Feed & Nutrition

Formulation, optimization of a poultry feed and analysis of spectrometry, biochemical composition and energy facts

February 13, 2024

795

Abstract

Poultry feed plays a key role in the success and economic profitability of poultry production. When formulating an effective feed, its cost and nutritional quality to cover the essentials of poultry must be taken into account. For this reason we tried to make an optimal poultry feed in quality and price, so we collected samples for this study The collection was organized to cover as much variability as possible related to seasons, suppliers, origins. Analysis must be done to discover the biochemical compositions and nutritional energy of the primary trades that lead to a formula to produce poultry feed, this formula is realized by linear programming technique to manufacture feed for three categories of poultry (starter (0–4 weeks), grower (4–18 weeks) and layer chickens (18 weeks to culling)). we analyze the samples by the near infrared spectrometry (NIRS), and to confirm the spectrometric results, we do deferent chemical analysis. Finally we obtain a database that guides us to interpret the optimal percentages for the manufacture of poultry feed.

Keywords

Formulation

Poultry feed

Near infrared spectrometry

Biochemical composition

Energy facts

1. Introduction

After air and water, food, as one of the most complicated sets of chemical elements, is the third most important thing for living (Quddusi, 2018). A feed is a substance that must provide the humans or animal with the energy and elements necessary to keep it alive and therefore cover its maintenance needs. For livestock, the feed must also provide enough nutrients to meet production needs (eggs or meat). The feed can be in different forms: raw materials, compound feed (a mixture of at least two raw materials), complete feed (compound feed which, because of its composition, is sufficient to cover the daily requirements) or supplementary feed (designed to supplement the raw materials given to the animal, for example cereals). The poultry industry has achieved tremendous progress in its production system during the last 50 years through improvements in genetic makeup, proper management and advancements in nutritional science (El-Tahawy et al., 2017; Gado et al., 2019).

Meat production requires large amounts of inputs and is therefore considered in many countries and cultures as a high-value food product. However, due to rising wages and world population, the global demand for animal protein continues to grow (OECD-FAO 2021), which creates many challenges to increase and suffice the needs of poultry (Altmann and Rosenau, 2022; Khan et al., 2022; Bryant et al., 2022; Abdelli et al., 2021; Kpomasse et al., 2021; Singh and Kim, 2021; Alhotan, 2021; Aboah and Enahoro, 2022).

Historically, the commercial poultry industry experienced tremendous changes in growth of all phases from the hatchery to broiler and layer farm practices along with meat and egg processing technological advances for long distance retail distribution (Ollinger et al., 2005; Diaz-Sanchez et al., 2015). As consumer demand has grown, the volume of poultry meat and eggs produced has also expanded to match this rise in retail demand (Erinle and Adewole, 2022; Madkour et al., 2021; Gürbüz and Korkmaz, 2022). This rapid expansion in commercial poultry production has required and will continue to depend on advances in bird genetics, nutritional management, processing technologies, and food safety (Havenstein et al., 1992; Havenstein et al., 2003; Bolder, 2007; Vandeplas et al., 2010; Cox et al., 2011; Chao et al., 2014; Ricke, 2017).

Current global predictions assert that by 2025 poultry meat will have the highest level of production and consumption, over beef, veal, pork and sheep (OCDE/FAO 2016). Increased poultry consumption may be attributed to the fact that chicken meat is an affordable and accessible source of protein with a low-fat content, and that there are few religious or cultural barriers related to its consumption. relative to other sources of dietary protein, chicken meat is also a low greenhouse gas emission food (Caro et al., 2017).

Feed additives are generally considered materials used to enhance the effectiveness of nutrients and exert their effects on improving poultry performance (Ashour et al., 2020; FARAG and ALAGAWANY, 2019; Khafaga et al., 2019). In addition, low levels of additives in poultry feed can contribute to an increase in the production of poultry protein for human consumption, which in some instances can decrease the cost of animal and poultry production (Ismail et al., 2021; Ismail et al., 2020; Johnson et al., 2019; F.M. Reda et al., 2020; F.M. Reda et al., 2020; F.M. Reda et al., 2020).

