Agriculture Reference
In-Depth Information
Despite some variances in actual vs
predicted responses which were expected
given the better management and health
status of the research facility compared to
the average commercial barn, the im-
provements were consistent. For the pro-
ducer, feeding higher amino acid:energy
ratio diets on a different feed budget im-
proved feed efficiency and reduced feed
costs.
Table 11.8.
Summary of the performance
predictions between the original programme and the
proposed Optimum programme based on optimizing
for margin over feed costs (MOFC). (From At a
Glance (an internal publication of Nutreco that is
distributed to their customers), February 2010.)
Original
Optimum
ADG, g/day
892
890
ADFI, kg/day
2.23
2.25
Feed:gain
2.50
2.51
Carcass weight, kg
95.4
95.4
Optimum nutrition strategy
MOFC improvement, $/pig
+2.38
Where there are strong business relation-
ships with large customers, models are of
particular benefit because predicting small
cost savings or increased revenue can
translate into significant monetary gains.
However, customers require time to build
up a level of trust in the accuracy of model
predictions. Should the model consist-
ently prove itself to be reliable and accur-
ate, then larger customers will integrate the
model into their decision making process,
frequently requesting simulations to be
made and the outcomes analysed before
implementing any nutrition or production
changes. In 2010, a large farrow to finish
operation (>
15,000
sows) in Canada that has
utilized the services of Watson
®
for the past
5
years, wanted to 'maximize the margin
over feed cost (MOFC) without sacrificing
biological performance'. The key to address-
ing this, and any, issue with an established
customer is being able to utilize a proven de-
scription of the current production environ-
ment already in the model. Running a series
of optimizations for an established customer
is a quick and efficient process. The results
of the optimization analysis yielded two
significant recommendations that would
achieve the desired objective: (i) change the
nutrient density (amino acids and energy) of
the diets; and (ii) add another phase in the
grow-finish programme. The end result was
an improvement in MOFC of $2.38/pig with-
out sacrificing performance
(Table 11.8
).
ADG = average daily gain; ADFI = average daily feed intake.
diets because of increasing feed ingredient
prices using Watson
®
and then validate
within a controlled research facility. The op-
timization objective was to reduce MOFC by
at least $0.50/pig through changes in the en-
ergy and/or amino acid levels of the diets.
The recommendation from the optimization
process was to change the amino acid levels.
A validation trial with Current, Optimum
and Average feeding programme treatments
was conducted using eight replications per
treatment (Nutreco Canada, 2010, unpub-
lished data). All simulations used to design
these programmes were based on the cur-
rent genetics, health status, physical envir-
onment and costs specific to the farm and
time of year. The predicted and actual trial
results are shown in
Table 11.9.
The differences between the actual and
predicted performance results were <4%.
The close alignment between the actual (1.3%)
and predicted (1.1%) improvement in the
producer's MOFC confirms the validity of
using the optimization process in commer-
cial practice.
Conclusions
For successful commercial application
of growth models it is not only imperative
to have an accurate biological model,
but also a well-defined commercialization
process that includes: (i) involving all
stakeholders (e.g. technical advisors, sales
managers, business leaders and the vari-
ous beneficiaries of the technology) in
Optimum nutrient density
A large integrator wanted to re-evaluate the
nutrient specifications of their grower-inisher
