US20060036419A1

US20060036419A1 - System and method for animal production optimization

Info

Publication number: US20060036419A1
Application number: US10/902,504
Authority: US
Inventors: David Cook; Daniel Barziza; Steve Burghardi; Gregory Engelke; Donald Giesting; John Hahn; Brian Knudson; Wade Martinson; Bruce McGoogan; Michael Messman; Mark Newcomb; Jennifer Ligt
Original assignee: CAN Technologies Inc
Current assignee: Cargill Inc; CAN Technologies Inc
Priority date: 2004-07-29
Filing date: 2004-07-29
Publication date: 2006-02-16
Also published as: US20060041412A1; WO2006015017A2; KR20070038173A; ZA200701705B; AU2005269279A1; JP2008508607A; BRPI0513775A; MX2007001076A; CN101014973A; CA2573897A1; RU2007107397A; EP1776664A2; US7904284B2; US7827015B2; ZA200701704B; CA2573901A1; US20060041413A1; RU2399289C2; KR20070055522A; BRPI0513797A

Abstract

A system for generating optimized values for variable inputs to an animal production system. The system includes a simulator engine configured to receive a plurality of animal information inputs and generate a performance projection, wherein at least one of the animal information inputs is designated as a variable input and an enterprise supervisor engine configured to generate an optimized value for the at least one variable input based on at least one optimization criteria and an animal feed formulation.

Description

BACKGROUND OF THE INVENTION

The present invention relates generally to the field of systems for and methods of animal production. More particularly, the present invention relates to systems for and methods of optimizing an animal production system based on one or more optimization criteria.
An animal production system may include any type of system or operation utilized in producing animals or animal based products. Examples may include farms, ranches, aquaculture farms, animal breeding facilities, etc. Animal production facilities may vary widely in scale, type of animal, location, production purpose, etc. However, almost all animal production facilities can benefit from identifying and implementing improvements to production efficiency. Improvements to production efficiency can include anything that results in increased production results, improved proportional output of desired products versus less desirable products (e.g. lean vs. fat), and/or decreased production costs.
A producer (i.e. a farmer, rancher, aquaculture specialist, etc.) generally benefits from maximizing the amount or quality of the product produced by an animal (e.g. gallons of milk, pounds of meat, quality of meat, amount of eggs, nutritional content of eggs produced, amount of work, hair/coat appearance/health status, etc.) while reducing the cost for the inputs associated with that production. Exemplary inputs may include animal feed, animal facilities, animal production equipment, labor, medicine, etc.
In order to maximize animal production over time, almost any input may be treated as a variable input. For example, the contribution of almost any input may be increased, decreased, or changed in some other way over time. For example, additional animal feed may be obtained, additional facilities may be constructed, additional labor may be hired, etc.
Every variable input may further be associated with one or more effects of variation. For example, for almost every variable input, an increase in the amount of the variable input is associated with an increase in the cost of the variable input. In a specific example, constructing additional facilities may be associated with building costs, financing costs, maintenance costs, etc. Additionally, the increase in the amount of the variable input is associated with an increase in the benefit provided by the variable input. Returning to our earlier example, the construction of the additional facilities may be associated with an increase in the number of animals that may be produced at the facility, or a reduction in animal crowding that will increase the production of each animal, etc.
What is needed is a system for and method of receiving inputs related to an animal production facility and processing the inputs to determine the effect of modifications to one or more of the inputs. What is further needed is such a system and method where the inputs are related to animal environment, animal type, animal feed ingredients, animal health, animal metabolic status, and/or animal economic data. Still further, what is needed is a system for and method of determining optimized inputs related to an animal production facility based on the minimization or maximization of an objective criteria.

SUMMARY OF THE INVENTION

One embodiment of the invention relates to a system for generating optimized values for variable inputs to an animal production system. The system includes a simulator engine configured to receive a plurality of animal information inputs and generate a performance projection, wherein at least one of the animal information inputs is designated as a variable input. The system further includes an enterprise supervisor engine that is configured to generate an optimized value for the at least one variable input based on at least one optimization criteria and an animal feed formulation.
Another embodiment of the invention relates to a method for determining optimized values for inputs to an animal production system. The method includes receiving a plurality of animal information inputs, wherein at least one of the animal information inputs is designated as a variable input. The method further includes generating at least one performance projection based on the animal information inputs and generating an optimized value for the at least one variable input based on the at least one performance projection and an animal feed formulation and at least one optimization criteria.
Yet another embodiment of the invention relates to a system for generating an animal feed formulation. The system includes a simulator engine configured to receive a plurality of animal information inputs and generate animal requirements based on the animal information inputs, a formulator engine, the formulator engine configured to receive a plurality of animal feed ingredient inputs and generate at least one animal feed formulation composed of the animal feed ingredients based on the animal requirements, wherein at least one of the animal feed ingredient inputs is designated as a variable input, and an enterprise supervisor engine configured to optimize the at least one animal feed formulation according to at least one optimization criteria, and further configured to generate an optimized value for the at least one variable input based on the at least one optimization criteria.
Yet another embodiment of the invention relates to an animal production optimization system. The system includes an optimization engine, having an objective function program therein, configured to receive a feed formulation input provided to the optimization engine. The system further includes an animal production modeling system configured to receive animal information input, including at least one variable input, receive feed formulation input, and provide modeling output to the optimization engine. The optimization engine optimizes the objective function to provide an optimized solution for the at least one variable input based on the modeling output.
Yet another embodiment of the invention relates to a method for generating optimized values for variable inputs to an animal production optimization system. The method includes the steps of receiving animal information input, including at least one variable input, generating modeling output based on the animal information input, receiving a feed formulation input to the objective function, and generating an objective function based on the modeling output and the feed formulation input. The method further includes optimizing the objective function to provide an optimized value for the at least one variable input.
Other features and advantages of the present invention will become apparent to those skilled in the art from the following detailed description and accompanying drawings. It should be understood, however, that the detailed description and specific examples, while indicating preferred embodiments of the present invention, are given by way of illustration and not limitation. Many modifications and changes within the scope of the present invention may be made without departing from the spirit thereof, and the invention includes all such modifications.

BRIEF DESCRIPTION OF THE DRAWINGS

The exemplary embodiments will hereafter be described with reference to the accompanying drawings, wherein like numerals depict like elements, and:
FIG. 1 is a general block diagram illustrating an animal production optimization system, according to an exemplary embodiment;
FIG. 2 is a general block diagram illustrating an enterprise supervisor for an animal production optimization system, according to an exemplary embodiment;
FIG. 3 is a general block diagram illustrating a simulator for an animal production system, according to an exemplary embodiment;
FIG. 4 is a general block diagram illustrating an ingredients engine and a formulator for an animal production system, according to an exemplary embodiment; and
FIG. 5 is a flowchart illustrating a method for animal production optimization, according to an exemplary embodiment.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