In order to calculate the cost of manufactured feed, it is necessary to choose what price to attribute to the raw materials produced on the farm. The method of calculating the price of manufactured feed will depend on the reasoning and objectives of the farmer. Does he want to maximize the incorporation of self-produced raw materials in the feed (use of production costs or cost price), or does he want to compare the interest of selling his raw materials or incorporating them in the feed, and differentiate the economic efficiency of the breeding workshop and the workshop. The success of a quality feed formulation depends strongly on a good knowledge of the physicochemical characteristics of raw materials (Ponka et al., 2016) and the effect of partial replacement of materials raw with other materials raw on the production efficiency and meat quality in broiler chickens (Biesek et al., 2022; Santos et al., 2022; Nasir et al., 2022; Kogut, 2022; Peng et al., 2022). For the present paper, the main objective is investigating an optimal poultry feed and analyze biochemical compositions and nutritional energy of the primary trades that lead to a formula to produce poultry feed and analyze the samples by the near infrared spectrometry (NIRS) and chemical analysis, to optimize percentages for the manufacture of poultry feed.

2. Formulation of poultry feeds

The purpose of feed formulation is to provide animals with a consumable product whose characteristics allow, under the given rearing conditions, to obtain a meat or egg production of meat or eggs with the highest profit (Larbier and Leclercq, 1992). To achieve this, it is important to know the consumer’s needs and to provide him with what he really needs, in the necessary quantity, in an appropriate form and at the right time.

The formulation of feeds consists in making, at a lower cost, palatable mixtures of feeds, with characteristics that facilitate their manufacture, handling and conservation and meet the needs of animals (maintenance, growth, production, reproduction) from available raw materials. It involves calculating the proportions of raw materials to be included in the manufacturing process in order to meet consumer requirements and production objectives

2.1. Steps in the formulation of poultry feed

2.1.1. Determining nutritional requirements

Nutritional requirements are influenced by genetics, sex, live weight, physiological stage, appetite and environmental factors (temperature, density, etc.). Nutrient requirements can be defined as the amount of nutrients needed to optimize a production factor, such as growth rate or feed conversion. Growth rate, meat or egg production performance, muscle or egg formation, fat or nutrient deposition in organs, feed intake are all characteristics that need to be defined before determining the requirement. Determining the requirement for several essential nutrients is therefore critical to the success of the formulation. These include metabolizable energy, crude protein, amino acids, calcium and phosphorus (…). Traditionally, nutritional requirements for poultry are indicated by INRA and NRC (INRA (Institut National de la Recherche Agronomique) 1989; NRC (National Research Council) 1994).

2.1.2. Determination of the nutritional values of available ingredients

For the need of feed formulation, the realization of a database of chemical composition, physical characteristic and digestibility of ingredients suitable for use in animal feed is necessary. There are databases of chemical compositions of poultry feed ingredients (NRC (National Research Council) 1994; FAO, INRA, CIRAD, AFZ 2022) which are continuously updated. However, the raw materials used for feed formulation depend on the country or region by depending on the potential of the ingredients (Brah and Houndonougbo, 2016). In this context, it is recommended that the formulator build his own database according to the available feed resources (Kaushik, 2000).

The number of nutrients to be considered for formulation varies, but the most used for formulation and performance monitoring are metabolizable energy, crude protein (Brah and Houndonougbo, 2016), amino acids (Sterling. et al., 2005), amino acids, calcium and phosphorus (Kaushik, 2000). These nutrients should be of the same nature as those considered in the definition of the nutritional requirements of the birds to be fed. In addition to the chemical composition and nutrient values of the ingredients, knowledge of the threshold of incorporation of each ingredient and the factors of variation in nutrient value are critical to the success of the food formulation. The maximum and minimum limits of each ingredient must be known in order to avoid toxicity, dietary imbalance, interference with other nutrients, under consumption due to in appetence or even environmental pollution after excretion through urine and droppings (Brah and Houndonougbo, 2016).

Several ingredients are mixed to formulate the poultry feed, it is important to take into account the variation factors related to the mixture; specifically, the sum of anti-nutritional factors that may be present. Indeed, their presence reduces the digestibility and absorption of nutrients (Brah and Houndonougbo, 2016) inhibit enzymes, cause liver and intestinal damage, or lead to stunted growth in children (Pimpukdee et al., 2004) or cause growth retardation in chickens (Miazzo et al., 2000).

