FEED MANUFACTURING TO LOWER FEED COST Presented at the Leman Conference 2007 By Dr. Charles Stark Feed Science and Management Department of Poultry Science, North Carolina State University Raleigh, North Carolina 27695 E‐mail: [email protected] Abstract The increasing demand for cereal grain for ethanol and fat for bio‐diesel production will create a new set of challenges within the animal agriculture sector. Remaining competitive over the next several years will require greater flexibility in both feed mills and animal production units. Purchasing agents, nutritionists, veterinarians, feed mill managers, and animal production operations must develop an integrated approach to purchasing ingredients, formulating, manufacturing, and delivering feed. Animal production and feed manufacturing operations must develop a synergistic relationship in order to meet these new challenges. Feed manufacturing practices that were effective in the past may no longer be beneficial to the animal producer when faced with rising ingredient, energy, and transportation costs. Feed mills and animal production units must evaluate the value of ingredients, particle size, mash vs. pelleted feed, and cost of delivery in today’s market environment. Feed mill managers must cultivate and implement processes that align with the long range goals of their customers’ or own company’s business plan. Long term success of the industry requires goals and objectives that focus on: 1) ingredient purchasing and logistics, 2) feed processing that improves animal performance, and 3) feed delivery and ordering that decreases cost. 1
Introduction The traditional role of feed manufacturing has changed as animal production units have increased in size. The buyer‐seller relationship is disappearing as companies integrate the feed manufacturing process into their operation through ownership of feed mills or through toll milling relationships with commercial feed mills. Developing a feed manufacturing process that is integrated into an animal production unit is a key element to lowering feed cost and remaining competitive. A company’s operating and management philosophy will determine how the feed manufacturing operation will interact with the live production unit. The underlying management philosophy of the company will determine the feed mill manager’s approach to manufacturing costs, employee motivation, and management interaction with production units. The two management philosophy’s most prevalent in the industry tend to be lowering the feed manufacturing costs at the feed mill or using the feed mill as a tool to maximize animal profitability. Focusing on lowering manufacturing costs alone will result in only small savings due to the limited number of variable costs a feed mill manager can directly affect within their operation. Modern large scale feed mills are very efficient and have very few variable costs that can be reduced in order to lower the cost of the final feed. The strategy a company uses to lower feed costs will vary based on the economic risk and benefit model of the company as well as how they interpret research. The key to lowering feed costs is to understand the interrelationships that exist from the time ingredients are purchased to the time the animal consumes the feed. A team comprised of individuals from different areas of responsibility within a company can help identify savings in: 1) ingredient procurement and logistics, 2) feed processing, and 3) feed ordering and delivery. The team should consist of a feed mill manager (commercial or integrated), nutritionist, purchasing agent, veterinarian, and animal production specialist. Once the team is created, a mission statement based on company goals must be developed. The mission statements may be centered around ”lowest cost of production”, “best feed conversion” and/or “highest ranking”. The mission statement must be communicated to employees involved with the process in order to understand and appreciate how decisions in one segment of the company will affect the company or customer feed cost. The first task of the team is to create written procedures that support the goals and objectives for lowering feed costs. Written documents should include standard operating procedures, ingredient specifications, a quality assurance program, and preventive maintenance program. The second step is to implement these procedures and programs through employee training. The final step is conducting follow‐up audits to ensure compliance with the procedures and programs. 2