In the following description, for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present invention. It will be evident to one skilled in the art, however, that the exemplary embodiments may be practiced without these specific details. In other instances, structures and devices are shown in diagram form in order to facilitate description of the exemplary embodiments.
In at least one exemplary embodiment illustrated below, a computer system is described which has a central processing unit (CPU) that executes sequences of instructions contained in a memory. More specifically, execution of the sequences of instructions causes the CPU to perform steps, which are described below. The instructions may be loaded into a random access memory (RAM) for execution by the CPU from a read-only memory (ROM), a mass storage device, or some other persistent storage. In other embodiments, multiple workstations, databases, processes, or computers can be utilized. In yet other embodiments, hardwired circuitry may be used in place of, or in combination with, software instructions to implement the functions described. Thus, the embodiments described herein are not limited to any particular source for the instructions executed by the computer system.
Referring now to FIG. 1, a general block diagram is shown illustrating an animal production optimization system 100, according to an exemplary embodiment. System 100 includes an enterprise supervisor 200, a simulator 300, an ingredient engine 400, and a formulator 500.
System 100 may be implemented utilizing a single or multiple computing systems. For example, where system 100 is implemented using a single computing system, each of enterprise supervisor 200, simulator 300, ingredient engine 400, and formulator 500 may be implemented on the computing system as computer programs, discrete processors, subsystems, etc. Alternatively, where system 100 is implemented using multiple computers, each of enterprise supervisor 200, simulator 300, ingredient engine 400, and formulator 500 may be implemented using a separate computing system. Each separate computing system may further include hardware configured for communicating with the other components of system 100 over a network. According to yet another embodiment, system 100 may be implemented as a combination of single computing systems implementing multiple processes and distributed systems.
System 100 is configured to receive animal information input including at least one variable input and analyze the received information to determine whether variation in one or more of the variable input will increase animal productivity or satisfy some other optimization criteria. Animal productivity may be a relative measure of the amount, type, or quality of output an animal produces relative to the expense associated with that production. Animal information input can include any type of information associated with an animal production system. For example, animal information input may be associated with a specific animal or group of animals or type of animals, an animal's environment, an economy related to the animal production, etc. Animal productivity may further be configured to include positive and negative outputs associated with the production. For example, animal productivity may be configured to represent harmful gaseous emissions as an expense (based on either financial costs associated with clean up or the negative impact on the environment), reducing the overall productivity.
Information associated with a specific animal or a group or type of animals may include, but is not limited to, a species, a state, an age, a production level, a job, a size (e.g. current, target, variability around, etc.), a morphology (e.g. intestinal), a body mass composition, an appearance, a genotype, a composition of output, a collection of microbial information, health status, a color, etc. The information associated with a specific animal may be any type of information relevant for determining the productivity of the animal.
Species information can include a designation of any type or class of animals such as domestic livestock, wild game, pets, aquatic species, humans, or any other type of biological organism. Livestock may include, but is not limited to, swine, dairy, beef, equine, sheep, goats, and poultry. Wild game may include, but is not limited to, ruminants, such as deer, elk, bison, etc., game birds, zoo animals, etc. Pets may include, but are not limited to, dogs, cats, birds, rodents, fish, lizards, etc. Aquatic species may include, but are not limited to, shrimp, fish (production), frogs, alligators, turtles, crabs, eels, crayfish, etc. and include those species grown for productive purposes (e.g., food products).
Animal state may include any reference or classification of animals that may affect the input requirement or production outputs for an animal. Examples may include, but are not limited to, a reproductive state, including gestation and egg laying, a lactation state, a health state or stress level, a maintenance state, an obese state, an underfed or restricted-fed state, a molting state, a seasonal-based state, a compensatory growth, repair or recovery state, a nutritional state, a working or athletic or competitive state, etc. Animal health states or stress level may further include multiple sub-states such as normal, compromised, post-traumatic (e.g. wean, mixing with new pen mates, sale, injury, transition to lactation, etc.), chronic illness, acute illness, immune response, an environmental stress, etc.
Animal age may include an actual age or a physiological state associated with an age. Examples of physiologic states may include a developmental state, a reproductive state including cycles, such as stage and number of pregnancies, a lactation state, a growth state, a maintenance state, an adolescent state, a geriatric state, etc.
Animal job may include a physiologic state as described above, such as gestation, lactation, growth, etc. Animal job may further include the animal's daily routine or actual job, especially with reference to canine and equines. Animal job may also include an animal movement allowance, such as whether the animal is generally confined versus allowed free movement in a pasture, or, for an aquatic animal, the different water flows the aquatic animal experiences, etc.
Animal size may include the actual weight, height, length, circumference, body mass index, mouth gape, etc. of the animal. The animal size may further include recent changes in animal size, such as whether the animal is experiencing weight loss, weight gain, growth in height or length, changes in circumference, etc.
Animal morphology includes a body shape exhibited by an animal. For example, a body shape may include a long body, a short body, a roundish body, etc. Animal morphology may further include distinct measurement of internal organ tissue changes like the length of intestinal villi or depth of intestinal crypts.
Animal body mass composition may include a variety of composition information such as a fatty acid profile, a vitamin E status, a degree of pigmentation, a predicted body mass composition, etc. The body mass composition generally is a representation of the percentage or amount of any particular component of body mass, such as lean muscle, water, fat, etc. The body mass composition may further include separate representations composition for individual body parts/sections. For example, body mass composition may include edible component compositions such as fillet yield, breast meat yield, tail meat yield, etc.
Animal appearance may include any measure or representation of an animal appearance. Examples can include the glossiness of an animal's coat, an animal's pigmentation, muscle tone, etc.
Animal genotype may include any representation of all or part of the genetic constitution of an individual or group. For example, an animal genotype may include DNA markers associated with specific traits, sequencing specific segments of DNA, etc. For example, the genotype may define the genetic capability to grow lean tissue at a specific rate or to deposit intramuscular fat for enhanced leanness or marbling, respectively. Additionally, genotype may be defined by phenotypic expression of traits linked to genotypic capacity such as the innate capacity for milk production, protein accretion, work, etc.
Composition of output may include the composition of a product produced by an animal. For example, the composition of output may include the nutrient levels found in eggs produced by poultry or milk produced by dairy cows, the amount, distribution, and/or composition of fat in meat products, etc.
Microbial and/or enzyme information may include current microbial populations within an animal or within an animal's environment. The microbial and/or enzyme information may include measures of the quantity or proportion of gram positive or negative species or other classifications such as aerobes, anaerobes, salmonella species, E. coli species, etc. Enzyme information may include the current content, quantity and/or composition of any enzyme, such as protease, amylase, and/or lipase, produced by the pancreas, produced within the gastrointestinal tract, enzymes produced by a microbial population, etc. Microbial and/or enzyme information may further include information about potential nutritional biomass represented by the microbial community that may be used as a feed source for some species (e.g., ruminants, aquatic species, etc.). The microbial and/or enzymatic environment may be monitored using any of a variety of techniques that are known in the art, such as cpn60, other molecular microbiological methods, and in vitro simulation of animal systems or sub-systems.
Animal information input associated with an animal or group of animals' environment may include, but is not limited to, factors related specifically to the environment, factors related to the animal production facility, etc. Animal environment may include any factors not associated with the animal that have an effect on the productivity of the animal or group of animals.
Examples of animal information input related to the environment may include ambient temperature, wind speed or draft, photoperiod or the amount of daylight exposure, acclimation, seasonal effects, air quality, water quality, water flow rate, aeration rate, system substrate, filter surface area, filtration loan capacity, geographic location, mud score, etc. The environmental information may further include detailed information regarding the system containing the animal or animals, such as system size (e.g. the size in square meters, hectares, acres, volume, etc.), system preparation such as using liming, discing, etc., aeration rate, system type, etc. Although some environmental factors are beyond the control of a producer, the factors can usually be modified or regulated by the producer. For example, the producer may reduce draft by closing vents, raise ambient temperature by including heaters or even relocating or moving certain animal production operations to a better climate for increasing productivity. According to another example, an aqua producer may modify nutrient input to an aquatic environment by altering a feed design or feeding program for the animals in the environment. According to an exemplary embodiment, animal information input related to the environment may be generated automatically using an environmental appraisal system (EAS) to calculate a thermal impact estimate for an animal and to provide measurements for the animal's current environment.
Examples of animal information input related to a production facility may include animal density, animal population interaction, feeder type, feeder system, feeder timing and distribution, pathogen loads, bedding type, type of confinement, feathering, lighting intensity, lighting time patterns, etc. Animal information input for a production facility may be modified by a producer to increase productivity or address other production goals. For example, a producer may build additional facilities to reduce population density, obtain additional or different types of feeding systems, modify the type of confinement, etc.
Animal information input associated with economic factors may include, but is not limited to, animal market information. Animal market information may include, but is not limited to, historical, current and/or projected prices for outputs, market timing information, geographic market information, product market type (e.g., live or carcass-based), etc.
Animal information inputs may further include any of a variety of inputs that are not easily classifiable into a discrete group. Examples may include an animal expected output (e.g., milk yield, product composition, body composition, etc.), a user defined requirement, a risk tolerance, an animal mixing (e.g., mixing different animals), variations with an animal grouping, etc., buyer or market requirements (e.g. Angus beef, Parma hams, milk for particular cheeses, etc.), expected and/or targeted growth curves, survival rates, expected harvest dates, etc.
The above described animal information input may include information that is directly received from a user or operator through a user interface, as will be described below with reference to FIG. 2. Alternatively, the animal information input or some part of the input may be retrieved from a database or other information source.
Further, some of the inputs may be dependent inputs that are calculated based on one or more other inputs or values. For example, an animal's stress level may be determined or estimated based on population density, recent weight loss, ambient temperature, etc. Each calculated value may include an option enabling a user to manually override the calculated value.
Yet further, each animal information input may include a variety of information associated with that input. For example, each animal information input may include one or more subfields based on the content of the animal information input. For example, where an indication is provided that an animal is in a stressed state, subfields may be received indicating the nature and severity of the stress.
According to an exemplary embodiment, the animal information input includes a capability to designate any of the animal information inputs as a variable input. A variable input may be any input that a user has the ability to modify or control. For example, a user may designate ambient temperature as a variable input based on the ability to modify the ambient temperature through a variety of methods such as heating, cooling, venting, etc. According to an alternative embodiment, system 100 may be configured to automatically recommend specific animal information inputs as variable inputs based on their effect on productivity or satisfying the optimization criteria, as will be further discussed below with reference to FIG. 2.
Designation of a variable input may require submission of additional information, such as a cost and/or benefit of variation of the variable input, recommended degrees of variation for optimization testing, etc. Alternatively, the additional information may be stored and retrievable from within system 100 or an associated database.
The animal information inputs may further include target values as well as current values. A target value may include a desirable level for animal productivity or some aspect of animal productivity. For example, a producer may wish to target a specific nutrient level for eggs produced by poultry. Therefore, the producer may enter current nutrient levels for eggs currently being produced as well as target nutrient values for the eggs. According to another example, a current size breakdown for shrimp in a pond versus a potential size breakdown. The target values and current values may be utilized by system 100 to make changes in an animal feed formulation or to make changes to variable inputs as will be described further below. Further, the target values may be viewed as equality constraints and/or inequality constraints for the optimization problem.

Table 1 below lists exemplary animal information inputs that may be provided as inputs to animal production optimization system 100. This listing of potential animal information inputs is exemplary and not exclusive. According to an exemplary embodiment, any one or more of the listed animal information inputs can be designated as a variable input.