2.1.3. Formulation methods

This step involves making calculations to determine the combination of different ingredients that best covers the recommended requirements for the animal category. Efficiency aspects are integrated in some methods to obtain a feed at lower cost

2.1.4. Empirical methods of formulation

These methods are based on the experience of the trainers in proposing a mixture of ingredients that needs production, and to readjust these proportions according to the observed performances and the evolution of the raw material market. The incremental choices simplify the complexity of the formulation requirements by focusing on a limited number of focusing on a limited number of criteria to be optimized. Companies that base their formulation approaches on intuitive experience and modification of formulas tend to perceive the volatility of raw material prices as a threat to their as a threat to the profitability of their production and often do not seize the opportunities it can offer in terms of competitive advantage

2.1.5. Manual methods of feed formulation

•
Pearson square method

A method determines the proportion of two ingredients to be mixed to meet the requirement of a nutrient. It is performed with the following Pearson square (Fig. 1) (Lee, 2009).

In X the desired solution: the nutritional needs to be satisfied. In A and B the nutrient contents of the two sources of ingredients I₁ and I₂ to satisfy the nutritional need X; C represents the difference between A and X without taking into account sign; it is the share of ingredient I₂ in the mixture; D represents the difference between B and X without taking into account sign; it is the share of ingredient I₁ in the mixture.

The proportion (%) of ingredient I₁ containing nutrient A is obtained by(1) $\frac{C}{C + D} \times 100$

The proportion (%) of ingredient I₂ containing nutrient B is obtained by(2) $\frac{D}{C + D} \times 100$

•
Method of simultaneous algebraic equations

This is an alternative method to Pearson’s square method using a simple algebraic equation (Afolayan and Afolayan, 2008). It allows us to satisfy a nutritional need by combining two ingredients (Omidiora et al., 2013) through a system of equations (Lee, 2009).(3) ${\begin{matrix} a X + b Y = n e e d t o b e s a t i s f i e d \\ X + Y - 100 o f n e e d t o s a t i s f y \end{matrix}$

X and Y are the unknowns of the equation and represent the percentages of the two sources of ingredients to be mixed; a and b are the nutrient contents of the two respective ingredients. This is a simple method. It can be used by farmers (Omidiora et al., 2013). It may have several ingredients to satisfy a need (Afolayan and Afolayan, 2008). Its limitations are that it can only satisfy one need and cannot be used in a formulation with several needs to be satisfied (Abd Rahman et al., 2010).

•
Matrix method

This method finds the solution leading to covering two nutritional requirements with two ingredients (Omidiora et al., 2013). It can even be used for several requirements with more than two ingredients (Afolayan and Afolayan, 2008).

Roush et al. (Roush, 1982) used it to meet the broiler’s need for energy, protein, calcium and phosphorus by mixing corn, soybean meal, monocalcium phosphate and limestone (Roush, 1982). The equation system he used is as follows:(4) ${\begin{matrix} a_{11} X_{1} + a_{12} X_{2} + \dots \dots \dots \dots \dots . . + a_{1 n} X_{n} = b_{1} \\ a_{21} X_{1} + a_{22} X_{2} + \dots \dots \dots \dots \dots . . + a_{2 n} X_{n} = b_{2} \\ \cdot \\ \cdot \\ \cdot \\ a_{n 1} X_{1} + a_{n 2} X_{2} + \dots \dots \dots \dots \dots . . + a_{nn} X_{n} = b_{n} \end{matrix}$

With the nutrient content, X the proportions to be found and b the requirements to be met. It is an efficient method to define the desired energy level and other nutrients with the algebraic equations (Abd Rahman et al., 2010). However, it has the disadvantage of being complex and time-consuming to solve if the nutritional requirement exceeds two elements.

•
Trial and error method

This method is the most popular and used for the formulation of poultry feeds (Afolayan and Afolayan, 2008). It satisfies the poultry requirement by manipulating the nutrient values of the ingredients according to the set percentages (Omidiora et al., 2013). It can be done manually or by using spreadsheets such as Excel, Lotus123 or Quattro pro (Abd Rahman et al., 2010). This method can be used to meet all the needs of the chickens, but has the disadvantage of being tedious and time-consuming to arrive at a satisfactory solution (Afolayan and Afolayan, 2008).

2.1.6. Mathematical programming methods

•
Linear programming

Linear programming is a statistical method for choosing, allocating and evaluating limited resources and several constraints to obtain a linear algebraic function. It determines how to get a result in the form of a mathematical equation system (Olorunfemi, 2006). It minimizes the cost of the food by balancing the percentage, nutritional value and constraints of the ingredient used in the formulation (Al-Deseit, 2009).

