variables are constantly studied, related to spectral and reference data (a process we call chemometrics) and are combined through multilinear regressions. From this, the equations of determination go through significance tests to determine the coefficient of determination (in statistics we call it R²) obtained in the validation and calibration steps. Remembering an important statistical premise: the R² will express the amount of data variance, so the higher the R², the better the determination equation fits the sample and better the calibration plans of the equipment being used.
Of course, there are other points of attention that must be taken into account in this calibration process: the standard error of the method, the standard deviation of the sample population, the diversity of sample composition, changes in external conditions that can impact on the measurement of results, among others. Exactly why this calibration process is so relevant to the accuracy of the instruments – many companies, such as AB Vista, Adisseo and Evonik, have NIR data reading platforms with calibration curves and determination equations already in a proprietary database. This approach turns out to be interesting for decreasing data variability and having a higher analytical relevance for determining chemical composition results, becoming more accurate when correlating a base obtained from wet analysis and spectral evaluations.
NIR Equipment Types and Benefits
There are several models, from several brands and that will use different analytical methods (or traditional and non-traditional composition prediction). However, let’s talk a little better about 3 types of equipment that are the most common in both the rural and livestock producers as well as in the animal and human nutrition industry today: the in-line, bench-top, and portable NIR.
Starting with the last one, portable NIR is gaining more and more space, especially in field use, by farmers, veterinarians, and breeders, for example. Many instruments come pre-calibrated and are excellent tools for quick analysis of raw materials such as silage and grass. They usually do not require complex sample preparation and deliver reliable results within minutes, apart from the fact that they are completely mobile, i.e. they can be used anywhere in the field and production quietly. The big disadvantage is that the spectral range they work with is smaller, so the calibration plans and on-board analyses are also limited.
For medium-sized factories, growing and even large organizations, for sure the most used NIR is the benchtop NIR. The concept of the bench NIR is similar to the portable NIR (and will be equal to what we will talk about later, the in-line NIR), the big difference is that it already works in a higher resolution and spectral range, going up to 2500 nm. It is usually on the bench of the chemical analysis laboratory, receiving samples of inputs and finished products for accurate and robust analysis. Thus, the accuracy of the results obtained in the bench NIR, in theory, will be higher and have greater predictive power. In contrast, the benchtop NIR is a more expensive instrument and, moreover, restricted to receiving the samples in the laboratory where it is located.
Finally, one of the big trends within the analytical prediction process is in-line NIR, which is installed directly in the process production line – either at the mixer, on the scale conveyor, or at the silo exit – to perform continuous measurements of frequently used inputs. This approach gives the formulator a much more accurate picture of process fluctuations and variability, bringing benefits such as:
- Trends in nutritional variations
- Calibration curves with larger sample populations
- Real-time raw material analyzer, enabling more efficient supplier qualification and inspection in a shorter period of time
- Decreased differences in assurance levels between formulated and produced product
- Smarter and faster decision making regarding feedstock composition levels, with data being generated by analyses that will arm the formulator at the very moment the product is being produced
Imagem retirada do vídeo “Multipoint inline NIR System – Control your entire process most economically “, disponível no canal buchilabequipment.
In fact it is a piece of equipment that has a higher investment and demands a more customized implementation process for each company and process. However, it usually brings a fast return on investment due to the possibilities of data analysis and cost savings that it brings to the production process.
Even though it is a new piece of equipment and still a high investment, it is always worth remembering the famous Moore’s Law: technology doubles its processing capacity every 18 months for the same price that was paid today. In other words, in all areas of technology three things end up happening: the digitalization of processes, the exponential growth of performance, and the decrease in price at the same time that the scale of use grows.
We have observed that many of the companies that use NIR still do not know the full potential of the tool and the investment made, and often use it only for measurement and analytical control of inputs (which in fact is essential) within the laboratory, but with data that does not go beyond the laboratory. It is worth remembering that modern quality is strategic, collaborative, and part of a culture that involves different areas and agents (which is why top-down company participation is crucial). We wrote earlier how quality and formulation (just citing an example) are connected. Within this same context and expanding the formulator’s own scope of work, we know that formulation will communicate with purchasing to qualify suppliers, with production for a more assertive PCP, with quality for variability analysis, and so on
Thus, the big challenge for companies is not only to use NIR to generate the data needed to monitor physical-chemical safety parameters and develop consistent and regular calibration – because this is essential. But also, from a set of structured data, allow the decision making between various areas of the company to be increasingly strategic and to bring an integrated look at the impact of these data for different activities. In summary: NIRS technology is becoming a management tool within companies for intelligent decision-making and in real time.