TABLE 1


General Characteristics

Impact of the ration on the greater	Quantity and/or composition	Quantity and/or
environment:	(e.g. nitrogen, phosporus,	composition of urine
	etc.) of manure per animal
	Quantity and/or quality of
	odor from facility

Swine Characteristics

Sow reproductive performance

No. of pigs born	No. of pigs born alive	No. of pigs weaned
Piglets birth weight	Uniformity of baby pigs	Mortality of baby pigs
Piglets weaning weight	Sow body condition score	Sow lactation back fat loss
Sow lactation weight loss	Interval weaning to estrus	Sow longevity

Working boar

Body condition score

Working frequency

Semen quality

Finisher

Average daily gain	Average daily lean gain	Average daily feed intake
		per weight gain
Average daily feed intake per	Feed wastage	Feed form
lean gain
Mortality	Days to market	Feed cost per kg gain
Feed cost per kg lean gain	Medication usage per pig	Dressing percentage
Lean percentage	Back fat thickness	Fatty acid composition

Evaluation Criteria for Environment

Thermal environment (Draft,	Air quality (Dust, Humidity,	Pig/pen
Floor type, Bedding, Insulation)	Ammonia, Carbon dioxide,
	etc)
Pig density	Health condition	Feeder type
Pigs/feeder hole	Water quality and quantity

Evaluation Criteria for appearance

Hair coat condition	Skin color	Ham shape
Body shape and length

Evaluation Criteria for meat/fat quality

Meat and fat color	Iodine value	Fatty acid profile
PSE	Juiciness	Flavor
Tenderness	Marbling score	Water holding capacity

Evaluation Criteria for Health

Suckling piglets

Eye condition (dry and dirty or	Skin condition (elastic or	Hair condition (dense or
bright and vital eyes)	dry) and color (pink or	coarse)
	pale)
Dirtiness of around anus	Breathe with open mouth	Belly condition

Finisher

Respiratory disease	Body temperature	Cannibalism (tail, ear, belly
		biting)
Skin and hair condition (mange	Stool condition	Swollen knee and ankle
and parasites)		joint
Dirtiness of around eyes	Nose condition	Respiratory sound
		(Difficulties in breathing)
Activity

Sows

MMA (Mastitis, Metritis,	Stool condition	Abortion and stillbirth
Agaclactia)	(constipation)
Wet belly	Body shaking	Vaginal and uterine
		prolapse
Body condition score	Interval weaning to estrus	Feed intake (sick sows eat
		less)
Leg problem	Body temperature

Dairy Characteristics

Cow reproductive performance

Breeding per conception	Live birth	Days to first estrous
Calf birth weight	Days open	Days to cleaning
Calf weaning weight	Cow body condition score	MUN and BUN
Cow body reserve change	Calving interval	Blood hormones
		progesterone and estrogen

Lactation

Milk per day	Body fatty acid loss or gain	Average daily feed intake
		per kg milk
Average daily feed intake per	Feed wastage	Feed form
kg milk
Mortality	Lactation length	Feed cost per kg milk
Milk per year and lifetime milk	Morbidity	Body amino acid loss or
		gain
Fatty acid composition of milk
(CLA, EPA and DHA, 18:2 to
18:3 ratio of milk)

Evaluation Criteria for Environment

Thermal environment (Draft,	Air quality (Dust, Humidity,	Blood cortisol, NEFA
Floor type, Bedding, Insulation)	Ammonia, Carbon dioxide,
	etc)
Animal density	Health condition	Feed presentation method
Cows per bunk or waterer	Water quality and quantity	Cow care and comfort
space		score card

Evaluation Criteria for appearance

Hair coat condition	Skin color	Body condition score
Body shape and length	Color of mucus memberance	Appearence of eyes and
		ears

Evaluation Criteria for milkquality

Milk color	Milk protein composition	Milk fat yield
Milk flavor	Milk lactose	Milk protein yield
Milk fatty acid composition	Total milk solids

Evaluation Criteria for Health

Calves

Heifers

Respiratory disease	Body temperature
Skin and hair condition (mange	Stool condition	Swollen knee and ankle
and parasites)		joint
Dirtiness of around eyes	Nose condition	Respiratory sound
		(Difficulties in breathing)
Activity

Cows

Mastitis, Metritis	Stool condition	Abortion and stillbirth
	(constipation) manure
	screener
Blood measures EX: cortisol,	Body shaking	Vaginal and uterine
NEFA, BHBA, alkaline		prolapse
phosphitase, progesterone
estrogen bun
Body condition score	Calving interval	Feed intake (sick cows eat
		less)
Leg problem	Body temperature	Milk urea nitrogen

Companion Animal and Equine Characteristics

Hair coat shine	Hair coat-fullness	Skin scale/flake level
Fecal consistency	Gas production	Breath
Immune status	Antioxidant status	Body condition (thin,
		normal, obese)
skeletal growth rate	endurance	Digestive health status
Circulatory health status	Hoof quality	Hair quality
Body fluid status	Workload (NRC specifies
	light, medium and heavy
	workloads)

Characteristics to optimize for athlete animals:

Speed	Sprint	Muscular glycogen spare
Muscular glycogen recovery	Decrease recovery time	Endurance
	after exercise
Body condition

Health and Welfare of the Animal:

Welfare and behavior (calmer or	Relationship between	Dry matter intake
energetic diet):	NDF/starch or forage/grain
	Long fiber intake	Electrolytes
General health status:	Low allergenicity	Digestive health
	Improving inmunologic	Increasing antioxidant
	status	status
	Minimize digestive upset
Immunologic status

Beef Characteristics

Cow reproductive performance

Conception rate	Weaning rate	Calf birth weight
Calf mortality	Calf weaning weight	Cow body condition score
Interval weaning to estrus	Calving interval

Bulls

Body condition score

Breeding Soundness

Growing

Average daily lean gain	Average daily feed intake	Feed cost per unit gain
	per lean gain
Feed cost per unit lean gain

Evaluation Criteria for Environment

Air quality

Nutrient excretion

Evaluation Criteria for appearance

Hair coat condition

Height

Height/weight ratio

Evaluation Criteria for meat/fat quality

Meat and fat color	Fatty acid profile	Juiciness
Flavor	Tenderness	Marbling score
Dressing percentage	Red meat yield	Muscle pH
Intra muscular fat	Antioxidant status

Evaluation Criteria for Health

Mortality

Medication cost

Morbidity

Poultry Characteristics

Egg and reproduction

Egg number	Fertility	Hatchability
Egg weight	Egg mass	Egg internal quality
		(Haugh Units)
Egg yolk color	Eggshell quality	Egg bacteriological content
		(Salmonella-fee)
Fertile eggs breakout analysis

Performance

Average daily gain	Average daily feed intake	Feed conversion
Mortality	Occurrences of Leg	Feed cost per kg gain live
	problem	weight
Feed cost per dozen eggs	Yield of Eviscerated	Yield of body parts
	carcass	(breast, thigh, back etc.)
Flock Uniformity	Feed consumption

Environment

Temperature	Air quality (Dust,	Bird density
	Humidity, Ammonia,
	Carbon dioxide, etc)
Feeder space	Lighting program	Water quality and quantity
Litter quality (Wet droppings)	Biosecurity

Evaluation Criteria for appearance

Feathering score	Skin color	Skin scratching score
Feed appearance (color, texture,
etc.)

Aquaculture Animal Characteristics

Initial weight	Size variability	Developmental stage
Target weight	Stocking density	Body composition (or meat
		composition)
Body condition	Animal or meat color	Survival rate
Feedings per day	Feeding activity	Swimming Speed
Feed water stability	Desired shelf-life	Specific growth rate
Meat yield (e.g., fillet, tail meat,	Mouth gape	Cost per unit gain
etc.)
FCR	Days to market	Genotype
Pigmentation	Feed Consumption	Harvest Biomass
Number of days to “X” animal	$ cost/unit weight gain	% of yield of target
size		product (shrimp tails, fillet,
		etc.)
$ profit/unit production biomass	Return on investment	Cycles per year
$ of feed/unit weight of	$ of feed/$ of biomass	Total harvest biomass
production
% of animals in target size range	Mortality or survival rate	Specific growth rate
Average animal size	$ of profit/unit of culture	Average weight gain/week
	area or volume
Weight of production/unit of	Product shelf life
aeration

Aquaculture Environmental Characteristics

System type	Ammonia, pH, dissolved	Water flow rate
	oxygen, alkalinity, temp.,
	hardness, etc.
Water exchange rate	Nutrient load	Natural productivity
		biomass (species specific
		forage base)
Population health	Environmental pathogen	Temperature, oxygen, etc.
	load	variability
System Substrate	Water Filtration Rate	Feed on feeding tray
Total Filtration Capacity	Photoperiod	Processing form for feed
(Mechanical and Chemical)
Medicine application	Aeration rate	Water exchange rate
Aeration pattern	Feeding tray # and	Feed distribution pattern
	positioning
Secchi disc reading