The principles of least-cost linear programming with constraints are given by the following equations:(5) $Z = \sum_{i = 1}^{n} C_{i} X_{i} (for minimum costs)$

(6) $\sum_{i = 1}^{n} a_{i} X_{i} \geq B_{i} that X_{i} \geq 0 (for constraints)$

With Z: total cost of the food; Ci: the unit cost of the ingredient reported per kg; Xi: the quantity of ingredient i in the food in kg; ai: the nutrient values of the ingredient and Bi the levels of requirement to be satisfied for each ingredient (Olorunfemi, 2006; Al-Deseit, 2009; Almasad et al., 2011; Bhagat and Bajaj, 2014). Thomson et al. used the Solver function in Excel to design a spreadsheet for teaching and formulating poultry feeds at low cost by linear programming (Thomson and Nolan, 2001).

•
Non-linear programming

The chemical composition tables of the ingredients only present the average values of the nutritional parameters. They do not represent the totality of the samples (Peña et al., 2009). While linear programming can be used to optimize a food formula, it does not take into account the variability of ingredients (Guevara, 2004 ).

It is more used to describe the interdependencies that exist between the level of incorporation of an ingredient or the relationships between the nutritional values (energy, protein, calcium…) of a food and the induced performance or profit margin generated in the form of a mathematical function (Heydari, 2014). Afrouziyeh et al. (2010, 2011) used the Excel solver to optimize egg production by non-linear programming and to determine the relationships between egg weight and energy density of the layer feed during phase 1 (24 to 32 weeks of age) and phase 2 (32 to 44 weeks of age) (Afrouziyeh et al., 2011).

•
Multi-objective programming

Multi-objective programming was used in 1983 to address nutritional imbalance in human nutrition (Abd Rahman et al., 2010). In animal feeding, multi-objective programming is used to take into account the least cost formulation and nutritional imbalance of feeds (Peña et al., 2009).

The model used in multi-objective programming is comparable to that of linear programming by taking into account the dietary imbalance (Abd Rahman et al., 2010). The variation in nutritional requirement is modeled by (Zhang and Roush, 2002):(7) $\sum_{i = 1}^{n} V_{ij} X_{i}^{2}$

Where: V is the variance of nutrient j in ingredient i; Xi the percentage of ingredient i in the food. When many ingredients are used, some percentages become small, hence the use of the variance of the nutrients and the square of the ingredients (Zhang and Roush, 2002).

Zhang et al. (Zhang and Roush, 2002) used the multi-objective programming technique to take into account the constraints related to the variation in nutritional requirements between broilers by finding an optimum in the group

3. Optimization of a poultry feed

In order to elucidate the process of integrating the prices of raw materials in the optimization of food formulas optimization of feed formulas, we considered the case of feed formulation for a starter, grower and layer pullet. This section presents the nutritional constraints considered, the raw materials and their incorporation constraints, as well as the databases of their characteristics

3.1. Nutritionnel contraints use

In poultry farming, feeds are often formulated according to the age of the animal and taking into a large number of quality criteria to be optimized. These feeds are considered complete when their formulation allows them to be administered to maintain the animals or their production without requiring any other food than water.

3.2. Raw materials used

Foods can be classified according to their particularities, namely those that provide energy, sources of protein, calcium and phosphorus and finally those that provide other minerals, trace elements and vitamins. According to their nature and the way they are obtained, the raw materials used in poultry feed can be classified into seven groups:

•
Cereals: The basic constituents of poultry feed and rich in starch, their use is limited only by the need to maintain the energy-protein balance. use is limited only by the need to maintain the energy-protein balance. This category includes corn, wheat, barley, rice, rye, triticale and sorghum.
•
Protein crops: In this group we find seeds rich in protein and fat, including legumes such as soybeans, peas, peanuts, alfalfa among which legumes such as soybeans, peas, peanuts, alfalfa, sunflower and rapeseed occupy a prominent place. Sunflower and rape occupy a place of choice.
•
By-products of the cereal industry: Resulting from the manufacture of cereal flours, This category includes wheat bran, cracked wheat, wheat remolding, and rice and corn byproducts. With new cereal by-products such as wheat gluten, corn gluten feed and distillers dried grains with soluble are emerging.
•
Oil cakes: By-products of the vegetable oil extraction industry, they are low in fat, especially when they come from a low in fat, especially when they come from a solvent extraction process. They contain high proportions of proteins, which is what makes them so interesting. These proteins are of unequal value depending on the plant species of origin. In this category, we find soybean cake, rapeseed cake, sunflower cake and palm kernel cake.
•
Vegetable oils and fats: They provide fatty acids and metabolizable energy. These oils also contribute to the structuring of the mixtures and reduce dust emissions. They have therefore, beyond their nutritional functions, functional abilities. Depending on the region, palm, sunflower, rapeseed, peanut and soybean oil.
•
Animal meal: Animal meals include all by-products of the meat, fish and milk industries. This category mainly includes fishmeal, which has an excellent nitrogen content and is sought after for their high content of essential amino acids and mineral content.
•
Mineral concentrates, amino acids and vitamins of biosynthesis: In order to compensate for probable deficiencies of mineral elements, amino acids and vitamins in traditional raw materials, the food formulation industry resorts to additives concentrated in these elements. These are either mineral rocks and salts such as calcium phosphate, sodium chloride salt or even amino acids and vitamins of biosynthesis. The addition of these concentrates in the formulation is very limited and can be in the form of pre-mixes already formulated.

3.3. Incorporation constraints

Some raw materials may have deleterious effects on animal health or on a property of the formulation when their incorporation thresholds in the formula do not respect certain limits. In accordance with the recommendations of INRA and DSM (INRA (Institut National de la Recherche Agronomique) 1989) we considered the incorporation thresholds mentioned in Table 1. Given the lack of compromise on these incorporation thresholds, the incorporation limits may vary from one from one organization to another.

Table 1. Thresholds for incorporation of raw materials (INRA (Institut National de la Recherche Agronomique) 1989).

Raw materials	Lower limit	Upper limit
Empty Cell	Thresholds for the incorporation of raw materials in%
Corn grain	20	40
Wheat grain	0	40
Peas	0	10
Soybeans	0	8
Rapeseed	0	8
Sunflower oil	0	4
Rapeseed oil	0	4
Sodium chloride salt	0.1	1

4. Analysis of the optima obtained

For the present study are 9 raw materials were selected, the analysis connected to the simples are based on spectrometry (Fig. 2), biochemical and energy fact which are detailed in (Fig. 3). The restriction to these 9 raw materials was imposed by the availability of information on their compositions mentioned in Fig. 3 are those describing the composition and nutritional value of raw materials for farm animals.

4.1. Origin of the samples

The samples for this study were collected in Morocco and by different countries that export the primary trades (Brazil, Argentina, and America) in 2021 and 2022. The collection was organized to cover as much variability as possible related to seasons, suppliers, origins. Nearly 300 samples were recovered, covering the following categories: – cereals (corn, wheat, rice, soya, and rapeseed) – cereal by-products – oil cakes (sunflower, rapeseed, and soya) – oils- miscellaneous.

4.2. Analysis

The samples were ground with a Retsch ZM200 mill with a 1 mm grid. The analyses carried out were dry matter (DM) by oven drying, total mineral matter (TM) by calcination at 550 °C, insoluble minerals (InsCl) by attacking the TM with hydrochloric acid, total nitrogenous matter (TNM) by the Kjeldahl method, fat (MG) by extraction with petroleum ether (after acid hydrolysis if necessary), crude cellulose (CB) by the Weende method, starch (AMI) by polarimetry and total sugars (SUC) by the Luff-Schoorl method. The methods used correspond to the AFNOR standards (AFNOR 1999) for animal feed. The results reported in this document are always expressed on a dry matter basis. The near infrared reflection spectrum (NIRS) of the samples was acquired on a BRUKER MPII spectrometer, between 400 and 2500 nm, in quartz cups. Each sample was read twice (2 cell fills) and the spectra were averaged

•
Use of Near Infrared Spectrometry (NIRS)

Near infrared spectrometry (NIRS) is an analytical technique based on the absorption of light (wavelengths from 800 to 2500 nm) by organic matter. As the absorption depends on the nature of the molecular bonds of the sample, it is related to its chemical composition. The technique requires calibrations for each type of raw material, which is quite a lot of work. When applied to the samples in our survey, SPIR allowed the effective prediction of the chemical composition of the samples in most cases. Fig. 2 presented spectra using Near Infrared Spectrometry (NIRS) for the raw materials and products obtained

4.3. Results and discussion

The composition of macronutrients and micronutrients varies according to the raw materials (Table 2 and Fig. 3)

Table 2. Biochemical composition% (g/100) and energy value (kcal/ kg) of different types of raw materials.

