AMINOTec FEED STRUCTURE AND PRODUCTION PROCESSES IMPACT ON POULTRY LIVE PRODUCTION CONTENT

Transcription

1 vessel/wagon discharge intake and storage lorry discharge dosing grinding mixing premix additive cooling granulating conditioning and pelletizing finished products load AMINOTec TOPICS ON FEED PRODUCTION AND OPERATIONS EDITION 1 NOVEMBER 2018 FEED STRUCTURE AND PRODUCTION PROCESSES IMPACT ON POULTRY LIVE PRODUCTION FEED COST IS THE MOST DOMINANT FACTOR IN POULTRY PRODUCTION COST. WHILE FEED PRODUCTION COST IS ONLY A MINOR PART OF THIS, FEED PRODUCTION PROCESSES HAVE A MAJOR INFLUENCE ON FEED QUALITY, CONTENT 03 Feed structure and production processes impact on poultry live production How to Manage Reduce Shrink Reduce Shrink 34 Leading Brazilian Feed Production chooses Evonik

2 2 AMINOTec EDITION 1 NOVEMBER 2018 Certain aspects are common to feed producers around the world and the necessity to supply healthy feed to the food supply chain at competitive cost is probably the most prominent. DETLEF BUNZEL, HEAD OF APPLIED FEED TECHNOLOGY, EVONIK NUTRITION & CARE GMBH DEAR READER, THE GLOBAL FEED INDUSTRY IS QUITE DIVERSE AS AGRICULTURAL PRODUCTION CHANGES WITH CLIMATIC CONDITIONS AND AS REGIONAL OR NATIONAL REGULA- TIONS AND HABITS INFLUENCE LIVESTOCK PRODUCTION AS WELL AS CONSUMER PREFERENCES. Certain aspects, though, are common to feed producers around the world and the necessity to supply healthy feed to the food supply chain at competitive cost is probably the most prominent. The structure of feed has an impact on both aspects, production cost and animal health. While pelleted feed is the standard in broiler production, Bunzel highlights options of reducing feed production cost by optimizing particle size distribution of the mash before pelleting or even introducing whole grains into the diet. Mubarak discusses the impact of such scenarios on broiler livestock production as well as health status of broilers. There seem to be opportunities for synergies in feed and livestock production in considering coarser particle sizes for broiler feed or even whole grain feeding. inventory losses in feed production. Such losses, commonly referred to as shrink, add to the cost of feed production without adding any value. Bunzel points out reasons for shrink and describe best pratices to manage and reduce shrink throughout the process steps. For decades Evonik has been providing consulting and handling solutions dedicated and optimized to support customers with using their products in feed production. Recently the hundredth system was started up by Evonik s team at a leading Brazilian feed producer. Lara gives some background on the project. You may also want to access our website for more details by scanning the QR code. P.S.: Please come back to us with questions or comments (aminotec@evonik.com). As always, suggestions for topics of your interest for future editions of AMINOTec are highly welcome. Another topic that is high on the agenda of feed mill managers is how to manage and explain

3 3 FEED STRUCTURE AND PRODUCTION PROCESSES IMPACT ON POULTRY LIVE PRODUCTION Author: Mubarak Ali, PhD Senior Technical Service Manager Evonik (SEA) Pte. Ltd.Singapore; co-author: Detlef Bunzel FEED COST IS THE MOST DOMINANT FACTOR IN POULTRY PRODUCTION COST. WHILE FEED PRODUCTION COST IS ONLY A MINOR PART OF THIS, FEED PRODUCTION PROCESSES HAVE A MAJOR INFLUENCE ON FEED QUALITY, WHICH IN TURN HAS AN IMPACT ON THE MAIN KPIS OF LIVE PRODUCTION, SUCH AS FEED INTAKE, BODY WEIGHT AND FEED CONVERSION RATE. KEY INFORMATION Mixing, grinding and pelleting are the core processes involved in in feed production. Pelleted feed is particularly prevalent in broiler production, because of the ease of handling and distribution of pellets, as well as its positive effect on feed intake and weight gain. It is well accepted that grinding has a major impact on mixing and pellet quality. At the same time, feed producers adapt feed structure to the specific needs of different species. Broiler feed typically has a coarser structure compared to swine or aqua feed, considering the digestive system of broilers and, specifically, the grinding function of the gizzard. roller mill Other feeding regimes for broilers such as whole grain feeding have developed, based on regional availability of raw materials and preferences of farmers. In the context of recent discussions and research, however, further feeding plans have been developed for broilers which have been shown to have positive effects on animal health and welfare. hammer mill After describing different approaches of grain handling on processes in feed and live production, nutritional aspects of various feeding regimes will be discussed. PICTURE 01 Roller mill and hammer mill complement specific strengths in step grinding process (CPM)

4 4 AMINOTec EDITION 1 NOVEMBER 2018 GRINDING From a process perspective, an appropriate mash structure is an important precondition to ensure good miscibility and homogeneity, as well as stability of the mixture during further transportation and processing. Engster (2014) discussed the importance of homogeneity of the mash in terms of its effect on animal performance in AMINOTec, Edition 1, In AMINOBatch Working Precision Tests, Evonik found that the particle size of the mash has an influence on coefficients of variation. The evidence for this included the analysis of supplemental amino acids (Bunzel, AMINONews Review 2009, AMINOBatch Working Precision Test A Valuable Tool to Evaluate Working Precision of your Mixing Plant). According to Fahrenholz (2017), particle size reduction is the secondmost common feed manufacturing process (mixing is first), and therefore has a substantial performance impact in the global sense. Grinding is most typically associated with the size reduction of cereal grains. He further explains that the reduction in particle size leads to increased ingredient surface area, leading to more intimate interaction with digestive acids and enzymes in the GIT (gastro-intestinal tract). The lack of adequate gizzard function, and resultant reduced GIT retention time, has a negative impact on live performance. As a result, interest in coarse particle inclusion has grown considerably recently. These nutritional aspects of the grinding process will be discussed in more detail at a later stage. On the other hand, particle size is known to be related to pellet quality, which may have an effect on animal performance (McKinney and Teter, 2004, figure 1). As pellet quality and durability depends on various factors, we can find different evaluations of their relative impact. Reimer (1992) states that 20 % of pellet quality is related to grinding (only dominated by formulation and on the same level as conditioning, figure 2). According to Fahrenholz (2012) particle size has the least impact out of seven influence factors he evaluated; the highest impact in this comparative trial was assigned to compression ratio of the die, followed by conditioning temperature and level of fat addition (figure 3) Weight gain d [g] FCR d ,05 2,00 y = 0,0014x + 1,9976 R 2 = 0, y = 0,7x+651 R 2 = 0,9225 1,95 1, Pellet Quality [% Pellets] 1, Pellet Quality [% Pellets] Weight gain Linear [Weight gain] FCR Linear (FCR) FIGURE 01 McKinney and Teeter, 2004: Impact of pellet quality on weight gain and feed conversation rate.