Referring now to the components of system 100, supervisor 200 may be any type of system configured to manage the data processing function within system 100 to generate optimization information, as will be further discussed below with reference to FIG. 2. Simulator 300 may be any type of system configured to receive animal information or animal formulation data, apply one or more models to the received information, and generate performance projections such as animal requirements, animal performance projections, environmental performance projections, and/or economic performance projections as will be further discussed below with reference to FIG. 3. Ingredient engine 400 may be any kind of system configured to receive a list of ingredients and generate ingredient profile information for each of the ingredients including nutrient and other information. Formulator 500 may be any type of system configured to receive an animal requirements projection and ingredient profile information and generate animal formulation data, as will be further discussed below with reference to FIG. 4.
Referring now to FIG. 2, a general block diagram illustrating an enterprise supervisor 200 for an animal production optimization system 100 is shown, according to an exemplary embodiment. Enterprise supervisor 200 includes a user interface 210 and an optimization engine 230. Enterprise supervisor 200 may be any type of system configured to receive animal information input through user interface 210, submit the information to simulator 300 to generate at least one animal requirement, submit the at least one animal requirement to formulator 500 to generate least cost animal feed formulation given the animal requirement, submit the optimized formulation to simulator 300 to generate a performance projection and to utilize optimization engine 230 to generate optimized values for one or more variable inputs.
According to an alternative embodiment, optimization or some portion of the optimization may be performed by a different component of system 100. For example, optimization described herein with reference to supervisor 200 may alternatively be performed by simulator 300. Further, optimization of animal feed formulation may be performed by formulator 500.
Enterprise supervisor 200 may include or be linked to one or more databases configured to automatically provide animal information inputs or to provide additional information based upon the animal information inputs. For example, where a user has requested optimization information for a dairy production operation, enterprise supervisor 200 may be configured to automatically retrieve stored information regarding the user's dairy operation that was previously recorded to an internal database and also to download all relevant market prices or other relevant information from an external database or source.
User interface 210 may be any type of interface configured to allow a user to provide input and receive output from system 100. According to an exemplary embodiment, user interface 210 may be implemented as a web based application within a web browsing application. For example, user interface 210 may be implemented as a web page including a plurality of input fields configured to receive animal information input from a user. The input fields may be implemented using a variety of standard input field types, such as drop-down menus, text entry fields, selectable links, etc. User interface 210 may be implemented as a single interface or a plurality of interfaces that are navigable based upon inputs provided by the user. Alternatively, user interface 210 may be implemented using a spreadsheet based interface, a custom graphical user interface, etc.
User interface 210 may be customized based upon the animal information inputs and database information. For example, where a user defines a specific species of animal, enterprise supervisor 200 may be configured to customize user interface 210 such that only input fields that are relevant to that specific species of animal are displayed. Further, enterprise supervisor 200 may be configured to automatically populate some of the input fields with information retrieved from a database. The information may include internal information, such as stored population information for the particular user, or external information, such as current market prices that are relevant for the particular species as described above.
Optimization engine 230 may be a process or system within enterprise supervisor 200 configured to receive data inputs and generate optimization information based on the data inputs and at least one of the optimization criteria. According to an exemplary embodiment, optimization engine 230 may be configured to operate in conjunction with simulator 300 to solve one or more performance projections and calculate sensitivities in the performance projection. Calculating sensitivities in the performance projections may include identifying animal information input or variable inputs that have the greatest effect on overall productivity or other satisfaction of the optimization criteria. Optimization engine 230 may further be configured to provide optimized values for the animal information inputs or variable inputs based on the sensitivity analysis. Optimization may include any improvement to productivity or some other measure according to the optimization criteria. The process and steps in producing the optimized values are further discussed below with reference to FIG. 5.
Optimization criteria may include any criteria, target, or combination of targets or balanced goals that are desirable to the current user. In a preferred embodiment, the optimization criteria is maximizing productivity. Maximizing productivity may include maximizing a single or multiple factors associated with productivity such as total output, output quality, output speed, animal survival rates, etc. Maximizing productivity may further include minimizing negative values associated with the productivity, such as costs, harmful waste, etc. Alternative optimization criteria may include profitability, product quality, product characteristics, feed conversion rate, survival rate, growth rate, biomass/unit space, biomass/feed cost, cost/production day, cycles/year, etc. Alternatively, the optimization criteria may include minimizing according to an optimization criteria. For example, it may be desirable to minimize the nitrogen or phosphorus content of animal excretion.
Optimization engine 230 maybe configured to implement its own optimization code for applications where feed ingredient information from formulator 500 is combined with other information and/or projections calculated in simulator 300. Optimization problems that coordinate several independent calculation engines, referred to as multidisciplinary optimizations, may be solved using gradient-based methods, or more preferably simplex methods such as Nelder-Mead or Torczon's algorithm. Preferably, optimization engine 230 may be configured to implement a custom combination of a gradient-based method for variables on which the optimization criteria depends smoothly (decision variables fed to simulator 300) and a simplex method for variables on which the objective function has a noisy or discontinuous dependence (diet requirements fed to formulator 500). Alternatively, other optimization methods may be applied, including but not limited to, pseudo-gradient based methods, stochastic methods, etc.
Enterprise supervisor 200 may be further configured to format the optimization results and provide the results as output through user interface 210. The results may be provided as recommended optimized values for the variable inputs. The results may further include recommended values for additional animal information inputs, independent of whether the animal information input was designated as a variable input. The results may further include a projection of the effects of implementation of the optimized values for the variable inputs.
Enterprise supervisor 200 may be configured to implement a Monte Carlo method where a specific set of values is drawn from a set of distributions of model parameters to solve for optimized values for the variable inputs. This process may be repeated many times, creating a distribution of optimized solutions. Based on the type of optimization, enterprise supervisor 200 maybe used to select either the value most likely to provide the optimal solution or the value that gives confidence that is sufficient to meet a target. For example, a simple optimization might be selected which provides a net energy level that maximizes the average daily gain for a particular animal. A Monte Carlo simulation may provide a distribution of requirements including various net energy levels and the producer may select the net energy level that is most likely to maximize the average daily gain.
Enterprise supervisor 200 may further be configured to receive real world feedback based on the application of the optimized values for the variable inputs. The real world feedback may be compared to the performance projections, further discussed below with reference to FIG. 3. Real world feedback can be provided using any of a variety of methods such as automated monitoring, manual input of data, etc.
Further, enterprise supervisor 200 may be configured to enable dynamic control of models. After setting an initial control action, for example the feed formulation, as will be discussed below with reference to FIG. 5, the animal response may be monitored and compared with the prediction. If the animal response deviates too far from the prediction, a new control action, e.g., feed formulation, may be provided. For example, if the performance begins to exceed prediction, some value may be recovered by switching to a less costly feed formulation, different water flow rate, etc. If performance lags prediction, switching to higher value feed formulation, may help to ensure that the final product targets are met. Although the control action is described above with reference to a feed formulation, the control action may be for any control variable, such as water flow rate, feeding rate, etc. Similarly, the adjustments may be made to that control variable, such as by increasing or decreasing the flow rate, etc.
Referring now to FIG. 3, a general block diagram illustrating a simulator 300 is shown according to an exemplary embodiment. Simulator 300 includes a requirements engine 310, an animal performance simulator 320, an environment performance simulator 330, and an economic performance simulator 340. Generally, simulator 300 may be any process or system configured to apply one or more models to input data to produce output data. The output data may include any performance projection, such as animal requirements and/or performance projections, including animal performance projections, economic performance projections, environmental performance projections, etc.
Specifically, simulator 300 is configured to receive animal information input from enterprise supervisor 200, process the information using requirements engine 310 and an animal requirements model to produce a set of animal requirements. Further, simulator 300 may be configured to receive feed formulation data from enterprise supervisor 200 and process the feed formulation data using any combination of animal performance simulator 320, environment performance simulator 330, and economic performance simulator 340 to produce at least one performance projection.
An animal requirements model, used by simulator 300 to convert input values into one or more outputs, may consist of a system of equations that, when solved, relate inputs like animal size to an animal requirement like protein requirement or a system requirement like space allotment or feed distribution. A specific mathematical form for the model is not required, the most appropriate type of model may be selected for each application. One example is models developed by the National Research Council (NRC), consisting of algebraic equations that provide nutrient requirements based on empirical correlations. Another example is MOLLY, a variable metabolism-based model of lactating cow performance developed by Prof. R. L. Baldwin, University of California-Davis. A model may consist of a set of explicit ordinary differential equations and a set of algebraic equations that depend on the differential variables. A very general model may consist of a fully implicit, coupled set of partial differential, ordinary differential, and algebraic equations, to be solved in a hybrid discrete-continuous simulation.
A model may be configured to be independent of the functionality associated with simulator 300. Independence allows the model and the numerical solution algorithms to be improved independently and by different groups.
Preferably, simulator 300 may be implemented as an equation-based process simulation package in order to solve a wide variety of models within system 100. Equation-based simulators abstract the numerical solution algorithms from the model. This abstraction allows model development independent from numerical algorithms development. The abstraction further allows a single model to be used in a variety of different calculations (steady-state simulation, dynamic simulation, optimization, parameter estimation, etc.). Simulators may be configured to take advantage of the form and structure of the equations for tasks such as the sensitivity calculations. This configuration allows some calculations to be performed more robustly and/or efficiently than is possible when the model is developed as a block of custom computer code. An equation-based process simulation package is software configured to interact directly with the equations that make up a model. Such a simulator typically parses model equations and builds a representation of the system of equations in memory. The simulator uses this representation to efficiently perform the calculations requested, whether steady-state simulations, dynamic simulations, optimization, etc. An equation-based process simulation package also allows incorporation of calculations that are more easily written as combination of procedures and mathematical equations. Examples may include interpolation within a large data table, calling proprietary calculation routines distributed as compiled code for which equations are not available, etc. As newer and better solution algorithms are developed, these algorithms may be incorporated into simulator 300 without requiring any changes to the models simulator 300 is configured to solve.
According to an exemplary embodiment, simulator 300 may be a process simulator. Process simulators generally include a variety of solution algorithms such as reverse mode automatic differentiation, the staggered corrector method for variable sensitivities, automatic model index reduction, robust Newton iteration for solving nonlinear systems from poor initial values, error-free scaling of variable systems, and the interval arithmetic method for locating state events. Process simulators utilize sparse linear algebra routines for direct solution of linear systems. The sparse linear algebra routines can efficiently solve very large systems (hundreds of thousands of equations) without iteration. Process simulators further provide a particularly strong set of optimization capabilities, including non-convex mixed integer non-linear problems (MINLPs) and global variable optimization. These capabilities allow simulator 300 to solve optimization problems using the model directly. In particular, the staggered corrector algorithm is a particularly efficient method for the sensitivities calculation, which is often the bottleneck in the overall optimization calculation.
Variable inputs for optimization to be solved by simulator 300 may include both fixed and time-varying parameters. Time varying parameters are typically represented as profiles given by a set of values at particular times using a specific interpolation method, such as piecewise constant, piecewise linear, Bezier spline, etc.
Simulator 300 and the associated models may be configured and structured to facilitate periodic updating. According to an exemplary embodiment, simulator 300 and the associated models may be implemented as a dynamic link library (DLL). Advantageously, a DLL may be easily exported but not viewed or modified in any structural way.
Requirements engine 310 may be any system or process configured to receive animal information input and generate animal requirements by applying one or more requirements models to the set of animal information input. A requirements model may be any projection of potential outputs based upon any of a variety of set of inputs. The model may be as simple as a correlation relating milk production to net energy in an animal feed or as complex as a variable model computing the nutrient requirement to maximize the productivity of a shrimp aquaculture pond ecosystem. Requirements engine 310 may be configured to select from a plurality of models based on the animal information inputs. For example, requirements engine 310 may include models for swine requirements, dairy requirements, companion animal requirements, equine requirements, beef requirements, general requirements, poultry requirements, aquaculture animal requirements, etc. Further, each model may be associated with a plurality of models based on an additional categorization, such as developmental stage, stress level, etc.
Animal requirements generated by requirements engine 310 may include a listing of nutrient requirements for a specific animal or group of animals. Animal requirements may be a description of the overall diet to be fed to the animal or group of animals. Animal requirements further may be defined in terms of a set of nutritional parameters (“nutrients”). Nutrients and/or nutritional parameters may include those terms commonly referred to as nutrients as well as groups of ingredients, microbial measurements, indices of health, relationships between multiple ingredients, etc. Depending on the degree of sophistication of system 100, the animal requirements may include a relatively small set of nutrients or a large set of nutrients. Further, the set of animal requirements may include constraints or limits on the amount of any particular nutrient, combination of nutrients, and/or specific ingredients. Advantageously, constraints or limits are useful where, for example, it has been established at higher levels of certain nutrients or combination of nutrients could pose a risk to the health of an animal being fed. Further, constraints may be imposed based on additional criteria such as moisture content, palatability, etc. The constraints may be minimums or maximums and may be placed on the animal requirement as a whole, any single ingredient, or any combination ingredients. Although described in the context of nutrients, animal requirements may include any requirements associated with an animal, such as space requirements, heating requirements, etc.
Additionally, animal requirements may be generated that define ranges of acceptable nutrient levels. Advantageously, utilizing nutrient ranges allows greater flexibility during animal feed formulation, as will be described further below with reference to FIG. 3.
Requirements engine 310 may be further configured to account for varying digestibility of nutrients. For example, digestibility of some nutrients depends on the amount ingested. Digestibility may further depend on the presence or absence of other nutrients, microbes and/or enzymes, processing effects (e.g. gelatinization, coating for delayed absorption, etc.), animal production or life stage, previous nutrition level, etc. Simulator 300 may be configured to account for these effects. For example, simulator 300 may be configured to adjust a requirement for a particular nutrient based on another particular nutrient.
Requirements engine 310 may also be configured to account for varying digestion by an animal. Animal information inputs may include information indicating the health of an animal, stress level of an animal, reproductive state of an animal, methods of feeding the animal, etc. as it affects ingestion and digestion by an animal. For example, the stress level of an animal may decrease the overall feed intake by the animal, while gut health may increase or decrease a rate of passage.