2.1
Ingredient Purchasing and Logistics Alternative Ingredient Selection Nutritionists, purchasing agents, and feed mill managers must develop an ingredient purchasing strategy that does not negatively impact the feed mill and animal performance. Replacing cereal grains with alternative ingredients and by‐products that can be segregated based on nutrient content at feed mill has the potential to lower feed cost. Reducing the cost of a $150/ton diet by 1% would result in a $1.50 savings whereas it would take a 15% reduction in feed manufacturing cost ($10/ton) to achieve similar savings. Understanding the constraints or limitations of a feed mill prior to purchasing ingredients will limit trucker driver wait times, demurrage, formulation changes, and negative impacts on animal performance. Communication with the feed mill manager, quality assurance laboratory, and production units prior to purchasing new or alternative ingredients will minimize the hidden costs that could be associated with an ingredient change. Nutritionists, purchasing agents, and feed mill managers must recognize and address the following challenges associated with new ingredients: •
•
•
•
•
•
•
•
2.2
Added analytical costs to develop a new ingredient matrix Additional receiving time and logistics management Slower batching times due to additional ingredients or slow flowing bins Reduced inventory of primary ingredients Reduced pellet mill throughput Changes in pellet quality and flow characteristics Changes in feed density Changes in feed palatability Ingredient Specification Sheets Ingredient purchasing specifications provide guidance for purchasing agents, suppliers, transporters, and receiving personnel. Ingredient specifications are the cornerstone to producing a high quality finished feed, limiting product liability, and lowering the cost of feed. Ingredient specifications sheets (Appendix A) should include: 1) Product description, 2) Typical nutrient analysis and reference analytical methods, 3) Physical characteristics, and 4) Basis for rejection. Including the analytical method used for each nutrient in the specification will minimize disputes with suppliers regarding product guarantees. The American Feed Industry Association (AFIA), Renewable Fuels Association (RFA), and National Corn Growers Association (NCGA) recently completed an evaluation of common analytical methods used to determine the nutrient content of DDGS samples and found differences in the percent moisture and crude fat based on different methods (Table 1). Table 1. Summary of moisture and fat results of DDGS samples. Method Description AOAC 934.01 AOAC 935.29 NFTA 2.2.2.5 AOAC 930.15 AOAC 2001.12 Loss on Drying (Vacuum) Loss on Drying (103C/5 hrs) Loss on Drying (105C/3 hrs) Loss on Drying (135C/2 hrs) Moisture (Karl Fischer) Method AOAC 2003.05 AOAC 954.02 AOAC 945.16 AOAC 2003.06 Crude Fat (Ethyl Ether) Fat (Acid Hydrolysis) Crude Fat (Pet Ether) Crude Fat (Hexane) Moisture, % Avg Value 10.67 10.17 9.87 12.69 9.03 StdDev Crude Fat, % Avg Value 9.22 13.03 8.85 9.00 StdDev .25 .15 .18 .19 .08 .28 .57 .24 .19 A feed mill should not accept ingredients that do not meet specification and should have procedures in place for rejecting shipments that do not meet specification. Routine acceptance of ingredients that do not meet specification is a signal to suppliers that a mill is a dumping ground for ingredients that are rejected at other feed mills. The routine also sends the wrong message to employees about the importance of manufacturing quality feed. Receiving off‐specification ingredients can also lead to overtime for bin cleaning, repairing equipment, and disposal of bad product. 2.3
Inbound Ingredient Logistics Feed mills that have the ability to segregate and manage ingredients based on supplier and/or plant location can capture savings in least cost formulation. Simple inexpensive quality assurance tests at the point of receiving can help segregate ingredients based on moisture, protein, fat, and starch. Equipment such as the grain moisture analyzer, NIR, or moisture balance can be used for rapid analysis of nutrients, typically in less than ten minutes. Mycotoxins tests can also be conducted on ingredients prior to unloading if the feed mill suspects a problem based on previous analysis. 2.4