5 5 5 % Drying/Cooling 40 % Formulation 10,9 0,5 5,4 particie size 15 % Die 2,8 fat DDGS feed rate 20 % Conditioning 20 % Grinding 1,1 0,6 7 conditioning temp retention time die L:D FIGURE 02 Relative impact of process parameters on pellet quality (Reimer, 1992) FIGURE 03 Factors influencing pellet daurability (Fahrenholz, 2012) In a review by Amerah et. al.(2007) of the implications of feed particle size on the digestion and performance of poultry, a general industry consensus is assumed that in broiler diets based on maize or sorghum, optimum particle size should be between 600 and 900 μm. Here the authors conclude that available data clearly show that grain particle size is more critical in mash diets than in pelleted or crumble diets and suggest that coarser grinding to a more uniform particle size improves the performance of birds maintained on mash diets. In the same review, the research authors point out that the reduction of feed particle size is the second largest energy cost after that of pelleting in the broiler industry (Reece et al., 1985) and likely to be the largest user of energy in the layer industry where pelleting is not performed (Deaton et al., 1989). Dozier (2002) estimated that the utility usage comprised 25 to 30 % of the manufacturing cost of broiler feed. Thus, any reduction in energy consumption from grinding could significantly lower feed cost. Again, nutritional effects will be addressed in more detail at a later stage. Djuro M. Vukmirović et. al. conducted research in 2014 into the Influence of grinding method and grinding intensity of corn on mill energy consumption and pellet quality at the University of Novi Sad. Here the experts examined how grinding different structures of corn on a hammer mill versus a roller mill affects subsequent conditioning and pelleting quality. Six different particle size distributions were compared with GMD (geometric mean diameter) varying from fine (671 µm) to very coarse (2108 µm), three of which were produced on a hammer mill, three on a roller mill. To achieve differences in particle size distribution, openings were used on the hammer mill sieves with 3, 6 and 9 mm. On the roller mill, the gap between the lower pair of rollers was set to 1.4, 2.0 and 2.6 mm (table 1 and figure 4). THUS, ANY REDUCTION IN ENERGY CON- SUMPTION FROM GRINDING COULD SIGNIFICANTLY LOWER FEED COST.

6 6 AMINOTec EDITION 1 NOVEMBER 2018 Table 01: Geometric mean diameters (GMD) and geometric standard deviations (GSD) obtained with different mill type (hammer mill (HM) or roller mill (RM)) and different grinding intensity (fine (F), medium (M), coarse (C) and very coarse (VC)) GRINDING TREATMENT GMD / μm GSD / μm HM-F HM-M RM-M HM-C RM-C RM-VC FIGURE 04 Comparison of particle size distributions for different grinding treatments (different mill type (hammer mill (HM) or roller mill (RM)) and different grinding intensity (fine (F), medium (M), coarse (C) and very coarse (VC)).

7 7 kwh/t a Specific energy consumption of mill b c c d HM-F HM-M/RM-M HM-C/RM-C d RM-VC Mill type and grinding intensity Pellet durability index [%] ±0.47 ad ±0.65 ª ±0.85 ad ±0.27 b ±0.19 d ±1.27 c HM-F HM-M/RM-M HM-C/RM-C Mill type and grinding intensity RM-VC FIGURE 05 Influence of mill type (hammer mill (HM) or roller mill (RM)) and grinding intensity (fine (F), medium (M), coarse (C) and very coarse (VC)) on specific energy consumption of mill and on pellet durability index. Values with different letters are significantly different (p <0.05). Authors conclude that high energy savings are possible by coarser grinding (figure 5). Energy consumption and mill throughput increase over-proportionally with increasing sieve opening (and GMD) on a hammer mill. With roller mills, however, further energy savings are possible with comparative GMD. Furthermore, roller mills produce more uniform particle size distributions with lower GSD (geometric standard deviation), and corn ground on the roller mill clearly showed better pellet durability under comparative pelleting conditions. This research seems to suggest a way forward for those committed to saving energy in the grinding process. Specifically, it involves improving the structure of the mash by aiming for coarser particle size distributions, but without affecting pellet quality. Moreover, this way forward would reduce production costs and may also have positive effects on live production as further detailed below.

8 8 AMINOTec EDITION 1 NOVEMBER 2018 WHOLE GRAIN FEEDING In 2014 Singh et. al. published a comprehensive review paper on whole grain feeding. Here they state that In broilers, whole wheat feeding is common commercial practice in Europe (particularly Scandinavia, United Kingdom and the Netherlands), Australasia and Canada under a variety of forms. In some parts of the world, whole wheat feeding is also used as part of a strategy for broiler breeders and table egg layers. The main driver of this development is the potential of reducing feed cost by eliminating the grinding stage. Furthermore, the authors quote Gabriel et al., 2008: this also meets consumer demands for a natural feeding system and improved animal welfare. There are several strategies used to introduce whole grain into the feeding system, which depend on the existing infrastructure at feed mills and farms, as well as the availability and cost of raw materials (figure 6). Most research on whole grain feeding focuses on whole wheat. According to a review by Addollahi (2018), only three studies were conducted evaluating the effects of maize inclusion into pelleted broiler diets either in the form of cracked (Dozier III et al., 2006; Clark et al., 2009) or whole grain (Singh et al., 2014d). It is important to note that whole or cracked grains will replace only part of the grain of the diet. The rest of the diet will be a concentrate to provide amino acids, minerals, vitamins and trace elements as needed. It is also important to consider that the proportion of whole, or cracked grain in the diet, needs to be adapted to the age of the birds. In the feed mill whole grain (namely wheat), or cracked maize, could be applied at the pre-pelleting or postpelleting stages. In the case of postpelleting, it usually involves mixing with crumbled pellets to ensure good handling properties and mix stability. Adding whole grain pre-pelleting might be a good compromise for feed millers who wish to benefit from reduced feed production costs, while maintaining the advantages of pelleted feed. These include, for example, advantages related to handling and distribution, utilizing existing equipment and infrastructure and well-proven processes. When using cracked corn, it may be necessary to assess detrimental effects on pellet quality and possible negative impacts on feed intake of broilers versus energy savings. Whole Grain feeding Free choice feeding Mixed feeding Sequential feeding Mash Pellet Pre pelleting inclusion of whole grain Post pelleting inclusion of whole grain FIGURE 06 Different methods for feeding of whole grain in poultry

9 9 In terms of farming, three feeding strategies have been implemented with each requiring different infrastructure and management processes in live production. With sequential feeding, commercial feed concentrate and whole grain are handled on the same feeding system alternately, or in sequence, one after the other. With free choice feeding, commercial feed concentrate and whole grain are offered in parallel on separate feeding lines. With mixed feeding, whole grain is added proportionally into the feeding line, or is topped up onto the silo load while the silo is being replenished. Again, nutritional aspects of such strategies will be discussed at a later stage. In any case, when considering farm mixing and feed production, it is important for the feed mill, as well as the farmer, to manage interfaces and adapt diets. Here this can involve taking into account the nutritional requirements based on the feeding strategy, namely the proportion and nutritional value of grains that may be added in farms. The FAO has published a guideline, laying down requirements for on-farm feeding. Figure 7 depicts necessary planning and production processes that must be followed to ensure feed safety and quality. Implementation and daily maintenance of such processes certainly comes at a cost; and farmers will evaluate such costs and balance them against the potential savings they may have when adding whole grain in farms. This will most probably only pay off in the case that they use home grown grains. Production Plan Specification for material Purchase of the materials Receipt & Inspection of material Materials for processing Quality Control Formulation Re-processing if possible Weighing Return to supplier Qualified personnel could perform these tasks Mixing Check list Quality Control Fall Labeling & packing Pass Storage Feed animals FIGURE 07 Source: Avitech Animal Health Pvt. Ltd, adapted