Table 2 below includes an exemplary listing of nutrients that may be included in the animal requirements. According to an exemplary embodiment, within the animal requirements, each listed nutrient may be associated with a value, percentage, range, or other measure of amount. The listing of nutrients may be customized to include more, fewer, or different nutrients based on any of a variety of factors, such as animal type, animal health, nutrient availability, etc.

TABLE 2


Nutrients Suitable for Generating Animal Requirements

ADF	Animal Fat	Ascorbic Acid
Arginine (Total and/or Digestible)	Ash	Biotin
Calcium	Calcium/Phos ratio	Chloride
Choline	Chromium	Cobalt
Copper	Cystine (Total and/or	Dry Matter
	Digestible)
Fat	Fiber	Folic Acid
Hemicellulose	Iodine	Iron
Isoleucine (Total and/or	Lactose	Lasalocid
Digestible)
Leucine (Total and/or Digestible)	Lysine (Total and/or	Magnesium
	Digestible)
Manganese	Margin	Methionine (Total and/or
		Digestible)
Moisture	Monensin	NDF
NEg (Net Energy Gain)	NEl (Net Energy Lactation)	NEm (Net Energy
		Maintenance)
NFC (Non-Fiber Carbohydrate)	Niacin	Phenylalanine (Total
		and/or Digestible)
Phosphorus	Phosphate	Potassium
Protein	Pyridoxine	Rh Index (Rumen Health
		Index)
Riboflavin	Rough NDF	Rum Solsug (Rumen
		Soluble Sugars)
Rumres NFC (Ruminant Residual	RUP (Rumen	Salt
Non-Fiber Carbohydrate)	Undegradable Protein)
Selenium	Simple Sugar	Sodium
Sol RDP (Soluble Rumen	Sulfur	Sw ME (Metabolizable
Degradable Protein)		Energy)
Thiamine	Threonine (Total and/or	Total RDP
	Digestible)
Tryptophan (Total and/or	Valine (Total and/or	Vitamin A
Digestible)	Digestible)
Vitamin B12	Vitamin B6	Vitamin D
Vitamin E	Vitamin K	Zinc
Gut Health Index

Requirements engine 310 may be configured to generate the animal requirements based on one or more requirement criteria. Requirement criteria can be used to define a goal for which the requirement should be generated. For example, exemplary requirement criteria can include economic constraints, such as maximizing production, slowing growth to hit the market, or producing an animal at the lowest input cost.
The requirements engine 310 may further be configured to generate the animal requirements based on one or more dynamic nutrient utilization models. Dynamic nutrient utilization may include a model of the amount of nutrients within an animal feed that are utilized by an animal based on information received in the animal information inputs, such as animal health, feeding method, feed form (mash, pellets, extruded, etc.), water stability of feed, uneaten food, etc.
Animal performance simulator 320 may be a process or system including a plurality of models similar to the models described above with reference to requirements engine 310. The models utilized in animal performance simulator 320 receive an animal feed formulation from formulator 300 through enterprise supervisor 200 and the animal information inputs and apply the models to the feed formulation to produce one or more animal performance projections. The animal performance projection may be any predictor of animal productivity that will be produced given the animal feed formulation input and other input variables.
Environment performance simulator 330 may be a process or system including a plurality of models similar to the models described above with reference to requirements engine 310. The models utilized in environment performance simulator 330 receive animal feed formulation from formulator 300 through enterprise supervisor 200 and apply the models to the feed formulation and animal information inputs to produce a performance projection based on environmental factors. The environmental performance projection may be any prediction of performance that will be produced given the animal feed formulation input, animal information inputs, and environmental factors.
Economic performance simulator 340 may be a process or system including a plurality of models similar to the models described above with reference to requirements engine 310. The models utilized in economic performance simulator 340 receive animal feed formulation from formulator 300 through enterprise supervisor 200 and apply the models to the feed formulation and animal information inputs to produce a performance projection based on economic factors. The economic performance projection may be any prediction of performance that will be produced given the animal feed formulation input, animal information inputs, and the economic factors.
The performance projections may include a wide variety of information related to outputs produced based on the provided set inputs. For example, performance projections may include information related to the performance of a specific animal such as the output produced by an animal. The output may include, for example, the nutrient content of eggs produced by the animal, qualities associated with meat produced by the animal, the contents of waste produced by the animal, the effect of the animal on an environment, etc.
According to exemplary embodiment, simulators 320, 330, and 340 may be run in parallel or in series to produce multiple performance projections. The multiple animal performance projections may remain separated or be combined into a single comprehensive performance projection. Alternatively, performance projections may be generated based on a single simulator or a combination of less than all of the simulators.
Requirements engine 310 may further include additional simulators as needed to generate performance projections that are customized to satisfy a specific user criteria. For example, requirements engine 310 may include a bulk composition simulator, egg composition simulator, meat fat composition, waste output simulator, etc.
Referring now to FIG. 4, a general block diagram illustrating an ingredients engine 400 and a formulator 500 is shown, according to an exemplary embodiment. Ingredients engine 400 is configured to exchange information with formulator 500. Ingredients engine 400 and formulator 500 are generally configured to generate an animal feed formulation based on available ingredients and received animal requirements.
Ingredients engine 400 includes one or more listings of available ingredients at one or more locations. The listing further includes additional information associated with the ingredients, such as the location of the ingredient, nutrients associated with the ingredient, costs associated with the ingredient, etc.
085 Ingredients engine 400 may include a first location listing 410, a second ingredient location listing 420, and a third ingredient location listing 430. First ingredient listing 410 may include a listing of ingredients available at a first location, such as ingredients at a user's farm. The second ingredient listing 420 may include a listing of ingredients that are available for purchase from an ingredient producer. Third ingredient listing 430 may include a listing of ingredients that are found in a target animal's environment such as forage in a pasture, plankton, zoo plankton, or small fish in an aquaculture pond, etc.
Referring now to third ingredient listing 430, an example of a listing of ingredients that are found in a target animal's environment may include a listing of the mineral content of water. An animal's total water consumption can be estimated based on known in consumption ratios, such as a ratio of water to dry feed matter consumed. This ratio may be either assigned an average value or, more preferably, calculated from known feed and animal properties. The mineral content of the water provided by producer may be measured on-site. This water, with measured mineral content and calculated intake level, may be incorporated as third ingredient listing 430.
Alternatively, third ingredient listing 430 may include an aquatic ecosystem total nutrient content. The ecosystem contribution to total nutrition may be included in several ways. For example, a sample may be drawn and analyzed for total nutrient content and included as third listing 430. Preferably, the models solved in simulator 300 may be expanded to include not only that species being produced but other species that live in the ecosystem as well. The model may include one or more of the following effects: other species competition for feed, produced species consumption of other species in ecosystem, and other species growth over time in response to excretion, temperature, sunlight, etc.
Third ingredient listing 430 may further include performance projections generated by simulator 300. For example, the nutrient content of milk may be modeled for the particular animals for an individual producer. This milk nutrient content model may be used as a third ingredient listing 430 for consumption by a nursing animal.
Each listing of ingredients may further include additional information associated with the ingredients. For example, a listing of ingredients may include a listing of costs associated with that ingredient. Alternatively, an ingredient at the first location may include a costs associated with producing the ingredient, storing the ingredient, dispensing the ingredient, etc., while an ingredient at the second location may include a cost associated with purchasing the ingredient, and an ingredient at the third location may include a cost associated with maintaining the environment. The additional information may include any type of information that may be relevant to later processing steps.