Empty Cell	Corn	Soft wheat	Soybean	Soybean cake	Canola cake	Wheat middlings	Alfalfa 17–18%	Wheat bran	Rice bran	Salt	Limestone	Canola oil
Variables	X1	X2	X3	X4	X5	X6	X7	X8	X9	X10	X11	X12
Energy kcal/kg	3303	3200	3868	2244	1564	2150	927	1930	3422	0	0	8964
Protein%	7.35	11.35	33.9	45.63	34.6	15	14.1	15.42	12.9	0	0	0
Moisture%	14.12	11.61	11.1	11.49	12.1	11.8	10.5	13.3	7.2	0	2.5	0
Fat content%	3.66	1.1	18.2	1.19	2.6	3.9	2.7	4.81	16.1	0	0	99.6
Ash%	1.66	1.5	4.2	6.76	6.2	4.2	12	5.2	8.8	0	0	0.2
Cellulose%	1.86	1.18	9.2	3.73	12.1	7	27.4	10.05	7.3	0	0	0
Starch%	63.17	60.3	19.1	4.1	0	31.8	2	19.55	12.2	0	0	0

Protein contents (Fig. 3(a)) vary from 7.35 (corn) to 45.63% (soybean cake). Soybean, soybean cake and canola cake are the richest raw materials in protein with contents of 33.6, 45.63 and 34.6% respectively. The protein content of soybean cake is higher than the value of 39.38% found by Ponka et al. (Ponka et al., 2016) in soybean cake in Cameroon. The protein content of soybean is higher than those available in the FAO food composition table (FAO 2022). These high protein contents can be attributed to the soybean variety used. For wheat bran, the protein content (15.42%) is slightly lower than the current data (Ngom, 2004). Indeed, the protein content may depend on the efficiency of the hulling process, which prevents whole or broken kernels from ending up in the wheat bran, which would contribute to a lower protein content.

The fat contents (Fig. 3(d)) vary from 1.1 to 18.2% (rapeseed oil contains 99.6%). Soybean and rice bran are the richest raw materials in fat content with 18.2 and 16.1% respectively. The fat content of soybean cake is lower than the values found in the different types of soybean cake (Ponka et al., 2016). These low-fat contents of soybean cake can be explained by the fact that the oil extraction of soybeans was optimal during the pressing process. The fat contents of rice and wheat bran are respectively higher than the values found in rice and wheat bran in Senegal (Mpouok, 1999). This could be due to the fact that sprouts were found in the rice and wheat bran during deshelling, which would contribute to the higher fat content.

Starch contents (Fig. 3(g)) vary from 4.1 to 63.17%. Corn is the richest in starch with a content of 63.17%. The carbohydrate (starch) content of corn is close to the value (73.3%) found in French corn (Bastianelli et al., 2008). These contents are relatively high, which is appreciable given that carbohydrates are the main source of energy necessary for basic metabolism and the functioning of the nervous system (Ponka et al., 2016). Some carbohydrates are involved in the formation of cartilage, nucleic acids, mucus, glycoproteins and immunoglobulins in organisms (Sguera, 2008).

The ash contents (Fig. 3(e)) vary from 1.5 to 12%. Alfalfa is the raw material that contains more mineral matter. The ash content of rice bran is lower than that (24.94%) found in Senegalese rice bran (Ngom, 2004). Alfalfa 17–18% is richer in ash compared to the other raw materials analyzed. This also reflects a high content of mineral salts, which are essential elements for the proper functioning of the body, and the growth of poultry.

The cellulose contents (Fig. 3(f)) vary from 1.18 to 27.4%. Alfalfa and canola cake contain the highest cellulose contents of 27.4 and 12.1% respectively. The cellulose contents from wheat bran and alfalfa 17–18% are higher than the value 8.9% found in Senegal (Mpouok, 1999). This high cellulose content of wheat bran can be attributed to the variety used. Unlike other cereals, the cellulose content of rice bran is comparable to the value found in rice bran from other countries (Huart, 2004).