10 10 AMINOTec EDITION 1 NOVEMBER 2018 Effect of particle size and feed form on growth performance The importance of particle size in poultry diets has been recognized because of its various benefits in terms of gizzard development and improvement in growth performance (Abdollahi et al., 2018). Various studies over the last few decades have demonstrated that particle size has a positive impact on the feed intake of broilers fed mash diets. Table 2 shows that feed intake in broilers fed mash diets can be increased by also increasing particle size, and the table also shows that effect varies with the age of birds and type of grain. Attawong et al. (2014) recommends that corn particle size of around 805 micron is sufficient for younger birds (<28 days). However, the older the birds get, a larger corn particle size is most likely to be required. Feed intake of crumbled or pelleted diets, for example, was not affected by particle size of maize or wheat (as shown in Table 3). Weight gain and feed efficiency improved with increases in particle size of grains in either in mash or pelted diets, but here the results are contradictory (Tables 2 and 3). Auttawong et al. (2013) tested two dietary levels of coarse corn (0 or 35 %) on broiler performance and noted an improvement in levels of feed efficiency with coarse corn (1080 microns) with ad libitum feeding over restricted feeding. Similarly, Xu et al. (2015) observed improved zootechnical parameters and digestive functions of broiler birds when crumble-pelleted diets contained 50 % coarse corn. The authors reported improvement in body weight (130 g) and feed conversion ratio (12 points) in broilers aged 42 days and over. This correlates well with the increased gizzard weight (2.75 mg/g body weight), increased digesta retention time (0.78 h), improved apparent ileal digestibility of energy (8.2 %) and also improved apparent ileal digestibility of nitrogen (12.4%) in broilers fed diets with 50% coarse corn compared to those fed without any coarse corn. PARTICLE SIZE HAS A POSITIVE IMPACT ON THE FEED INTAKE OF BROILERS FED MASH DIETS. PICTURE 02 Feed structure from mash to pellets impacts production cost, handling properties and KPIs in live production

11 11 Table 02: Effect of particle size on feed intake, weight gain and feed conversion ratio (FCR) of broilers fed mash diets GRAIN TYPE AGE (d) MILL TYPE / SCREEN SIZE GMD PARTICLE SIZE (µm) GSD FEED INTAKE (g) WEIGHT GAIN (g) FCR REFERENCE Maize, wheat and sorghum 7 21 H (2 mm) a 427 a 1.55 b Nir et al. (1994) H (0.8 mm) b 357 b 1.65 a H (3 mm) a 401 a 1.60 ab Wheat 1 21 Maize 1 21 H (3 mm) NR NR 777 b 453 b 1.72 a Amerah et al. H (7 mm) NR NR 877 a 539 a 1.63 b (2007b) H (3 mm) b 1.36 Amerah and H (7 mm) a 1.38 Ravindran (2009) Maize 1 21 H (2 mm) H (5 mm) H (8 mm) Naderinejad et al. (2016) H, Hammer mill; R, roller mill; NR, not reported Table 03: Effect of particle size on feed intake, weight gain and feed conversion ratio (FCR) of broilers fed crumbled or pelleted diets GRAIN TYPE AGE (d) MILL TYPE / SCREEN SIZE GMD PARTICLE SIZE (µm) GSD FEED INTAKE (g) WEIGHT GAIN (g) FCR REFERENCE Maize 1 42 H (3.18 mm) a 1.89 a Reece et al. H (6.35 mm) b 1.91 b H (9.53 mm) a 1.89 a (1986) Maize 1 21 Wheat 1 21 Wheat 1 21 Maize 1 21 H (3.18 mm) 716 NR a 1.40 Lott et al. H (9.59 mm) 1196 NR b 1.42 (1992) H (3 mm) NR NR Amerah et al. H (7 mm) NR NR (2007b) H (1 mm) NR NR Amerah et al. H (7 mm) NR NR (2008) H (1 mm) NR NR Amerah et al. H (7 mm) NR NR (2008) Maize 1 21 H (2 mm) H (5 mm) H (8 mm) Naderinejad et al. (2016) H, Hammer mill; R, roller mill; NR, not reported

12 12 AMINOTec EDITION 1 NOVEMBER 2018 Various examinations have also been made into the effect of particle size reduction of commonly used by-products in poultry feed, such as distillers dried grains with solubles (DDGS) and soybean meal (SBM). Pacheco et al. (2013, 2014) conducted experiments to evaluate the effect of particle size of DDGS and SBM on live performance of broilers. In the first experiment, two particle sizes (700 and 350 microns) of DDGS were tested. Results showed that birds fed pelleted diets containing 350 micron DDGS had both greater body weight and higher performance levels. Pacheco et al. (2014) also evaluated the effect of particle size of SBM (410 or 1025 microns) and DDGS (480 or 745 microns) on pellet quality and live performance. Here the inclusion of fine SBM (410 microns) was shown to improve pellet quality, whereas coarse SBM (1025 microns) had positive effect on live performance. For DDGS, however, fine particle size decreased pellet durability whereas overall pellet durability index values were similar. Fine DDGS (480 microns) also increased feed intake and body weight without any impact on feed efficiency. Here coarse DDGS (745 microns) in broiler diet, however, increased gizzard weight. Therefore, it can be concluded that coarse particle size of DDGS and SBM in broiler diets results in better growth performance. The physical form of feed also has significant impact on broilers growth performance (Dozier et al. 2010). Many studies, for example, have reported improvement in broiler performance when fed pellet feed Average daily feed intake (g/d) Average daily gain (g/d) a c a b a b a c a b a b 90 Fine Medium Coarse 55 Fine Medium Coarse Feed particle size Feed particle size Crumble-pellet Mash Crumble-pellet Mash Feed conversation ratio 1,69 1,68 1,67 1,66 1,65 Fine Medium Coarse Feed particle size Crumble-pellet Mash FIGURE 08 This shows the effect of feed form (mash and crumble-pellet) and feed particle size (fine, medium and coarse) on average feed intake, average daily gain and feed conversion ratio of broilers (aged 1-40 days). Fine, medium and coarse particle sizes were achieved by grinding the macro ingredients (wheat, corn and soybean meal) in the hammer mill to pass through 2, 5 and 8 mm sieves, respectively. (Mingbin et al., 2015)

13 13 compared to mash feed (Amerah et al., 2008; Chewning et al., 2012; Mingbin et al., 2015). Use of crumble or pellet form of feed was found to reduce the feed wastage and prevents particle selection. The pelleting process also improves palatability and increases nutrient digestibility (Mingbin et al., 2015). Zang et al. (2009), for example, also reported improvement in feed intake, weight gain and feed efficiency when broilers were fed pelleted diets. Mingbin et al. (2015) conducted a study to evaluate the effects of feed form (mash and crumble-pellet) and feed particle size (fine, medium and coarse) on performance and digestive tract development of broilers. Results showed that feed form had a greater effect than feed particle size on broiler performance and digestive tract development. Here the study also concluded that the effect of feed particle size was dependent on feed form. Figure 8 shows on effect of feed form and feed particle size on growth performance of broilers. Impact of feed particle size is more critical in birds fed mash diets. This observation was in line with Svihus et al. (2004) wherein the authors concluded that the pelleting of mash diet could negate the effect of feed particle size. particle size on broiler performance gained little attention among the researchers (Amerah et al., 2007a). Published studies indicate that broiler performance increases with an increase in particle size of grains. The effect of the particle size of grains on a broiler s feed intake, is far clearer for mash diets than for pelleted diets. As a result, coarse mash feeding has been practiced for a long time in many countries. Mash diets have been replaced with crumble-pelleted diets in current feeding practices of broilers. Broilers fed pelleted diets show higher feed intake and weight gain levels compared to those fed mash diets at same particle size (Figure 8). Pelleting has been shown to improve feed intake, weight gain and feed efficiency (Zang et al., 2009). Coarse particle size of grains or by-products, either in mash or pelleted diets, may improve weight gain and feed efficiency. Similarly, the physical form of feed also has a significant effect on a broiler s growth performance. PHYSICAL FORM OF FEED HAS SIGNIFICANT IMPACT ON GROWTH PERFORMANCE OF BROILERS It should be recognized that it is not only particle size which is important, but also that the uniformity of particle size in feed is of great significance when striving to fully exploit the genetic potential of fastgrowing broilers. Chickens are known to prefer for large particle sized feed. However, due to current feeding practices whereby crumble/pelleted diets are used, exploring the uniformity of grain PICTURE 03 Pelleted feed is prevalant in broiler production