Table 3 below includes an exemplary list of ingredients which may be used in generating the animal feed formulation. The listing of ingredients may include more, fewer, or different ingredients depending on a variety of factors, such as ingredient availability, entry price, animal type, etc.

TABLE 3


Exemplary Ingredients Suitable for
Use in Formulating Custom Feed Mixes

Acidulated Soap Stocks	Active Dry Yeast	Alfalfa Meal
Alfalfa-Dehydrated	Alimet	Alka Culture
Alkaten	Almond Hulls	Ammonium Chloride
Ammonium Lignin	Ammonium Polyphosphate	Ammonium Sulfate
Amprol	Amprol Ethopaba	Anhydrous Ammonia
Appetein	Apramycin	Arsanilic Acid
Ascorb Acid	Aspen Bedding
Avizyme	Bacitracin Zinc	Bakery Product
Barley	Barley-Crimped	Barley-Ground
Barley-Hulless	Barley-Hulls	Barley-Midds
Barley-Needles	Barley-Rolled
Barley-Whole	Barley-With Enzyme	Baymag
		Beet
Beet Pulp	Biotin	Biscuit By Product
Black Beans	Blood-Flash Dry
Bone Meal	Brewers Rice	Brix Cane
Buckwheat		Cage Calcium
Calcium Cake	Calcium Chloride	Calcium Formate
Calcium Iodate	Calcium Sulfate	Calcium Prop
	Canadian Peas	Cane-Whey
Canola Cake	Canola Fines	Canola Meal
Canola Oil	Canola Oil Blender	Canola Oil Mix
Canola Screenings	Canola-Whole	Carbadox
Carob Germ	Carob Meal	Cashew Nut Byproduct
Catfish Offal Meal	Choline Chloride	Chromium Tripicolinate
Citrus Pulp	Clopidol	Cobalt
Cobalt Carbonate	Cobalt Sulfate	Cocoa Cake
Cocoa Hulls	Copper Oxide	Copper Sulfate
Corn Chips	Corn Chops	Corn Coarse Cracked
Corn-Coarse Ground	Corn Cob-Ground	Corn Distillers
Corn Flint	Corn Flour	Corn Germ Bran
Corn Germ Meal	Corn Gluten	Corn-High Oil
Corn Kiblets	Corn Meal Dehulled	Corn Oil
Corn Residue	Corn Starch	Corn/Sugar Blend
Corn-Cracked	Corn-Crimped	Corn-Ground Fine
Corn-Ground Roasted	Corn-Steam Flaked	Corn-Steamed
Corn-Whole	Cottonseed Culled	Cottonseed Hull
Cottonseed Meal	Cottonseed Oil	Cottonseed Whole
Coumaphos	Culled Beans	Danish Fishmeal
Decoquinate	Dextrose	Diamond V Yeast
Disodium Phosphate	Distillers Grains	Dried Apple Pomace
Dried Brewers Yeast	Dried Distillers Milo	Dried Porcine
Dried Whole Milk Powder	Duralass	Enzyme Booster
Epsom Salts		Extruded Grain
Extruded Soy Flour	Fat	Feather Meal
Feeding Oatmeal	Fenbendazole	Fermacto
Ferric Chloride	Ferrous Carbonate	Ferrous Carbonate
Ferrous Sulfate	Fine Job's Tear Bran	Fish Meal
Fish	Flavoring	Folic Acid
	Fresh Arome	Fried Wheat Noodles
Gold Dye	Gold Flavor	Grain Dust
Grain Screening	Granite Grit	Grape Pomace
Green Dye	Green Flavor	Guar Gum
Hard Shell	Hemicellulose Extract
Herring Meal	Hominy	Hygromycin
Indian Soybean Meal	Iron Oxide-Red	Iron-Oxide Yellow
Job's Tear Broken Seeds		Kelp Meal
Kem Wet	Lactose	Larvadex
Lasalocid	Levams Hcl	Limestone
Linco	Lincomix	Lincomycin
Linseed Meal	Liquid Fish Solubles	Lupins
Lysine	Magnesium	Magnesium Sulfate
Malt Plant By-Products	Manganous Ox	Maple Flavor
Masonex	Meat And Bone Meal	Meat Meal
Mepron	Methionine	Millet Screenings
Millet White	Millet-Ground	Milo Binder
Milo-Coarse Ground	Milo-Cracked	Milo-Whole
Mineral Flavor	Mineral Oil	Mixed Blood Meal
Molasses	Molasses Blend	Molasses Dried
Molasses Standard Beet	Molasses Standard Cane	Molasses-Pellet
Mold	Monensin	Monoamonum Phos
Monosodium Glutamate	Monosodium Phosphate	Mung Bean Hulls
Mustard Meal High Fat	Mustard Oil	Mustard Shorts
Narasin	Natuphos	Niacin
Nicarbazin	Nitarsone	Oat Cullets
Oat Flour	Oat Groats	Oat Hulls
Oat Mill Byproducts	Oat Screenings	Oat Whole Cereal
Oatmill Feed	Oats Flaked	Oats-Ground
Oats-Hulless	Oats-Premium	Oats-Rolled
Oats-Whole	Oyster Shell	Paddy Rice
Palm Kernel	Papain	Papain Enzyme
Paprika Spent Meal	Parboiled Broken Rice	Pea By-Product
Pea Flour	Peanut Meal	Peanut Skins
Pelcote Dusting	Phosphate	Phosphoric Acid
Phosphorus	Phosphorus Defluorinated	Pig Nectar
	Poloxalene	Popcorn
Popcorn Screenings	Porcine Plasma; Dried	Pork Bloodmeal
Porzyme	Posistac	Potassium Bicarbonate
Potassium Carbonate	Potassium Magnesium	Potassium Sulfate
	Sulfate
Potato Chips	Poultry Blood/Feather	Poultry Blood Meal
	Meal
Poultry Byproduct	Predispersed Clay	Probios
Procain Penicillen	Propionic Acid	Propylene Glycol
pyran Tart	Pyridoxine	Quest Anise
Rabon	Rapeseed Meal	Red Flavor
Red Millet	Riboflavin	Rice Bran
Rice By-Products Fractions	Rice Dust	Rice Ground
Rice Hulls	Rice Mill By-Product	Rice Rejects Ground
Roxarsone	Rumen Paunch	Rumensin
Rye	Rye Distillers	Rye With Enzymes
Safflower Meal	Safflower Oil	Safflower Seed
Sago Meal	Salinomycin	Salt
Scallop Meal	Seaweed Meal	Selenium
Shell Aid	Shrimp Byproduct	Silkworms
Sipernate	Sodium Acetate	Sodium Benzoate
Sodium Bicarbonate	Sodium Molybdate	Sodium Sesquicarbonate
Sodium Sulfate	Solulac
Soy Flour	Soy Pass	Soy Protein Concentrate
Soybean Cake	Soybean Curd By-Product	Soybean Dehulled Milk By-Product
Soybean Hulls	Soybean Mill Run	Soybean Oil
Soybean Residue	Soybeans Extruded	Soybeans-Roasted
Soycorn Extruded	Spray Dried Egg	Standard Micro Premix
Starch Molasses	Steam Flaked Corn	Steam Flaked Wheat
Sugar (Cane)	Sulfamex-Ormeto	Sulfur
Sunflower Meal	Sunflower Seed	Tallow Fancy
Tallow-Die	Tallow-Mixer	Tapioca Meal
Tapioca Promeance	Taurine	Terramycin
Thiabenzol	Thiamine Mono	Threonine
Tiamulin	Tilmicosin	Tomato Pomace
Trace Min	Tricalcium Phosphate	Triticale
Tryptophan	Tryptosine	Tuna Offal Meal
Tylan	Tylosin	Urea
Vegetable Oil Blend	Virginiamycin	Vitamin A
Vitamin B Complex	Vitamin B12	Vitamin D3
Vitamin E	Walnut Meal	Wheat Bran
Wheat Coarse Ground	Wheat Germ Meal	Wheat Gluten
Wheat Meal Shredded	Wheat Millrun	Wheat Mix
Wheat Noodles Low Fat	Wheat Red Dog	Wheat Starch
Wheat Straw	Wheat With Enzyme	Wheat-Ground
Wheat-Rolled	Wheat-Whole	Whey Dried
Whey Permeate	Whey Protein Concentrate	Whey-Product Dried
Yeast Brewer Dried	Yeast Sugar Cane	Zinc
Zinc Oxide	Zoalene