Ingredient Variation Ingredient variation exists due to differences in growing conditions, base raw material, and plant processes. Ingredient matrix nutrient values should be based on samples collected prior to or during the receiving process. Least cost ingredient matrix values should take into account both supplier and plant variation. Growing conditions, drying, processing, and storage methods can affect the nutrient content of both cereal grains and ingredients. Batal and Dale (2006) reported nutrient variation in 17 samples of DDGS obtained from six ethanol plant in the Midwest (Table 2), illustrating the differences that exist between plants. In addition to supplier and plant variation, analytical variation should be taken into account when making purchasing decisions and matrix value adjustments. The energy content of feed fat will vary based on the raw material from which it is derived; in addition to energy content, swine producers should also be aware of differences in the iodine value of fat which influences belly quality (Table 3). The nutrient content of cereal grains will change over time as they are stored in grain elevators and on‐farm bins. Table 2. Nutrient Variability Based on Ethanol Plant TME Kcal/lb TME Kcal/kg Average SD Range 1,282 82 1,132‐1,450 2,820 181 2,490‐3,190 Crude Protein (%) 27 2 23‐30 Crude Fat (%) 8.8 2.3 2.5‐10.6 Crude Fiber (%) 6.6 0.8 5.1‐8.1 Ash (%) 4.4 0.4 3.9‐5.4 Table 3. Energy and iodine value of feed fats based on source. Fat Source Yellow Grease Poultry Fat Choice White Grease Tallow 1
Meeker 2003 3
Poultry ME, kcal/lb1
3,582 3,539 3,424 3,167 Swine ME, kcal/lb1 3,663 3,641 3,585 3,452 Iodine1
Varies 77‐80 63‐65 43‐45 Feed Manufacturing Process 3.1
3.1.1
Grinding Cost of Particle Size Reduction Reducing the particle size of cereal grains is a small fraction of the overall cost of feed manufacturing. Grinding costs are inversely related to particle size; costs increase as the target particle size decreases. The cost of particle size reduction is dependent on the target particle size, type of equipment (hammermill vs. roller mill), and ingredients that must be processed. The additional costs are associated primarily with electricity, hammermill screens, and hammers. Feed mills can shift the particle size in all feeds by simply changing the screen size (hole diameter) and adjusting the hammer setting. Adding a different grind size to a specific formula is similar to adding another ingredient to the feed mill. The extra labor cost in screen changes, grinder downtime, and overtime due to decreased mill efficiency may have a greater impact on manufacturing costs than electricity, hammers, and screens. Grinder selection should be based on the standard ingredient mix, target particle size, and the final feed form (meal vs. pellet). Considerations should also be made for operational costs (electricity, labor, and maintenance) and capital investment. Roller mills produce a granular product and should be considered when producing mash diets. Roller mills will efficiently grind material to 600 – 1,000 microns but are limited to low fiber ingredients. Hammermills can reduce the particle size of ingredients to less than 400 microns when properly sized with the correct air handling system. 3.1.2
Particle Size Determination The mean particle size of ground material is expressed by the geometric mean diameter and log normal geometric standard deviation (ASAE S319). The standard method involves the analysis of the material utilizing 15 sieves and a Ro‐tap® shaker. Managers should determine the particle size of their ground grains on a regular basis at a company, commercial or university laboratory that performs this analysis. 3.1.3