14 14 AMINOTec EDITION 1 NOVEMBER 2018 Table 04: Effect of feed particle size on the relative weight (g/kg body weight) of gastrointestinal tract (GIT) segments in broilers (Adapted from Zaefarian et al., 2016) GIT SEGMENTS GRAIN TYPE FEED PARTICLES REFERENCE FINE MEDIUM COARSE Gizzard Wheat Svihus et al. (2004) Crop Proventriculus Gizzard Duodenum Wheat Jejunum Ileum Caeca Crop Proventriculus Gizzard Duodenum Wheat Jejunum Ileum Caeca Crop Proventriculus Gizzard Duodenum Maize Jejunum Ileum Caeca Proventriculus Gizzard Maize Small intestine Caeca Proventriculus Gizzard Maize Jejunum Ileum Pancreas Proventriculus Gizzard Maize Small intestine Caeca Amerah et al. (2007) Amerah et al. (2008) Amerah et al. (2008) Amerah and Ravindran (2009) Xu et al. (2015) Naderinejad et al. (2016)

15 15 Table 05: Effect of corn particle size on cecal microbial populations in 21-d-old chicks in experiment 1 as determined by quantitative PCR 1 (Adapted from Jacobs et al., 2010) CORN PARTICLE SIZE LOG 10 CFU/g OF DM BIFIDOBACTERIA LACTOBACILLI ESCHERICHIA COLI 557 µm 8.25 a a µm 7.69 ab ab µm 7.44 ab ab µm 7.23 b b Pooled SEM a, b Means within a column with no common superscript, and which thus differs significantly (P < 0.05). 1 Means represents 4 replicates of 6 chicks each per treatment. Effect of particle size on gastrointestinal tract development and gut health Feed particle size influences the development of the gastrointestinal tract to a greater extent when the broilers are fed mash diets, as opposed to pelleted diets. Reducing the particle size property during the pelleting process may result in suboptimal gizzard development and changes in the morphology and microbiota profile of the intestinal tract (Zaefarian et al., 2016). A large particle size helps the gizzard to function properly, and is also beneficial to gut health development in broilers (Svihus et al., 2004; Choct, 2009). Naderinejad et al. (2016) also found that coarse grinding of maize in pelleted diets had a positive effect on gizzard development and functionality, which is in turn beneficial for nutrient utilization and growth performance. The gizzard is extremely good at grinding the feed to a consistent particle size (Hetland et al., 2004). A welldeveloped gizzard also improves grinding activity and gut motility (Ferket, 2000). It increases cholecystokinin release, which stimulates the secretion of pancreatic enzymes and gastroduodenal reflexes (Duke, 1982; Svihus et al., 2004). Coarse particles reduce the digesta rate in gizzard and lower the ph of gizzard (Nir et al., 1994). The low ph of gizzard may increase pepsin activity (Gabriel et al., 2003) and protein efficiency. It also reduces the risk of coccidiosis (Cumming, 1994) and feedborne pathogens (Engberg et al., 2002). The effect of feed particle size on segments of broiler gastrointestinal tract is presented in Table 4. Feed particle size also affects the intestinal tract segments other than the gizzard, but the results are contradictory. Nir et al. (1995) reported no difference in the weight of intestinal segments with an increase in maize particle size from 0.6 to 2.17 mm. Xu et al. (2015) also reported non-significant differences for intestinal length, and the ratio of intestinal length to body weight when 50% coarse corn is included in pelleted diets. However, Amerah et al. (2007a) reported a reduction in the relative weight of small intestinal segments with an increase in wheat particle size. Limited data has been reported on the effect of feed particle size on intestinal morphology and microbiota profile. Amerah et al. (2007a) reported a non-significant change in villus height, crypt depth, and epithelial thickness in the duodenum with an increase in maize particle size. However, Liu et al. (2006) and Xu et al. (2015) reported a

16 16 AMINOTec EDITION 1 NOVEMBER 2018 Relative gizzard weight (g/kg) a ab ab bc c d e WW 4.5 % WW 4.5 % WW 9.0 % WW 18.0 % WW 4.5 % WW 9.0 % WW 18.0 % Control Pre-pellet Post-pellet Gizzard digesta (ph) c bc c bc ab ab a WW 4.5 % WW 4.5 % WW 9.0 % WW 18.0 % WW 4.5 % WW 9.0 % WW 18.0 % Control Pre-pellet Post-pellet FIGURE 09 Effect of pre- and post-pellet inclusion levels of whole wheat (WW) on gizzard weight, and ph of gizzard digesta. (Truong et al., 2017) Relative gizzard weight (g/kg) c bc bc ab abc a ab a b b b b b b a 1,55 1, b b b b b b 1,45 0 Control WW 4.5 % WW 9.0 % WW 18.0 % WW 4.5 % WW 9.0 % WW 18.0 % Pre-pellet Post-pellet 1,4 Feed intake (g) weight gain (g) Feed conversation ratio FIGURE 10 Effect of pre- and post-pellet inclusion levels of whole wheat (WW) on growth performance from 7 to 28 days post-hatch. (Truong et al., 2017)

17 17 positive effect of coarse maize on intestinal morphology. The inclusion of coarse maize reduced the number of mast cells in the duodenum, jejunum, and ileum compared with finely ground maize (Liu et al., 2006). Mast cells translate stress signals to release a wide range of neurotransmitters and pro-inflammatory cytokines, which can have dramatic effects on the gastrointestinal physiology of broilers (Vicuna, 2015). Reduction in mast cells in the small intestine is also beneficial. An increase in mast cell numbers reflects Eimeria infection in broilers (Morris et al., 2004). Furthermore, particle size of ingredients also affects the intestinal microbiota profile. Jacobs et al. (2010), for example, reported changes in cecal microbiota profiles when using different corn particle sizes. The effects of corn particle size on cecal microbial populations are presented in Table 5. The lactobacilli population was significantly increased (P < 0.05) when the largest corn particle size of 1,387 µm was included in the diet. Increase in lactobacilli spp. concentration is considered to be beneficial because it can prevent colonization of pathogens such as E. coli (Engberg et al., 2002). The bifidobacteria population was significantly decreased with increasing corn particle size from 557 µm to 1,387 µm. However, E. coli populations was not affected by corn particle size. Singh et al. (2014) also found that the counts of Lactobacillus spp. and Bifidobacteria spp. increased and those of Clostridium, Campylobacter and Bacteroides species were decreased with increas- AN INCREASE IN MAST CELL NUM- BERS REFLECTS EIMERIA INFEC- TION IN BROILERS Intestine Crop Gizzard FIGURE 11 Gastro intestinal tract of chicken features crop and gizzard to grind and digest coarse particles