Finally, the moisture contents (Fig. 3(c)) vary between 7.2 and 14.12%. The energy values of the samples (Fig. 3(a)) vary from 927 to 3868 kcal/kg. The energy values are comparable to those found by (KY et al., 2020). In sum, the high-energy value of some of the raw materials analyzed may be explained by the fact that they have a high fat content. However, in animal nutrition, a feed is said to be energy-rich when it contains a high carbohydrate content (especially starch). Thus, the energy foods for monogastric animals are cereals (corn, soy, wheat, rice, etc.). On the other hand, poultry diets are mainly composed of a mixture of several feeds such as cereals, soybean cake, animal product derivatives (fishmeal, bones, shells, etc.), fats, vitamin and mineral premixes. All these feeds provide the energy and nutrients necessary for the growth, reproduction and well-being of the animals (Ponka et al., 2016).

•
Presentation of optima obtained

The recommended intakes are explained for each nutrient in the form of minimum and/or maximum concentrations of the nutrient in the food (% nutrient). They are calculated based on a daily consumption a priori related to the energy concentration of the food in order to cover the daily requirements. For essential nutrients such as essential amino acids and vitamins, the recommendations set minimum incorporation levels necessary to achieve production objectives. In the particular case of vitamins, for which uncertainties are very important, recommendations are much higher than the theoretical needs in order to avoid deficiencies. The choice of nutrients and units used in the formulation makes it possible to adjust intakes more or less precisely to requirements. For example, amino acid intakes are expressed in digestible amino acids (actually usable by the animal) and not in total amino acids (present in the feed).

To perform a specific formula by linear programming using Excel solver to formulate different feeds for chicken starter (0–4 weeks), grower (4–18 week) and layer (from 18 weeks), the values obtained after chemical analysis of the feeds were listed in Table 3 and presented in Fig. 4, Fig. 5 which compared with international standards (Delphine, 2018).

Table 3. Energy value (kcal/kg) of different types of feed distributed.

Variables	Values	Min	Max	Values	Min	Max	Values	Min	Max
Empty Cell	a) Feed 1 (Starter)			b) Feed (Pullet Grower)			c) Feed (Layer)
Energy kcal/kg	2827	2750	2850	2869	2800	2900	2710	2700	2800
Protein%	20.02	20	22	17.9	16	18	14.8	14	16
Moisture%	10.5	0	14	11.1	0	14	11.4	0	14
Fat content%	3.05	2	5	3.93	2	7	3.9	2	7
Ash%	5.67	—	10	4.64	—–	10	5.13	—–	10
Cellulose%	3.98	3.5	5	3.47	3.5	7	4.34	3.5	7
Starch%	40.24	32	—-	42.03	32	—-	40.61	32	—–

The results are compared to the INRA (INRA (Institut National de la Recherche Agronomique) 1989) dietary guidelines and show that the levels in foods are above the standards in all food types.

The protein content (Fig. 4) varied depending on the type of feed between 20.02, 17.9 and 14.8% respectively for starter, grower and layer chickens. These contents are within the recommended range, so these results are valid. The protein and amino acid content of the feed is one of the elements whose impact is fundamental for the growth of broilers. In the starter feed, the protein content must be higher than in the growth feed to increase the feed conversion ratio and improve weight gain. Improving the composition of the ration would reduce the cost of the feed. Energy and protein requirements change with age, and there are three feeding periods.

•
Start-up: from 0 to 4 weeks: protein requirements are very high: 20 to 22%, distribute crumb food as much as possible.
•
Growth: from 4 to 18 weeks: protein requirements decrease (16 to 18%), as do energy requirements. Distribute commercial food in semolina or pellets at will or flattened or crushed grains + protein supplement.
•
Laying: from 18 weeks to culling, it lasts at least one year, protein requirements decrease further (14 to 16%).

The continuous renewal of body proteins requires the synthesis of proteins from their unitary components: amino acids. Amino acids have several functions: (1) a maintenance role through cell renewal; (2) a growth role through the increase in the number and size of cells; (3) a role in certain secretions necessary for protein-rich productions such as egg production.

The fat contents (Fig. 4) vary from 3.05, 3.93 and 3.90% respectively for starter, grower and layer chicken. These contents are within the recommended range, so these results are valid. These are important sources of metabolizable energy for poultry feed. They increase the energy value of rations while decreasing feed conversion ratesipids facilitate the use of protein-rich raw materials (oil cakes) with relatively low energy levels (Sakande, 1993). According to Sakande et al., (1993) the use of animal fats, therefore rich in saturated fatty acids, can lead to the formation of soaps that are poorly absorbed by chicks and cause a poor use of calcium and. Consequently, an increase in the incidence of tibial dyschondroplasia.