Animal Performance Reducing the particle size of cereal grains improves animal performance (Table 5). Ohh et al. (1983) and Healey et al. (1994) reported improvements in the performance of starter pigs when corn was reduced from 1000 to 500 microns. Healey et al. (1994) observed a negative effect on stomach mucosa at 300 microns. Lawrence et al. (2003) found no affect on starter pig performance when the particle size of solvent extracted SBM was reduced from 1,226 to 444 microns. Wondra et al. (1995) reported reducing particle size from 1000 to 400 microns resulted in an 8% improved in feed/gain (linear effect, P<.01). Callan et al. (2007) reported a better (P<.01) feed conversion when finishing pigs were fed a complete diet ground through a 3 mm vs. 6 mm hammermill screen. Table 5. Summary of particle size research trials. Research Microns
ADG, lbs Ohh et al., 1983 624 1.01 877 0.99 822 1.02 1147 1.04 Healey et al., 1994 300 .93 500 .96 700 .90 900 .94 Lawrence et al., 2003 444 1.06 797 1.07 1,226 1.07 Wondra et al., 1995 400 2.16 600 2.09 800 2.07 1000 2.11 400 2.18 F/G 1.70 1.78 1.81 1.92 1.44 1.41 1.46 1.50 1.53 1.52 1.52 3.22 3.43 3.41 3.39 3.01 Comments Hammermill , Corn, Starter pigs Hammermill, Corn, Starter pigs Roller mill, Corn, Starter pigs Roller mill, Corn, Starter pigs Hammermill, Corn, Starter pigs Hammermill, Corn, Starter pigs Hammermill, Corn, Starter pigs Hammermill, Corn, Starter pigs SBM, Starter pigs SBM, Starter pigs SBM, Starter pigs Hammermill, Corn, Finishing pigs, Meal diet
Hammermill, Corn, Finishing pigs, Meal diet
Hammermill, Corn, Finishing pigs, Meal diet
Hammermill, Corn, Finishing pigs, Meal diet
Hammermill, Corn, Finishing pigs, Pelleted diet
Callan et al., 2007 600 800 1000 2.24 2.22 2.18 2.10 2.02 3.13 3.14 3.32 2.61 2.80 Hammermill, Corn, Finishing pigs, Pelleted diet
Hammermill, Corn, Finishing pigs, Pelleted diet
Hammermill, Corn, Finishing pigs, Pelleted diet
Hammermill 3 mm screen, Feed, Finishing pigs
Hammermill 6 mm screen, Feed, Finishing pigs
Goodband et al. (2002) developed a regression equation to predict feed conversion from 400 to 1,200 microns in finishing pigs from 120 to 240 lbs (Graph 1). Graph 1 F/G = microns x .000415175 + 3.066333 (r=.61; P<.01) The optimal particle size for individual production units must be based on herd health status, genetics, stage of production (sow, nursery, and grow/finish), electrical costs, capital investment, and type of diet (meal vs. pellet). 3.2
Batching Reviewing batching records daily is not only a requirement of the FDA’s current Good Manufacturing Practices (cGMP’s) it also makes economic sense to insure animals are receiving the correct nutrients and medications. Feed mill managers should monitor batching reports for trends in shrink or gain of ingredients on a daily and weekly basis. Batches of feed that have a negative weight variance on ingredients will not provide the proper level of nutrients to the animals. Conversely over addition of ingredients can significantly increase the cost of the feed. From a global perspective a net .5% shrink of ingredients can add $.75/ton ($150/ton feed cost) to the cost of the feed (the cost of shrink is typically added back to the cost of ingredients or the manufacturing cost). On an individual ingredient basis the addition of 5 lbs extra fat at $.20/lb will add $1.00 to the cost of the feed batch. Depending on the size of the batch this could equate to $.10 ‐.50/ton of feed. 3.3
Mixing The mixer should be tested at least once a year using a single source ingredient. Salt or synthetic amino acids are commonly used to test mixer uniformity. Ten samples should be obtained from the mixer by probing the mixer or collecting samples at equally spaced time intervals during the discharge process. The samples should be analyzed for the selected ingredient to determine the uniformity of the mix. The coefficient of variation (CV%) for the ten samples should be less than 10%. Herrman and Behnke (1994) provided guidelines for interpretation of mixer uniformity results and potential corrective actions (Table 6). Table 6. Interpretation of mixer tests. CV < 10% 10-15% 3.4
RATING
Excellent 15‐20% Good Fair 20% + Poor CORRECTIVE ACTION
None
Increase mixing time by 25-30%
Increase mixing time by 50%, look for worn equipment,
overfilling, or sequence of ingredient addition
Possible combination of all the above
Consults extension personnel or feed equipment manufacturer