18 18 AMINOTec EDITION 1 NOVEMBER 2018 ing inclusion levels of coarse maize (0 to 600 g/kg). Coarse mash diets can also increase the Lactobacilli populations in the ceca and rectum (Engberg et al., 2002). It can be concluded that large particle size not only influence the development of gastro intestinal tract, but also change the cecal microbial populations. Large particle sizes also increase the Lactobacilli counts and reduce the pathogenic bacteria in a broiler s caecum. Effect of whole grain feeding on growth performance and gut health The primary aims of feeding whole grains to broilers is to reduce feed cost and to improve digestive functions (Singh et al. 2014). It also has a positive impact on gut health by encouraging the colonization of beneficial bacteria, and reduces incidences of coccidiosis (Cumming, 1989). Engberg et al. (2004) reported that whole wheat feeding can reduce intestinal numbers of Clostridium perfringens, which is important for prevention of necrotic enteritis. Another finding is that whole grain feeding practices meet consumer demands for a natural feeding system, and are also good for animal welfare (Gabriel et al., 2008). Whole wheat feeding as part of a free choice feeding system resulted in increases in weight and length of the segments of the small intestine (Singh et al., 2015). Fernandes et al. (2013) also reported that the weight of small intestine increases in case of broiler diets containing 50 or 100% of whole sorghum grain. Cecum length was also significantly increased when birds were fed diets containing whole sorghum grain. The intestine is the biggest immune organ inside a bird s body which contributes further to better health and immune response. It has been reported that a whole grain feeding approach helps in preventing the enlargement of proventriculus and atrophy of the gizzard which are common problems with pelleted diets (Singh et al. 2014). Whole wheat administered via free choice feeding methods increase the relative gizzard weight, irrespective of whether mash or pellet forms of feed are used (Singh et al., 2014). Another claim, is that the gizzard is a pace maker of gut motility and even reverses peristalsis in broilers (Ferket, 2000), whilst whole grain feeding may also influence starch digestive dynamics, and provide more gradually or slowly digestible starch. This effect on starch digestion may lead to improvements in energy utilization and feed efficiency (Liu et al. 2015). Wu et al. (2004) reported that, pre-pellet, 20% whole grain addition improved energy utilization. Published studies on whole grain feeding, however, reported contradictory results. Many studies, for example, reported beneficial effects of whole grain feeding on broiler performance, whereas others reported no advantage or even poorer performance (Singh et al. 2014). Whole grain can be incorporated into poultry feed either pre or post steam pelleting, and here offered as either intact pellets or as a whole grain pelleted concentrate blend (Moss et al., 2018). The practice of feeding whole grains to broilers along with balanced concentrate is currently becoming more accepted in certain regions, including Europe, Canada, Australia and New Zealand. Chicken meat production is an extensively integrated process in Australia and New Zealand. In New Zealand, whole wheat is usually incorporated into broiler diet prior to steam pelleting, whereas in Australia, whole wheat is added post pelleting. The effect of pre- and post-pellet inclusions of whole wheat in broiler diet on gizzard weight, gizzard digesta ph and growth performance is presented in Figures 9 and 10 (Truong et al. 2017). The post-pellet inclusion of whole wheat in broiler diets had a greater effect than pre-pellet inclusions. Compared to ground grain control diet, post-pellet whole wheat inclusions increased relative gizzard weight, reduced gizzard digesta ph and also reduced incidences of dilated proventriculus. Feed efficiency was also significantly improved in all diets containing whole wheat included (pre- and post-pellet).

19 19 Conclusions The particle size of grains and by-products used in broiler feed effect both pellet quality and growth performance. Particle size also has a positive impact on the development of the gastrointestinal tract, mainly the development and functionality of the gizzard. Large particles help broilers gizzards to function well, and also help their overall gut health development. For mash diets, however, the impact of particle size is more critical. The use of large particle size of grains increases the Lactobacilli count, and also reduces the population of pathogenic bacteria. Whole grain feeding improves digestive functions and encourages the colonization of beneficial bacterial. It reduces pathogenic challenges, including, mainly, the incidence of coccidiosis. The post-pellet inclusion of whole grain showed greater impacts on broiler performance compared to pre-pellet inclusions through increased relative gizzard weight, low gizzard digesta ph, and reduced incidences of dilated proventriculus. Energy costs involved in producing pelleted feed may be reduced by aiming for coarser particle size distributions and employing adapted grinding regimes. This would include the use of roller mills, while also making sure to maintain pellet quality and broiler performance. Further opportunities may also be discovered when looking into whole grain breeder feeding regimes which may offer beneficial effects on feed cost, on performance, and also on overall gut health aspects. In all of these cases, regional and infrastructure-related factors will play a key role.

20 20 AMINOTec EDITION 1 NOVEMBER 2018 HOW TO MANAGE AND REDUCE SHRINK IN FEED PRODUCTION Author: Detlef Bunzel (detlef.bunzel@evonik.com), VP Applied Feed Technology, Evonik Nutrition & Care GmbH, Hanau, Germany ALL MANAGERS RESPONSIBLE FOR OPERATIONS WHICH DEAL WITH COMMODITIES ARE FAMILIAR WITH THE TEDIOUS TASK OF KEEPING INVENTORIES AND, EVEN WORSE, OF RECONCILING INVENTORY DISCREPANCIES intake and storage dosing grinding mixing conditioning and pelletizing vessel/wagon discharge premix additive lorry discharge cooling granulating FIGURE 01 SIMPLIFIED FEED MILL PROCESS DIAGRAM

21 21 KEY INFORMATION Whenever inventories are lower than expected, we talk about shrink. Within the feed industry, shrink of up to 1 % of the feed mill output is regarded as being in the standard or expected range (Fahrenholz, 2017). To appreciate the concept of shrink from an accounting perspective, it helps to translate this into dollars. So, for an annual feed production of mt at a price of 250 $/mt, 1 % of shrink therefore accounts for an annual loss of $. In the following paragraphs we will examine the feed production process step by step, focusing on aspects that are most prone to have a serious impact on inventories. INTAKE & STORAGE 6 finished products load Products arrive at the feed mill in various forms of delivery, depending on a range of factors such as the local infrastructure, product origin and volumes handled. While in the Americas, and also in some Asian countries, standard feed formulation is based on corn and soy as macro ingredients, in Europe, however, a much higher number of main ingredients is common. In the US main ingredients, which are locally produced, typically arrive in railcars or trucks, whilst in northern European countries, large sized feed mills are often located at rivers or canals, and receive their bulk ingredients by ship. On the other hand, in some regions in Asia and Africa, often even main ingredients are manually handled in bags due to poor local transport infrastructure. Minor ingredients and micro ingredients are also typically handled in bags, big-bags or drums. Methods used to keeping stock of ingredients delivered in full bulk, or as packaged goods, vary considerably. While bulk deliveries are generally recorded by weight, for example, packaged goods are usually recorded by the amount of package units. Depending on who delivers the full bulk, the load may come with shipping documents and an invoice based on the weighing ticket of the supplier, or it may come directly from the farm without a weighing ticket. In the first case, the feed mill will check the weight of the load, then compare it with the net weight according to the bill of lading. With full bulk loads, two weighing tickets are normally necessary to secure the net weight at the point of loading, i. e. the weight of the full truck load after filling, minus the weight of the empty truck entering the bulk filling station. A related process is carried out at the feed mill when discharging, i. e. a full truck entering the feed mill minus empty truck leaving the feed mill. Truck scales with a weighing capacity of 60 mt typically have a resolution of 10 or 20 kg, i. e. the smallest weight change that can be detected on the scale is 10 or 20 kg. This depends on the regulatory standards that the weights and measures authority apply, which differs in the US and in Europe (cp. Table 1 and 2, Mettler Toledo, 2013). According to weights and measures regulations, the weighing accuracy of a truck scale must be within ± 1.5 x resolution during verification, and must stay within ± 3 x resolution within the maintenance period, i. e. in between verification intervals.

