Precision feeding in swine,
for a sustainable animal production
Jaap van Milgen
jaap.vanmilgen@rennes.inra.fr
DuPont/Danisco Animal Nutrition - Technical seminar
November 14, 2012
• Introduction
• Predicting the response of pigs to the nutrient supply
• Dealing with variation
• Monitoring the system
• Examples of precision feeding and precision pork production
• Conclusions
Outline
Animal production is facing new challenges
“The livestock sector is one of the most significant contributors to the most serious
environmental problems, at every scale from local to global”
“As it stands now, there are no technically or economically viable alternatives to
intensive livestock production for providing the bulk of the food supply”
products
resources
(intensive) livestock production systems
high flow rate
very efficient (/ha, /$, or /capita)
in regions with high production densities
1. Continuous monitoring of the process response or outcome
2. Mathematical model predicting the process outcome from inputs
3. The desired outcome
4. A mechanism to control inputs
Precision livestock farming
“Management of livestock production systems using the principles and technology
of process engineering”
(Wathes et al., 2008)
How do I feed a pig so that it will attain 110 kg
at 6 months of age?
• Introduction
• Predicting the response of pigs to the nutrient supply
• Dealing with variation
• Monitoring the system
• Examples of precision feeding and precision pork production
• Conclusions
Outline
age, d
b o d y w e ig h t, k g
Strathe et al., 2010
time
Empirical modeling of growth
Schulin-Zeuthen et al., 2008
feed
Empirical modeling of growth
intake deposition 0.0
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
protein lipid carbohydrate fiber ash
ra te , kg D M /d
The efficiency of nutrient transformation is low
lipid
protein starch sugars fiber
intermediary metabolism
lipid ATP
protein
heat
The transformation of organic matter into a pig
0 5 10 15 20 25 30 35 40 45 50 0
20 40 60 80 100 120
Energy intake (MJ/d)
P ro te in d ep o si ti o n ( g /d )
Whittemore & Fawcett, 1976
upper limit to protein deposition
(PDmax)
minimum LD:PD ratio
Protein deposition depends on energy intake
0 5 10 15 20 25 30 35 40 45 50 0
20 40 60 80 100 120 140
160 25 kg 60 kg 100 kg
Energy intake (MJ/d)
P ro te in d ep o si ti o n ( g /d )
Black et al., 1986
The response of the pig changes over time
The response of the pig changes over time
van Milgen et al., 2006 (InraPorc)
• Growth is determined by protein and lipid deposition
• There is an upper limit to protein deposition
• Energy partitioning rule between protein and lipid deposition
• Feed intake is a model input; lipid is an energy sink
Key concepts in these models
protein deposition
water deposition ash
deposition
lipid deposition
body weight gain lean gain
fat gain
backfat thickness
nutrient intake
The conceptual link between nutrients and tissues
has not been very strong
subcutaneous fatty acids intramuscular
fatty acids
intermuscular fatty acids
internal fatty acids
Lizardo et al., 2002; Halas et al., 2004
dietary fatty acids de novo
synthesized fatty acids
What determines the body fatty acid composition?
dietary C18:2
subcutaneous C18:2 intramuscular
C18:2
intermuscular C18:2
internal C18:2 ATP
What determines the body fatty acid composition?
Kloareg et al., 2007; Bruininx et al., 2011
Control Met- Carcass Body weight gain, g/d 292 211 -28%
Protein content, % 18.6 15.5 -17%
Methionine content, % 2.13 1.97 -8%
Methionine gain, mg/d 1153 645 -44%
The animal possesses different mechanisms to cope with an amino acid deficiency
LD Intestines
-47% 0%
-20% +8%
-12% +3%
-63% +11%
Conde-Aguilera et al., 2010
• Introduction
• Predicting the response of pigs to the nutrient supply
• Dealing with variation
• Monitoring the system
• Examples of precision feeding and precision pork production
• Conclusions
Outline
Variation among individuals is natural, essential,
and very well controlled
energy intake =
1.002 x energy expenditure
energy intake =
1.000 x energy expenditure
Variation among individuals is natural, essential,
and very well controlled
Used in genetic selection
Many production practices are applied to groups of pigs
Variation among individuals is natural, essential,
and very well controlled
1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 3.3 3.5 0.6
0.7 0.8 0.9 1.0 1.1 1.2 1.3
Average feed intake (kg/d)
A ve ra g e d ai ly g ai n ( kg /d )
Different pigs have different nutrient requirements
70 80 90 100 110 120 130 140 0.0
0.2 0.4 0.6 0.8 1.0 1.2 1.4
Age, d
L ys r eq u ir em en t, %
Different pigs have different nutrient requirements
• Introduction
• Predicting the response of pigs to the nutrient supply
• Dealing with variation
• Monitoring the system
• Examples of precision feeding and precision pork production
• Conclusions
Outline
outputs
inputs
Monitoring is crucial - no output without input
65 75 85 95 105 115 125 135 1.0
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Age, d
F ee d i n ta ke , kg /d
Daily feed intake is variable
65 75 85 95 105 115 125 135 0
20 40 60 80 100 120 140 160 180 200
Age, d
C u m u la ti ve f ee d i n ta ke , kg
Feed intake can be “predicted” (afterwards)
65 75 85 95 105 115 125 135 0
20 40 60 80 100 120 140 160 180 200
Age, d
C u m u la ti ve f ee d i n ta ke , kg
Feed intake is more difficult to foresee
Monitoring the system
• Introduction
• Predicting the response of pigs to the nutrient supply
• Dealing with variation
• Monitoring the system
• Examples of precision feeding and precision pork production
• Conclusions
Outline
Outputs:
• feed intake
• growth
• body composition Inputs:
• type of feed
• quantity of feed
• heating
Sensors Model
Controller
Target
Precision pork production
Wathes et al., 2008
Models have become mature and accessible …
Source: InraPorc, 2012
… and are used in in silico and practical solutions
0.60 0.80 1.00 1.20 1.40
1.60 feed carcass
€/ kg c ar ca ss
Morel et al., 2010
… and are used in in silico and practical solutions
Pomar et al., 2009, 2010, 2011 Hauschild et al., 2010, 2012
“Feeding pigs with daily tailored diets reduced N and P intake by 25 and 29% and nutrient excretion by more than 38%.
Feed cost was 10.5% lower for pigs fed daily tailored diets.”
• Introduction
• Predicting the response of pigs to the nutrient supply
• Dealing with variation
• Monitoring the system
• Examples of precision feeding and precision pork production
• Conclusions
Outline
Where do we come from and where do we go?
1850 1980
stochastic modeling
empirical growth models
deterministic growth models
animal monitoring (phenotyping)
models tools
precision pork production