Pelleting The pelleting process agglomerates ingredients that have different particle sizes, densities, and flowability. Pelleting allows nutritionists to formulate diets that include ingredients with poor flow characteristics, including ground grains less than 600 microns without affecting the flow characteristics of the finished feed. The effectiveness of the pelleting process is measured by pellet quality and percent fines at the mill or in the feeder on the farm. Feed mills that have pellet screeners will remove the fines after the pellets have been cooled and re‐pellet the fines, thus creating a product with minimal fines when it leaves the feed mill. Integrated operations typically do not remove fines and ship product that has both pellets and fines. This feed may contain from 10‐50% fines in the feeder at the farm. Pellet quality can be determined with the standard Pellet Durability Index developed at Kansas State University (ASAE 5269.3). The method may need to be modified through the use of additional hex nuts to create a model that is representative of a company’s manufacturing and delivery process. A pellet quality model should be created for each individual production system. Feed mills that have a method to estimate pellet quality at the feeder can make adjustments in the mill to meet the pellet quality specifications at the feeder. Additionally, the model can provide feedback to nutritionists and purchasing agents as to the negative effect an ingredient or formulation change had on pellet quality. Stark (1994) reported that feeding pelleted diets which contained 60% fines resulted in a feed conversion that was similar to feeding a mash diet, thus negating the benefit of pelleting. The consistency of pelleted feed is as important as the actual amount of fines. Inconsistency in the percentage of pellet fines between feed deliveries requires more feeder management to minimize feed wastage. Establishing a specification for percent fines at the feeder will allow the nutritionist, purchasing agent, and feed mill manager to formulate, purchase, and manufacture feed to the specification while exploring options to lower the cost of the feed. Animal production units that specify 100% pellets at the farm can expect higher manufacturing costs. These costs may be significantly higher if the feed mill is attempting to pellet ingredients with poor pelleting characteristics which require more re‐pelleting of the fines, thus decreasing the efficiency of the pellet mill. 4
4.1
Feed Ordering and Delivery Feed Ordering The feed ordering system used by an animal production company can impact the cost of feed delivered to the animal. Feed ordering systems which allow human inputs and/or decisions have a greater potential for errors as well as increased costs, lower animal performance, and compromised medication strategies. Substituting feed types and not shipping according to the nutritional plan are the most frequent errors in ordering systems that allows human intervention without multiple approval levels. Manufacturing problems, ingredient delivery issues, and emergency farm orders will always occur within a company; however written policies and procedures for managing these issues will minimize the feed cost associated with substituting the wrong feed. Delivering one truck load of a formula that costs $10/ton more than the formula specified in the nutritional plan will add $250 to the cost of that delivery and thus add $0.25/pig (1,000 pig barn) in feed costs. 4.2
Feed Delivery Feed delivery costs can increase significantly when trucks are not fully loaded to the legal weight limit (Table 7 and 8). The design and tare weight of the feed delivery equipment will determine the total weight of feed that can be hauled on a load. Selecting the correct type and design of equipment should be based on the condition of the roads, distance to farms, and operating philosophy of the company. Table 7. Transportation cost based on tons delivered to a barn for a long distance delivery. Tons Delivered Cost/Load, $ Delivery Cost, $/ton Additional Cost, $/ton 21 250 11.90 2.65 23 250 10.87 1.61 25 250 10.00 0.74 27 250 9.26 0 Table 8. Transportation cost based on tons delivered to a barn for a short distance delivery. Tons Delivered Cost/Load, $ Delivery Cost, $/ton 21 100 4.76 23 100 4.35 25 100 4.00 27 100 3.70 Additional Cost, $/ton 5