22 22 AMINOTec EDITION 1 NOVEMBER 2018 Table 01: NIST / NTEP Accuracy Tolerance Handbook 44 outlines a step-based tolerance that increases with every additional 5 % of the scale s capacity used. Table 02: There are various accuracy specification levels defined by OIML: C3, C4, and C6, and these differences can be observed in the range of truck scales available. Most truck scale applications are C3 (used in the chart above). A very small percentage of truck scales are C6, which provides a stricter degree of accuracy tolerance. SCALE CAPAC- ITY (lbs.) TRUCK WEIGHT (lbs.) ALLOWABLE ERROR (lbs.) SCALE CAPAC- ITY (kg.) TRUCK WEIGHT (kg.) ALLOWABLE ERROR (kg.) (multi-range) Based on all of this, weight differences between the bulk weight on the bill of lading, and feed mill weighing ticket of 2 x 3 x 20 kg = 120 kg, are still within the tolerance of scale accuracy within the maintenance interval. This is, of course, assuming that one scale has the maximum weight plus deviation, while the second scale is to the maximum weight minus deviation. With a full truck load of 20 to 25 mt, this difference is in the range of 0.5 % of the payload. The same range of accuracy of ± 0.5 % is typical for class 1 bulk hopper scales. Therefore, it is important to devise a system which can be used on a daily basis to account for such weight differences for incoming bulk ingredients. It is good practice to keep track of differences for each supplier separately. Differences should be random and balanced out over time. In case a pattern of regular shortages emerges, claims should be filed with suppliers on a monthly or quarterly basis depending on the number of bulk loads to compensate for shortages. Depending on the infrastructure of bulk storage systems, inventories of bulk products can only be calculated by measuring or estimating volumes of bulk and multiplying the amount by the bulk density. This process, however, is obviously prone to human error. Depending on the logistics of the plant, inventory differences may also be due to accounting differences of full bulk loads in railcars or trucks. Specifically, these may apply to loads which have arrived on-site, but whose full bulk has not been discharged and has therefore not been counted during stock taking. This is therefore irrespective of any shipping documents which may have been processed and therefore already show due inventory in the accounting system. Such differences relating to full truck loads or railcars typically are therefore usually balanced in the following month, when the stock has entered the system physically, i. e. when bulk has been transferred from railcars or trucks into the storage system. IT IS GOOD PRACTICE TO KEEP TRACK OF DIFFERENCES FOR EACH SUPPLIER SEPARATELY.

23 23 With packaged goods, stock is calculated based on counting package units. Despite this process, stock differences may still occur based on counting errors. Even with packaged goods, it is good practice to double-check the accuracy of the nominal weight of package units on a randomized basis. Again, depending on the nominal weight, filling tolerances of ± 1 % for an individual package unit are possible. E. g. with 25 kg bags the actual net weight may vary ± 250 g. On average bags of one delivery should contain the nominal weight of 25 kg. Thus, a full pallet of 40 x 25 kg bags should add up to kg. With packaged goods, damaged package units may result in product spills and losses. Such losses must also be accounted for in stock taking and claimed for from the suppliers, assuming the damage was obvious upon delivery, and before unloading. Depending on the delivery form (bulk or packaged) and the origin of ingredients, incoming raw materials are cleaned during intake, e. g. stones and other residues are removed. Here feed mill managers certainly have often developed their own methods for calculating acceptable losses for each product at this stage. A major Brazilian feed miller, for example, may accept losses of % for maize and classify this as normal. Hicks (1992) describes handling losses as being due to loss of volatile compounds such as oils, mechanical losses from broken kernels and foreign material, and possibly also due to respiration of the seed itself. The 3-year on-farm average (of handling losses) was 0.82 % compared to 0.88 % for the commercial facilities according to research published by Iowa State University. According to Weber (1995) grains are living beings and, as such, even biologically disrupted from the plant at the moment of their maturation, they breathe, release carbon dioxide (CO2), water (H2O) and heat. This reaction causes a loss of grain mass and is affected by storage conditions, such as moisture and temperature of the grain mass. Kongskilde describe this process in the technical publication Grain Biology, as follows: Germinative grain breathes. Each kernel gets oxygen from the air and burns food from its endosperm. This process gives off heat, water, and carbon dioxide. This process in grain is called respiration. In many countries, respiration during the storage period is causing a dry matter loss of 1 % or more. Figure 2.1 and 2.2 show the loss in weight due to CO2 respiration of wheat and barley depending on grain moisture contents and on grain temperature at a moisture of 17 %. Depending on ambient temperatures and relative humidity, i. e. depending on climatic conditions, grain will adapt and stabilize its moisture content to an equilibrium as shown in table 3 (Sadaka, FSA1074). At any given storage temperature, the equilibrium moisture of the grain will drop with the relative humidity of the ambient air and vice versa. Making use of this phenomenon, in some parts of the world incoming raw materials are sun-dried in the yard upon delivery, before filling them into the storage bins. This is done to avoid product losses due to excessive moisture of the grains in storage bins, and to reduce energy costs for venting and drying during storage. mg CO₂ Produced/1000g dry matter/24 hours 1500 mg CO₂ Produced/1000g dry matter/24 hours 100 Wheat, 17% mc Grain moisture (%) Respiration of wheat and barely Grain temperature Respiration depend on temperature FIGURE Respiration increases with grain moisture content at a given temperature and with grain temperature at a given moisture content (Kongskilde)

24 24 AMINOTec EDITION 1 NOVEMBER 2018 Moisture content of grain must be adjusted to the range as shown in table 4, by using ventilation, heating or cooling systems to ensure a safe storage considering ambient temperature and humidity for long term storage. By reducing the humidity of grain, the weight of the grain will decrease as a result of the water evaporating in the drying process. The amount of water loss or shrink can be seen in table 5: When reducing the humidity from 20 % to 12 % to a safe, warm climate at which to store grain (cp. table 4), 91 kg of water has evaporated, resulting in a shrink of 9.1 % (Kongskilde). To best manage the issue of shrink in grain storage, it is important to consider potential weight loss in grain mass through respiration, and through loss of moisture by drying. Furthermore, aeration procedures must be adapted to environmental conditions and control systems, must be implemented to avoid unnecessary losses by germination, fungal growth or insect infestations (figure 3, Kongskilde). Another effect that may occur with product intake and on-site transportation into storage systems, relates to appararent shrink (or gain) by incorrectly-positioned distribution systems, e. g. leaking outlet slides of trough chain conveyors or leaking turn-heads or two-way distributors (picture 1). Assuming product A is filled into bin 1, but the turn-head is incorrectly positioned in between bin 1 and bin 2, part of product A will end up in bin 2. This will be noted during stock taking as shrink in bin 1, and gain in bin 2. Furthermore, it is obvious that any leaking gates, couplings or pipe ducts, will cause spillages and result in product losses. This is not only true for the intake area, but for all process steps in the feed mill. ANY LEAKING GATES, COUPLINGS OR PIPE DUCTS, WILL CAUSE SPILL- AGES AND RESULT IN PRODUCT LOSSES. Table 03: Corn Equilibrium Moisture Content RELATIVE HUMIDITY (%) Temperature ( F)

25 25 Table 04: Safe moisture content for grains in long- and shortterm storage 1 GRAIN TYPE LONG TERM (ONE YEAR) WARM TEMPERATURES SHORT TERM (30 TO 60 DAYS) COOL TEMPERATURES Corn 12 % 12 % 15 % Sorghum 12 % 12 % 15 % Rough Rice 12 % 12 % 14 % Soybean 11 % 11 % 14 % Wheat 12 % 12 % 14 % 1 Table 05: Weight in kg of water lost when one ton of damp grain is dried. INITIAL MOISTURE CONTENT % FINAL MOISTURE CONTENT % PICTURE 01 Turnhead and two-way distributor

26 26 AMINOTec EDITION 1 NOVEMBER 2018 Temperature C A 10 B C 0 0 % 10 % 15 % 20 % 25 % A B C Safe Fall in germination Lower limit for insect heating. Lower limit for germination. Lower limit for fungal heating. Insect heating Fungal heating m.c. FIGURE 03 Values of grain temperature, relative humidity and moisture content for safe storage, insect and fungal heating, and fall in germination PICTURE 02 Grain storage system with elevator and trough chain conveyor distribution system

27 27 DOSING, GRINDING & MIXING With dosing systems, be it dry bulk or liquid dosing, whenever metering (weighing or volumetric metering systems) comes into play, errors may occur that are related to metering accuracy, as described in detail for truck scales. Any scale certified for legal-for-trade applications will be subject to regular inspection by weight and measures authorities, based on the respective regulations. In feed mills, this will usually apply for truck scales or hopper scales in bulk loading systems, as well as bagging equipment. Other in-process metering systems are not subject to regular inspection and verification by authorities. DEPENDING ON THE METERING/ WEIGHING SYSTEM, ERRORS MAY RANGE FROM % OF THE DESIGN CAPACITY OF THE EQUIPMENT. Depending on the metering / weighing system, errors may range from % of the design capacity of the equipment. Therefore, it is a good practice to do a thorough maintenance and check weighing / metering accuracy of process scales and liquid meters, at least once a year. As process scales or metering systems are part of a more complex process stage, the system accuracy also depends on other factors that may be related to the process design, or the equipment maintenance. PICTURE 03 Examples of badly maintained platform scales of micro dosing systems Picture 3 shows examples of several platform scales of AMINOSys dosing systems that Evonik engineers discovered during annual service visits to feed mills. Accumulated product spills block scale movement, which in turn affects weighing accuracy. Spills may even attract rodents or birds who then attempt to build their nests inside the protected area of the platform scale. Rodents even managed to do severe damage to connecting cables to load cells by gnawing at them, and this caused downtime for troubleshooting and corrective maintenance.