They are also a source of essential fatty acids for animals. In addition, they limit dust and machine wear during feed production. However, they are said to cause excessive fattening, leading to carcass depreciation, especially when visible fat deposits are significant (abdominal fat, subcutaneous fat, etc.) (Lessire, 2001).

The starch contents (Fig. 4) vary from 40.24, 42.03 and 40.61% for starter, grower and layer chicken respectively. These contents are within the recommended range, so these results are valid. These levels are relatively high, which is appreciable given that carbohydrates are the main source of energy required for basic metabolism and nervous system function (Ponka et al., 2016).

The ash (mineral) content (Fig. 4) varied depending on the type of feed between 5.65, 4.64 and 5.13% respectively for starter, grower and layer chickens. These contents are within the recommended range, so these results are valid. Minerals have different functions, such as maintaining osmotic pressure (sodium), maintaining ionic balance (chlorine) or building the skeleton and/or eggshell (calcium. phosphorus). Minerals must be provided in sufficient quantities by the feed to avoid harmful deficiencies. The onset of egg-laying results in a high calcium requirement (minimum 3.5% per kg of feed), which is used for eggshell formation. Separate calcium feeding has a triple advantage: ‘ Better shell strength ‘ Consumption adapted to individual needs ‘ Reduction of dietary calcium allowing energy and protein concentration. Separate calcium feeding consists in providing less calcium via the compound feed (1%) and offering the hens a source of calcium at will (oyster shell, calcium carbonate pellets). In practice, the calcium particles are distributed mixed with the compound feed (meal). In addition to the solidity of the shell, the quality of the egg product will depend on the size (essentially linked to age) and the color of the yolk. The color of the yolk is linked to the presence of carotenoid pigments, called xanthophyll and depends very precisely on the nature of these pigments (yellow or red) and the quantities supplied. In organic poultry feed, these pigments are provided by corn, alfalfa protein concentrate and additives (natural pigments).

The cellulose contents (Fig. 4) vary from 3.98, 3.47 and 4.34% respectively for starter, grower and layer chicken. These contents are within the recommended range, so these results are valid. Finally, the moisture contents vary between 10.5 and 11.4%.

The energy values of the samples (Fig. 5) vary from 2827, 2850 and 2710 kcal/kg for starter, grower and layer chicken respectively. These levels are within the recommended range, so these results are valid. All these feeds provide the energy and nutrients required for growth, reproduction and animal welfare (Ponka et al., 2016).

Recommended intakes are defined so that the feed provides sufficient nutrients to meet the animal’s requirements. These requirements are defined in relation to a given production objective. They depend on the animal (age, production stage, sex, strain, individual variability). The environment (ambient temperature. quality and use of the run) and the production objectives set (age at slaughter. yield of parts. meat quality for broilers. number and quality of eggs for layers). According to the solutions found. They allow minimizing the cost of the formulated feed and satisfying the nutritional requirements of the selected poultry category by using the available feed ingredients. Indeed, the formulation of feed by linear programming using the Excel solver allows minimizing the cost of the feed while satisfying the nutritional needs of the poultry. However, the more feed ingredients are used the better the balance between feed cost and nutritional requirements.

5. Conclusion

In the present study, a least-cost formulation approach for complete feeds for poultry was developed by applying the linear programming technique. This formulation program required the creation of two databases on the characteristics of the raw materials and the nutritional and incorporation constraints to be satisfied in the formulation.

This study collected information on a very large number of raw material samples. The 300 samples analyzed cover about 9 raw materials. Although only some of them are presented in this document.

The analytical methods involved should not be too cumbersome or expensive, not all analyses should be performed on all samples: reasoned analytical strategies should be developed to estimate the quality of a sample with a minimum of analytical work (and therefore cost). NIRS spectrometry can be of great help in this work. Allowing rapid and inexpensive estimation of sample properties. Food formulators can also manage this variability through formulation practices that limit the risk of significant error.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

Abdelli et al., 2021

N. Abdelli, D. Solà-Oriol, J.F. Pérez

Phytogenic feed additives in poultry: achievements, prospective and challenges

Animals, 11 (2021), p. 3471, 10.3390/ani11123471

View in ScopusGoogle ScholarAbd Rahman et al., 2010

R. Abd Rahman, L.A. Chooi, R Ramli

Investigating feed mix problem approaches: an overview and potential solution

World Acad. Sci. Eng. Tech., 70 (2010), pp. 467-475

USA