Application in the animal nutrition industry
NIRS technologies and equipment that use this technique are not new in the animal nutrition and production industry – on the contrary, they have been used for decades to determine bromatological parameters such as protein, moisture, ethereal extract, among others. What is observed in the industry more recently is the popularization of this equipment, that is, increasingly accessible tools that bring immediate results both in the control of validation parameters and in the qualification of raw material suppliers.
Besides already being applied to other types of analysis, such as mycotoxins, it is a methodology that does not require a complex sampling and preparation treatment, can be used in solid, liquid, pasty and grain samples, the results come out in a matter of minutes (if not seconds), and do not generate additional waste, thus being a more sustainable process. Moreover, with the popularization of tools for results management and analysis prediction (such as digestibility and amino acids), such as Labinfy, which allows a complete control of analytical results and statistical inferences from the NIR results, companies can make more complete, fast and powerful data analysis.
From the standpoint of the industry as a whole, the increased use of NIRS technology by more and more companies for the analysis of ingredients and/or finished products to know their basic constituents and contaminants has an important impact: it raises the required and delivered standard of quality of raw materials, with lower variability and better traceability, thus driving the culture of using data to develop more reliable strategies and results.
The impact on feed formulation
For formulators and product development teams, NIR becomes a great ally. We have seen the use of this technology in every type of feed company, from premix, commercial feeds, supplements, grains, and commodities, to the human food industry such as dairy products, ice cream syrup, and chocolate products. Whatever the purpose of use, the big impact of NIR for formulation is to enable greater accuracy and decrease the difference between formulated and produced product – taking away the faster monitoring of ingredient safety and quality parameters.
As we have seen in previous articles, NIR data aids in the adoption of increasingly efficient precision nutrition and formulation, driving the strategic use of data to uncover gaps, opportunities and insights. By observing a large discrepancy in the found result of a certain nutritional component compared to the average, the formulator can adjust their nutritional matrices and change their safety coefficients, thus ensuring that the formulation of the products is the most appropriate. By adjusting the nutritional matrices with the correct information, the optimization of the formulation can balance the nutritional requirements in a way that is more adjusted to reality and bring a higher financial performance and animal performance
The latest NIR techniques make it possible to perform the analysis on your own product production line – meaning you can update your nutritional matrices and reformulate your products instantly. Surely the impact this has on companies, financially and productively, is enormous, taking precision formulation to maximum levels. Formulating with the matrix of the ingredient that is going into the mixer or coming out of the bin – already adjusted for standard deviation and averages of past sample populations – is enabling the formulation to deliver a precision calculation with low variability and at an even lower cost; not to mention the gain in precision nutrition to meet the nutritional requirements of the animals.
Regardless of the equipment and the type of company, we highlight 4 points of attention that formulators and quality should have when using NIR as a formulation aid:
- Select representative samples of raw materials from more than one supplier to evaluate costs and availabilities
- See together with the lab the NIR results and compare them with good analytical results inferred by conventional reference methods and wet chemistry, to make sure that the results found are always reliable
- Implement with other areas of the company a collaborative analytical routine, that is, that the areas have access to the data that impacts the results of the company as a whole
- Revisit your nutritional matrices from time to time to ensure that the values entered are always up to date and appropriate to the standards and suppliers used in production
Of course there are other points of attention that need to be paid attention to, from the sampling process, laboratory activities that respect the parameters and self-monitoring programs, to regular calibrations and equipment maintenance. These are just a few practical tips that formulators can benefit from by using NIR and getting closer to laboratories and quality areas.
Every day we see more companies adopting and implementing NIRS technology to support the processes and records of the verification parameters provided in their quality controls. As we have seen, it is not a new technology, but it has had a significant growth driven by advances in mathematical and scientific models that enable an increasingly accurate and rapid detection capability of nutritional values. And like any technology that takes scale, the trend is that new equipment and suppliers are on the market and make prices increasingly affordable.
Today, we see many formulators and nutritionists that have in NIR an important ally for the company’s objectives to be achieved in a more efficient way and that brings real results to the production. Whether analyzing ingredients to determine if the levels evaluated are in accordance with those reported by the supplier, to update their nutritional matrices, or even to evaluate the guarantee levels of the product produced, NIR has been bringing great facilities for companies, leaving them more competitive and with better products.
However, we still see that there are huge opportunities with NIR. The process of expanding the use of this equipment and updating the technologies of proprietary laboratories enables faster results for increasingly assertive decisions. With fast and continuous analysis, the tendency is for companies to be able to adapt more quickly and, consequently, have greater operational productivity and competitiveness in the industry.
Precision formulation and nutrition are becoming more and more accessible to companies of different types and sizes. And for sure, NIRS technology will be a great tool for this.