28 28 AMINOTec EDITION 1 NOVEMBER 2018 Rio Grande do Sul Humidity difference grinding maizel (%) 0,20 0,10 0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,00 % 0,00 % 0,22 % 0,11 % 0,05 % 0,17 % 0,08 % 0,40 % 0,42 % 0,33 % 0,64 % 0,08 % 0,09 % 0,14 % 0,26 % Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Humidity difference conditioning (%) 0,15 0,14 0,14 0,13 0,13 0,12 0,12 0,11 0,13 % 12,66 % 13,31 % 13,80 % 12,80 % 12,92 % 13,80 % 13,34 % 13,68 % 13,66 % 13,58 % 12,16 % 12,78 % 13,25 % 13,98 % Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Humidity difference pelleting (%) 1,50 1,00 0,12 % 0,08 % 0,87 % 0,50 0,00 0,50 1,00 1,50 0,01 % 0,44 % 0,22 % 0,24 % 0,19 % 0,50 % 0,57 % 0,56 % 0,11 % 0,36 % 0,67 % 0,48 % Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec FIGURE 04 Humidity differences of process steps in two feed mills in Brazil

29 29 Minas Gerais Humidity difference grinding maize (%) 0,20 0,10 0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,00 % 0,00 % 0,20 % 0,05 % 0,11 % 0,03 % 0,52 % 0,30 % 0,27 % 0,10 % 0,15 % 0,08 % 0,54 % 0,32 % 0,20 % Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Humidity difference conditioning (%) 0,17 0,16 0,16 0,15 0,15 0,14 0,14 0,13 0, Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 0,14 % 14,84 % 14,99 % 15,03 % 15,95 % 15,36 % 14,92 % 14,69 % 14,78 % 15,08 % 14,20 % 15,66 % 14,87 % 15,17 % 14,11 % Humidity difference pelleting (%) 1,50 1,00 0,50 0,00 0,50 1,00 1,50 0,01 % 1,22 % 1,04 % 0,53 % 1,38 % 1,45 % 0,95 % 0,91 % 1,45 % 0,47 % 0,65 % 1,02 % 1,18 % 1,15 % 1,29 % Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

30 30 AMINOTec EDITION 1 NOVEMBER 2018 Grinding injects large amounts of mechanical energy into the product, which will increase product temperature and result in moisture losses, and hammer mills are often used to grind raw materials in feed mills. In order to improve grinding efficiency, these hammer mills feature aspiration systems to help the product pass through the screen. A major Brazilian feed miller, for example, quantifies shrink due to moisture losses in the grinding process in the range of %, depending on climatic conditions (figure 4, personal correspondence). Recycling fines from the aspiration system to the product flow is recommen ded (figure 5, CPM) If not recycled, these fines must also be classified as shrink. FIGURE 05 Hammer mill with aspiration system CONDITIONING, PELLETING & COOLING In order to prepare the mash for the pelleting process, steam, and occassionally water and other liquids are added to the conditioner, to improve pellet quality and increase pellet mill throughput. The best mash temperature for both pelleting and sanitation purposes depends on the feed formulation (whether broiler, swine or ruminant). To achieve this ideal temperature, depending on ambient conditions in the range of 2 4 % of saturated steam are added to the mash in the conditioner. In field tests conducted by Kanagaraj (2014) in India, mash and pellet moisture was checked in the feed mill process in samples taken after the mixer, after the conditioner, after the cooler and from the bagged product. In figure 6 results are summarized for three different climatic conditions, specifically, hot & humid, cold & dry and hot & dry. Clearly, almost all humidity added in the conditioner evaporates from the pellets in the cooler. This is most pronounced with cold and dry ambient conditions, as coolers usually operate with ambient air. Compared to the mash before conditioning, a slight increase in humidity of approximately 0.9 % of the bagged pellets was noted with hot and humid ambient conditions. In both hot & dry climates, as well as cold and dry climates, however, the moisture contents of the bagged product was found to be below the initial moisture contents of the mash before the conditioner. Here a moisture loss of between 1 2 % of moisture loss was identified. This picture is confirmed by data from two feed mills in different regions and climatic conditions in Brazil. While in Rio Grande do Sul there is a mix of slight gains and moderate losses, in the hotter and drier climate of Minas Gerais, moderate to high moisture losses were observed in the cooler (figure 4, personal correspondance) hot & humid cold & dry hot & dry In order to manage and minimize moisture losses in the grinding and pelleting / cooling process, a possible strategy is to check and adjust the moisture of the mash in the main mixer. Moisture that is added in the main mixer will be fully absorbed by the mash, and some of this moisture will be retained throughout the further process. Such moisture, increasing the moisture content of the mash, will also support sanitation processes in the conditioner. Here mash with higher moisture content will better conduct thermal energy into the centre of the mash kernels, and thus accelerate and support effects of sanitation. after mixer after cond after cooler bagged FIGURE 06 Mash and pellet moisture in feed production in different climatic conditions

31 31 Table 6 shows the effect of conditioning temperature and time on the mash at 3 different levels of moisture, specifically 5, 10 and 15 % of moisture. It is obvious that for each conditioning temperature and time, the kill rate of salmonella expressed as a percentage, was at its highest at the highest moisture level. In the cooling process dust is withdrawn from the cooler with the air flow. Table 06: Effect of time, temperature and moisture on the destruction of Salmonella enteritidis in animal feed (Reimann, 1996) from: D. Ziggers, Feed Tech Volume 5, Number 6 TEMPERATURE TIME PERCENT REDUCTION AT DIFFERENT MOISTURE LEVELS ( F) ( C) (seconds) 5 % moisture 10 % moisture 15 % moisture ,22 83,44 90, ,50 86,35 97, ,57 90,80 99, ,36 92,36 98, ,93 96,91 98, ,61 93,49 99, ,92 98,09 99, ,14 99,02 99, ,62 99,12 99,98 During the cooling process, a blower creates an airstrem. A cyclone is then used to separate the dust from the airstream, then recycles the air. Most of the fine dust particles are so tiny, that although they will be separated from the air stream initially, they will later be recycled back into the airstream via the cyclone. According to Fahrenholz (2017) even in high efficiency cyclones, lb of dust emissions may be expected per ton of feed (34 g/ton). With an annual production of mt this would add up to 3 to 4 mt shrink by dust emissions.

32 32 AMINOTec EDITION 1 NOVEMBER 2018 GENERAL REMARKS AND CONCLUSIONS The best way for feed mill managers to keep the issue of shrink under control is to implement meaningful and strict housekeeping routines. Among those are regular checks of equipment, connections, gates for leaks and spillages. Cleaning procedures should be followed and enforced as these help to minimize shrink and avoid losses by pest infestations and rodents that are attracted by spills of organic materials, such as grain or feed mash. YOU SHOULD BE AWARE OF THIS! Spillages of mash attract rodents HOW TO KEEP ISSUE OF SHRINK UNDER CONTROL: 1 COLLECTION AND RECYCLING OF START-UP PRODUCTS FROM THE PELLET MILL We need to define and implement procedures on how to deal with start-up products from the pellet mill, products needed for flushing of production lines and fines collected in aspiration systems etc. Whenever possible, such products should be c ollected and recycled, or used as ingredients in certain formulations, where appropriate.