1.06 0.64 0.30 0 Conclusion Remaining competitive requires flexibility in both feed mills and animal production units. Companies must develop a purchasing, manufacturing, and delivery model based on their organizational goals and objectives. The key to lowering feed costs is to understand the interrelationships that exist from the time ingredients are purchased to the time the animal consumes the feed. Feed mills may vary in their design, cost, and efficiency. However, the common denominator in feed mills must be standard operating procedures, ingredient specifications, quality assurance, and preventive maintenance programs that support animal production operations. 6
References AAFCO. 2006. 2006 Official Publication. Association of American Feed Control Officials Incorporated. AFIA. 2007. Evaluation of Analytical Methods for Analysis of Dried Distillers Grains with Solubles. American Feed Industry Association, Arlington VA. ASAE. 1983. Method of determining and expressing fineness of feed materials by sieving. ASAE Standard S319, Agriculture Engineers Yearbook of Standards. St. Joseph, MI. ASAE . 1987. Wafers, pellets, and crumbles‐definitions and methods for determine density, durability, and moisture content. Yearbook of Standards. P.325. American Society of Agricultural Engineers. Batal, A.B. and N.M. Dale. 2006. True metabolizable energy and amino acid digestibility of distillers dried grains with soluble. J. Appl. Poult. Res. 15:89‐93. Callan, J.J., B.P. Garry, and J.V. O'Doherty. 2007. The effect of expander processing and screen size on nutrient digestibility, growth performance, selected faecal microbial populations and faecal volatile fatty acid concentrations in grower–finisher. Animal Feed Science and Technology, 134 (3), p.223‐234. Goodband, R.D., R.D. Tokach, and J.L. Nelssen. 2002. The effects of diet particle size on animal performance. Kansas State University Extension Bulletin, MF‐2050. Healy, B.J., J.D. Hancock, G.A. Kennedy, P.J. Bramel‐Cox, K.C. Behnke, and R.H. Hines. 1994. Optimum particle size of corn and hard and soft sorghum for nursery pigs. J. Anim. Sci. 72:2227‐2236. Herrman , T. and K.C. Behnke. 1994. Testing Mixer Performance. Kansas State University Extension Bulletin, MF‐1172. Lawrence, K.R., C.W. Hastad, R.D. Goodband, M.D. Tokach, S.S. Dritz, J.L. Nelssen, J.M. DeRouchey, and M.J. Webster. 2003. Effects of soybean meal particle size on growth performance of nursery pigs. J. Anim. Sci. 2003. 81:2118‐2122. Meeker, D.L. and C.R. Hamilton. 2006. An overview of the rendering industry. Essential Rendering. D.L. Meeker, ed. National Renderers Assn. Arlington, VA. pp 1‐16. Ohh, S.J., G. Allee, K.C. Behnke, and C.W. Deyoe. 1983. Effect of particle size of corn and sorghum grain on performance and digestibility of nutrients for weaned pigs. J. Anim. Sci. 57(Suppl. 1):260 (Abstr.). Wondra, K.J., J.D. Hancock, K.C. Behnke, R.H. Hines, and C.R. Stark. 1995. Effect of particle size and pelleting on growth performance, nutrient digestibility, and stomach morphology in finishing pigs. J. Anim. Sci. 73:757‐763. Appendix A Example Ingredient Specification Sheet Distillers Dried Grains w/ Solubles AAFCO PRODUCT DESCRIPTION: Distillers dried grains with solubles is the product obtained after the removal the of ethyl alcohol by distillation from the yeast fermentation of a grain or a grain mixture by condensing and drying at least ¾ of the solids of the resultant whole stillage by methods employed in the grain distilling industry. AAFCO #27.6 IFN #5‐02‐843 Typical Nutrient Analysis Analytical Methods Moisture max 12% ‐ NFTA 2.2.2.5, Loss on Drying (105C/3 hrs) Protein min 25% ‐ AOAC 990.03, Protein ‐ Combustion Fat min 8% ‐ AOAC 945.16, Crude Fat (Pet Ether) Fiber max 10% ‐ AOAC 978.10, Crude Fiber (F.G. Crucible) Ash 5% Aflatoxin max 20 ppb Physical Properties Color: golden to tan color Odor: sweet odor Bulk Density: 32 – 36 lbs/ft3 Sieve: 90% through #10 US Basis for Rejection 1. Transportation method does not meet the FDA Regulations Governing the Transportation of Animal Proteins Prohibited From Use in Ruminant Feed. 2. Product is adulterated or misbranded. 3. Contains product that got wet during shipping. 4. Aflatoxin above 20 ppb. NCSU Feed Mill QA Manual