Ingredient engine 400 may further include an ingredient information database 440. Ingredient information database 440 may include any kind of information related to ingredients to be used in generating the feed formulation, such as nutrient information, cost information, user information, etc. The information stored in database 440 may include any of a variety of types of information such as generic information, information specifically related to the user, real-time information, historic information, geographically based information, etc. Ingredient information database 440 may be utilized by ingredient engine 400 to supply information necessary for generating an optimized feed formulation in conjunction with information supplied by the user.
Ingredient information database 440 may further be configured to access external databases to acquire additional relevant information, such as feed market information. Feed market information may similarly include current prices for ingredient, historical prices for output, ingredient producer information, nutrient content of ingredient information, market timing information, geographic market information, delivery cost information, etc. Ingredient information database 440 may further be associated with a Monte Carlo type simulator configured to provide historical distributions of ingredient pricing and other information that can be used as inputs to other components of system 100.
Ingredient engine 400 may further include a variable nutrient engine 450 configured to provide tracking and projection functions for factors that may affect the nutrient content of an ingredient. For example, variable nutrient engine 450 may be configured to project the nutrient content for ingredients over time. The nutrient content for some ingredients may change over time based on method of storage, method of transportation, natural leaching, processing methods, etc. Further, variable nutrient engine 450 may be configured to track variability in nutrient content for the ingredients received from specific ingredient producers to project a probable nutrient content for the ingredients received from those specific ingredient producers.
Variable nutrient engine 450 may be further configured to account for variability in nutrient content of ingredients. The estimation of variability of an ingredient may be calculated based on information related to the particular ingredient, the supplier of the ingredient, testing of samples of ingredient, etc. According to exemplary embodiment, recorded and/or estimated variability and covariance may be used to create distributions that are sampled in a Monte Carlo approach. In this approach, the actual nutrient content of ingredients in an optimized feed formulation are sampled repeatedly from these distributions, producing a distribution of nutrient contents. Nutrient requirements may then be revised for any nutrients for which the nutrient content is not sufficient. The process may be repeated until the desired confidence is achieved for all nutrients.
Referring now to formulator 500, formulator 500 is configured to receive animal requirements from simulator 300 through enterprise supervisor 200 and nutrient information from ingredients engine 400 based on available ingredients and generate an animal feed formulation. Formulator 500 calculates a least-cost feed formulation that meets the set of nutrient levels defined in the animal requirements.
The least-cost animal feed formulation may be generated using linear programming optimization, as is well-known in the industry. The least-cost formulation is generally configured to utilize a users available ingredients in combination with purchased ingredients to create an optimized feed formulation. More specifically, the linear programming will incorporate nutrient sources provided by a user such as grains, forages, silages, fats, oils, micronutrients, or protein supplements, as ingredients with a fixed contribution to the total feed formulation. These contributions are then subtracted from the optimal formulation; the difference between the overall recipe and these user-supplied ingredients constitute the ingredient combinations that would be produced and sold to the customer.
Alternatively, the formulation process may be performed as a Monte Carlo simulation with variability in ingredient pricing included as either historical or projected ranges to created distribution which are subsequently optimized as described above.
Referring now to FIG. 5, a flowchart illustrating a method 600 for animal production optimization is shown, according to an exemplary embodiment. Method 600 generally includes identifying optimized values for one or more animal information inputs according to at least one optimization criteria. Although the description of method 600 includes specific steps and a specific ordering of steps, it is important to note that more, fewer, and/or different orderings of the steps may be performed to implement the functions described herein. Further, implementation of a step may require reimplementation of an earlier step. Accordingly, although the steps are shown in a linear fashion for clarity, several loop back conditions may exist.
In a step 605, enterprise supervisor 200 is configured to receive the animal information inputs. The animal information inputs can be received from a user through user interface 210, populated automatically based on related data, populated based on stored data related to the user, or received in a batch upload from the user. The received animal information inputs include a designation of one or more of the animal information inputs as a variable input. The designation as a variable input may be received for single, multiple, or all of the animal information inputs.
In a step 610, enterprise supervisor 200 is configured to receive an optimization criteria through user interface 210 or, alternatively, receive a preprogrammed optimization criteria. The optimization criteria may include maximizing productivity, reducing expenses, maximizing quality of output, achieving productivity targets, etc. In an exemplary embodiment, the optimization criteria may be an objective function requiring minimization or maximization. The objective function may have constraints incorporated therein or may be subject to independent constraints. The objective function may be a function of any combination of variables of the animal production system.
In a step 615, enterprise supervisor 200 is configured to communicate the animal information inputs and optimization criteria to simulator 300. Upon receiving the animal information inputs and optimization criteria, simulator 300 is configured to generate a set of animal requirements in a step 620.
In a step 625, the set of animal requirements are communicated from simulator 300 through enterprise supervisor 200 to formulator 500. Formulator 500 is configured to generate a least cost animal feed formulation based upon the animal requirements and nutrient information received from nutrient engine 450 in a step 630.
In a step 635, enterprise supervisor 200 is configured to generate optimized values for the one or more variable inputs received in step 605, as discussed in detail above with reference to FIG. 2.
Although specific functions are described herein as being associated with specific components of system 100, functions may alternatively be associated with any other component of system 100. For example, user interface 210 may alternatively be associated with simulator 300 according to an alternative embodiment.
Many other changes and modifications may be made to the present invention without departing from the spirit thereof. The scope of these and other changes will become apparent from the appended claims.

Claims

1. A system for generating optimized values for variable inputs to an animal production system, comprising:

a simulator engine configured to receive a plurality of animal information inputs and generate a performance projection, wherein at least one of the animal information inputs is designated as a variable input; and

an enterprise supervisor engine configured to generate an optimized value for the at least one variable input based on at least one optimization criteria and an animal feed formulation.

2. The system of claim 1, further including a formulator engine, the formulator engine configured to receive animal feed ingredient information and generate the animal feed formulation composed of the animal feed ingredients based on the performance projection.

3. The system of claim 1, wherein generating an optimized value for the at least one variable input includes providing a projected effect of the modification to the at least one variable input.

4. The system of claim 1, wherein the variable input is one of an animal factor, an environmental factor, and an economic factor.

5. The system of claim 4, wherein the variable input is an environmental factor and includes at least one of a thermal environment, an animal population environment, a photoperiod environment, a feeding environment, an animal comfort environment, an animal structural environment, and an animal microbial environment.

6. The system of claim 4, wherein the variable input is an animal factor and includes at least one of an animal species type, an animal job, an animal genotype, and an animal composition.

7. The system of claim 4, wherein the variable input is an economic factor and includes at least one of a market timing, an animal feed price, an animal product valuation, and a market location.

8. The system of claim 1, wherein the simulator engine includes an animal performance simulator configured to generate an animal performance profile based upon the animal feed formulation information and the animal information input including at least one variable input.

9. The system of claim 8, wherein the enterprise supervisor engine is configured to actuate the simulator engine based upon variations in the variable input to generate a plurality of animal performance profiles.

10. The system of claim 9, wherein the enterprise supervisor is further configured to select an optimized value for the at least one variable input based on application of the at least one optimization criteria to the plurality of animal performance profiles.

11. The system of claim 10, wherein the optimization criteria includes at least one of weight gain per day, weight per week, value of product per cost of feed, value of product per unit of space, survival rate, environmental impact, cycles per year, animal size, stocking density, water exchange rate, total biomass of production, and product form.

12. The system of claim 1, wherein the simulator engine further includes an animal environment performance simulator configured to generate an animal performance profile based upon variations in a variable input associated with the environment for at least one animal.

13. The system of claim 12, wherein the enterprise supervisor engine is configured to actuate the simulator engine based upon variations in the variable input to generate a plurality of animal performance profiles.

14. The system of claim 13, wherein the enterprise supervisor is further configured to select a preferred value for the at least one variable input based on application of the at least one optimization criteria to the plurality of animal performance profiles.

15. The system of claim 14, wherein the optimization criteria includes at least one of productivity per unit of nitrogenous waste released, productivity per unit of phosphorous released, balance of bypass nutrients, water exchange rate, and aeration rate.

16. The system of claim 1, wherein the simulator engine further includes an animal economic performance simulator configured to generate an animal performance profile based upon variations in a variable input associated with economic factors for at least one animal.

17. The system of claim 16, wherein the enterprise supervisor engine is configured to actuate the simulator engine based upon variations in the variable input to generate a plurality of animal performance profiles.