33 33 2 IMPLEMENTING OF QUALITY CONTROL PROCESSES Sampling and quality control processes also need to be put in place for incoming raw materials, to make sure deviations from specifications are recognized and well documented, be it related to product quality, weight deviations or packaging quality. Only well documented deviations from specifications can be claimed for from suppliers. 3 DOCUMENTATION OF SHRINK RELATED TO CLEANING AND DRYING There is also a need to develop procedures to quantify and document shrink related to the cleaning and drying processes upon receipt of raw materials. Such losses cannot be avoided as grain moisture must be kept in a certain range to ensure safe storage conditions, and, also, to avoid losses through mould growth, insect infestations or germination. 4 IMPLEMENTATION OF APPROPRIATE MOISTURE CONTROL OF THE MASH AND WATER ADDITION Losses of moisture in the feed production processes (namely grinding and cooling) are also unavoidable. But those losses could be minimized by implementing appropriate moisture control of the mash, and, where necessary, adding water into the main mixer to adjust moisture and secure an appropriate level for the further mixing, conditioning and pelleting process. 5 RECONCELING INVENTORY Finally, reconciling inventory (in terms of ingredients and finished feed is an operational and administrative task that requires, and deserves, thorough and continuous attention. Stock taking is a tedious job, but results are only reliable if it is conducted responsibly by welltrained and experienced staff on a daily basis. To reconcile inventories in the accounting system, the amount of beginning stock plus incoming goods must be compared to ending stock plus consumption (how much stock has been used up). The difference between both, is therefore the shrink (or gain ). shrink Shrink (or gain) = (beginning stock + goods received) (ending inventory + consumption) A team effort of operators, accounting staff and managers is most definitely requiredto keep shrink under control. Causes for physical stock differences are multiple, as we have explored in detail in this article. In addition to this, accounting errors may well make life more complicated but addressing this issue would perhaps require a discussion entirely of its own.

34 34 AMINOTec EDITION 1 NOVEMBER 2018 LEADING BRAZILIAN FEED PRODUCTION CHOOSES EVONIK Marco Lara (marco.lara@evonik.com), Regional Manager Handling Solution and Feed Milling Technology, Evonik Degussa Brasil Ltda. Sao Paulo, Brazil ONE OF BRAZIL s TOP PORK PRODUCERS HAS CHOSEN EVONIK s LEADING DOSING SYSTEM TO SECURE AN INCREASE IN ITS FEED PRODUCTIVITY AND QUALITY LEVELS. PICTURE 01 The Alibem Alimentos feed mill and premix plant in Santa Rosa, Rio Grande The Alibem Alimentos feed mill and premix plant in Santa Rosa, Rio Grande do Sul, has implemented Evonik s high precision dosing system, AMINOSys. Customers need to improve the use of, supplemental amino acids and Evonik offers innovative, accurate and reliable solutions that bring proven economic results says Marco Lara, regional manager of handling solutions and feed milling technology. All living organisms require amino acids to synthesize proteins and secure the growth of lean tissue. By using the AMINOSys, Alibem will now be able to better manage handling and high precision dosing of crystalline or granular bulk amino acids, ensuring that their pork products are of the highest possible quality. The flexible AMINOSys system is part of an installation project comprised of consulting, hardware and software solutions which automatically dose amino acids in feed. AMINOSys is designed as a modular system that supports receiving amino acids in big bags, or in full bulk silos. It doses and conveys amino acids directly to the batch mixer or simply replenishes bins of an already existing micro dosing system. AMINOSys IS DESIGNED AS A MODULAR SYSTEM THAT SUPPORTS RECEIVING AMINO ACIDS IN BIG BAGS, OR IN FULL BULK SILOS.

35 35 3-IN-ONE SOLUTION FOR ALIBEM SINCE THE 1940s The Alibem Alimentos group specializes in the slaughter, processing, commercialization and distribution of quality pork and beef products, and produces animal protein in the form of pork (ALIBEM brand) and beef (AGRA brand). It is the second largest swine processing company in the state of Rio Grande do Sul, and the 5th largest in terms of slaughter numbers in the whole of Brazil. As part of the recent installation, three systems LOWER COSTS, HIGHER QUALITY As well as supplying the Brazilian market, Alibem also exports to several countries. Its demand for Evonik s equipment arose due to its rapid growth in recent years, resulting in a significant increase in feed production. The AMINOSys equipment will improve the production process, and the quality and volume of premix produced. It will also increase the company s confidence in dosing accuracy, reducing production costs and optimizing labour, since it has enabled workers to relocate from the premix area to other areas that need support said Fabricio Ruschel, executive director of the agriculture and livestock division. of Evonik s state-of-the art equipment have now been integrated into Alibem s feed production, supporting two compound feed mix lines, and a premix line. Our mission is to satisfy all the means involved, through sustainable growth, ensuring investments in the areas of environment, biosafety, genetics, production and marketing along with providing social development, said Mr. Maximiliano Chang Lee, CEO of the Alibem group. Paulo Teixeira, Business Director for Animal Nutrition in Brazil said the installation of the AMINOSys dosing system will provide to Alibem s business due to the decrease in operational costs, with an impact on the reduction of labour, availability of storage area for other ingredients, and precision in the dosage of amino acids. This installation in Brazil is officially Evonik s 100 th such project in the country. To celebrate this milestone Evonik recently held an inauguration ceremony involving 120 people, including executives from Evonik, Alibem and project partners. After a dinner party at the Santa Rosa Gaucho Traditions Center, the colourful event was brought to life with a wide range of traditional dance performances. LEADING SUPPORT TO THE ANIMAL NUTRITION SECTOR The improvement of feed quality and productivity of production plants are constant challenges for companies involved in the animal production supply chain. Since the 1940s, Evonik has been helping to meet this challenge by producing supplemental amino acids for animal feed. It has also provided optimized solutions for the feed production market. We act according to the new challenges, providing pioneering systems and services of excellence, contributing to help, develop and improve the business of companies, said Rogerio Ott, business manager of Evonik and responsible for Alibem account.

36 36 AMINOTec EDITION 1 NOVEMBER 2018 FIRST-CLASS REPUTATION IN BRAZIL In Brazil, Evonik is recognized by the feed production market as the leading company which supplies a quality dosing solution for dry bulk amino acids. With the first AMINOSys implemented in the 1990s, AMINOSys has gone on to establish itself as one of the most accurate and efficient microdosing solutions for amino acids. Proof of the robustness of the equipment and the AMINOSys solution is that the systems initially installed in Brazil onlyneeded an upgrade in This means they managed to operate reliably and successfully for over 20 years. This serves as testament not just to the quality of Evonik s products and systems, but also to the presence, motivation and dedication of its highly professional team. Their staff are fully committed to training customers, providing prompt support in troubleshooting and corrective maintenance, as well as implementing world-quality preventive maintenance programs. In addition to AMINOSys, Evonik has developed a range of tools to support feed mills such as AMINOBatch and AMINOBatch Working Precision Test, which help improve the blending process. Since they were first implemented, these tools have contributed to a significant improvement in the quality of food produced in Brazil. AMINOSys HAS GONE ON TO ESTABLISH ITSELF AS ONE OF THE MOST ACCURATE AND EFFICIENT MICRO-DOSING SOLUTIONS FOR AMINO ACIDS. Watch the AMINOSys video! front/article/8sqx1s59u