18. The system of claim 17, wherein the enterprise supervisor is further configured to select a preferred value for the at least one variable input based on application of the at least one optimization criteria to the plurality of animal performance profiles.

19. The system of claim 18, wherein the optimization criteria includes at least one of value of biomass per cost of feed, value of product per unit of space, value of product per hour of labor, return on investment, return on working capital, and value per kilogram of product produced.

20. The system of claim 1, wherein the simulator engine further includes an animal performance simulator, an animal environment performance simulator, and an animal economic performance simulator configured to generate a comprehensive animal performance profile based upon the animal feed formulation information and the animal information input including at least one variable input.

21. The system of claim 20, wherein the enterprise supervisor engine is configured to actuate the simulator engine based upon variations in the variable input to generate a plurality of animal performance profiles.

22. The system of claim 21, wherein the enterprise supervisor is further configured to select a preferred value for the at least one variable input based on application of the at least one optimization criteria to the plurality of animal performance profiles.

23. The system of claim 22, wherein the optimization criteria includes at least one of weight gain per day, weight gain per week, value of product per cost of feed, value of product per unit of space, survival rate, environmental impact, cycles per year, animal size, stocking density, water exchange rate, product form, productivity per unit of ammonia released, productivity per unit of phosphorous released, water exchange rate, aeration rate, biomass produced per cost of feed, value of biomass per cost of feed, value of product per unit of space, value of product per hour of labor, return on investment, return on working capital, and value per kilogram of product produced.

24. The system of claim 20, wherein the animal performance profile is generated based on at least two of animal performance, animal environment, animal economics.

25. A method for determining optimized values for inputs to an animal production system, comprising:

receiving a plurality of animal information inputs, wherein at least one of the animal information inputs is designated as a variable input;

generating at least one performance projection based on the animal information inputs; and

generating an optimized value for the at least one variable input based on the at least one performance projection and an animal feed formulation and at least one optimization criteria.

26. The method of claim 25, further including generating at least one animal feed formulation composed of the animal feed ingredients based on the at least one performance projection.

27. The method of claim 26, further including optimizing the at least one animal feed formulation according to at least one optimization criteria.

28. The method of claim 25, wherein generating an optimized value for the at least one variable input includes providing an effect of the modification to the at least one variable input.

29. The method of claim 25, wherein the variable input is one of an animal factor, an environmental factor, and an economic factor.

30. The method of claim 29, wherein the variable input is an environmental factor and includes at least one of a thermal environment, an animal population environment, a photoperiod environment, a feeding environment, an animal comfort environment, an animal structural environment, and an animal microbial environment.

31. The method of claim 29, wherein the variable input is an animal factor and includes at least one of an animal species type, an animal job, an animal genotype, and an animal composition.

32. The method of claim 29, wherein the variable input is an economic factor and includes at least one of a market timing, market pricing, and a market location.

33. The method of claim 26, further including generating a plurality of animal performance profiles based upon the animal feed formulation information and the animal information input including at least one variable input.

34. The method of claim 33, further including generating a plurality of animal performance profiles based on variations in the at least one variable input.

35. The method of claim 34, further including selecting a preferred value for the at least one variable input based on application of the at least one optimization criteria to the plurality of animal performance profiles.

36. The method of claim 35, wherein the optimization criteria includes at least one of weight gain per day, weight per week, value of product per cost of feed, value of product per unit of space, survival rate, environmental impact, cycles per year, animal size, stocking density, water exchange rate, and product form.

37. The method of claim 25, further including generating an animal performance profile based upon variation of a variable input associated with the environment for at least one animal.

38. The method of claim 37, further including iteratively generating a plurality of animal performance profiles based on variation of the at least one variable input.

39. The method of claim 38, further including selecting a preferred value for the at least one variable input based on application of the at least one optimization criteria to the plurality of animal performance profiles.

40. The method of claim 39, wherein the optimization criteria includes at least one of productivity per unit of ammonia released, productivity per unit of phosphorous released, water exchange rate, and aeration rate.

41. The method of claim 25, further including generating an animal performance profile based upon variations in a variable input associated with economic factors for at least one animal.

42. The method of claim 41, further including generating a plurality of animal performance profiles based on variation of the at least one variable input.

43. The method of claim 42, further including generating an optimized value for the at least one variable input based on application of the at least one optimization criteria to the plurality of animal performance profiles.

44. The method of claim 43, wherein the optimization criteria includes at least one of value of biomass per cost of feed, value of product per unit of space, value of product per hour of labor, return on investment, return on working capital, and value per kilogram of product produced.

45. The method of claim 25, further including generating a comprehensive animal performance profile based upon the animal information input including at least one variable input and at least two of an animal performance profile, an animal environment performance profile, and an animal economic performance profile.

46. The method of claim 45, further including generating a plurality of animal performance profiles based upon variation of the at least one variable input.

47. The method of claim 46, further including generating an optimized value for the at least one variable input based on application of the at least one optimization criteria to the plurality of animal performance profiles.

48. The method of claim 47, wherein the optimization criteria includes at least one of weight gain per day, weight per week, value of product per cost of feed, value of product per unit of space, survival rate, environmental impact, cycles per year, animal size, stocking density, water exchange rate, product form, productivity per unit of ammonia released, productivity per unit of phosphorous released, water exchange rate, aeration rate, value of biomass per cost of feed, value of product per unit of space, value of product per hour of labor, return on investment, return on working capital, and value per kilogram of product produced.

49. A system for generating an animal feed formulation, comprising:

a simulator engine configured to receive a plurality of animal information inputs and generate animal requirements based on the animal information inputs;

a formulator engine, the formulator engine configured to receive a plurality of animal feed ingredient inputs and generate at least one animal feed formulation composed of the animal feed ingredients based on the animal requirements, wherein at least one of the animal feed ingredient inputs is designated as a variable input; and

an enterprise supervisor engine configured to optimize the at least one animal feed formulation according to at least one optimization criteria, and further configured to generate an optimized value for the at least one variable input based on the at least one optimization criteria.

50. The system of claim 49, wherein suggesting a modification to the at least one variable input includes providing an effect of the modification to the at least one variable input.

51. The system of claim 49, wherein the variable input is the nutrient content of an ingredient.

52. The system of claim 49, wherein the simulator engine further includes an animal performance simulator configured to generate an animal performance profile based upon the animal feed formulation information and the animal information input including at least one variable input.

53. The system of claim 52, wherein the enterprise supervisor engine is configured to actuate the simulator engine based upon variations in the variable input to generate a plurality of animal performance profiles.

54. The system of claim 53, wherein the enterprise supervisor is further configured to select a preferred value for the at least one variable input based on application of the at least one optimization criteria to the plurality of animal performance profiles.

55. The system of claim 54, wherein the optimization criteria includes at least one of weight gain per day, weight per week, value of product per cost of feed, value of product per unit of space, survival rate, environmental impact, cycles per year, animal size, stocking density, water exchange rate, and product form.

56. An animal production optimization system, comprising:

an optimization engine, having an objective function program therein, configured to receive a feed formulation input; and

an animal production modeling system configured to

receive animal information input, including at least one variable input,

receive feed formulation input, and

provide modeling output to the optimization engine,

wherein the optimization engine optimizes the objective function to provide an optimized solution for the at least one variable input based on the modeling output.

57. The animal production optimization system of claim 56, further including a user interface configured to allow the selection, by a user, of one or more variable inputs.

58. The animal production optimization system of claim 56, further including a formulator engine configured to generate to the feed formulation input.

59. The animal production optimization system of claim 56, wherein optimizing the objective function includes iteratively generating modeling output based on variations to the one or more variable input.

60. The animal production optimization system of claim 56, wherein the variable input is one of an animal factor, an environmental factor, and an economic factor.

61. The animal production optimization system of claim 60, wherein the variable input is an environmental factor and includes at least one of a thermal environment, an animal population environment, a photoperiod environment, a feeding environment, an animal comfort environment, an animal structural environment, and an animal microbial environment.

62. The animal production optimization system of claim 60, wherein the variable input is an animal factor and includes at least one of an animal species type, an animal job, an animal genotype, and an animal composition.

63. The animal production optimization system of claim 60, wherein the variable input is an economic factor and includes at least one of a market timing, animal product valuation, and a market location.

64. A method for generating optimized values for variable inputs to an animal production optimization system, comprising:

receiving animal information input, including at least one variable input;

generating modeling output based on the animal information input;

receiving a feed formulation input to the objective function;

generating an objective function based on the modeling output and the feed formulation input; and

optimizing the objective function to provide an optimized value for the at least one variable input.

65. The method of claim 64, further including receiving a selection, by a user, of one or more variable inputs.

66. The method of claim 64, wherein optimizing the objective function includes iteratively generating modeling output based on variations to the one or more variable input.

67. The method of claim 64, wherein the variable input is one of an animal factor, an environmental factor, and an economic factor.

68. The method of claim 67, wherein the variable input is an environmental factor and includes at least one of a thermal environment, an animal population environment, a photoperiod environment, a feeding environment, an animal comfort environment, an animal structural environment, and an animal microbial environment.

69. The method of claim 67, wherein the variable input is an animal factor and includes at least one of an animal species type, an animal job, an animal genotype, and an animal composition.

70. The method of claim 67, wherein the variable input is an economic factor and includes at least one of a market timing, animal product valuation, and a market location.