ArticlePDF Available

Livestock Farming: A suite of electronic systems to ensure the application of best practice management on livestock farms

Abstract and Figures

The sophisticated global market place for livestock products demands safe, uniform, cheap, and environmentally-and welfare-friendly products. However, best-practice management procedures are not always implemented on livestock farms to ensure that these market requirements are consistently satisfi ed. Therefore, improvements are needed in the way livestock farms are managed. Information-based and electronically-controlled livestock production systems are needed to ensure that the best of available knowledge can be readily implemented on farms. New technologies introduced on farms as part of Precision Livestock Farming (PLF) systems will have the capacity to activate livestock management methods that are more responsive to market signals. PLF technologies encompass methods for measuring electronically the critical components of the system that indicate effi ciency of resource use, software technologies aimed at interpreting the information captured, and controlling processes to ensure optimum effi ciency of resource use and animal productivity. These envisaged real-time monitoring and control systems should dramatically improve production effi ciency of livestock enterprises. However, as some of the components of PLF systems are not yet suffi ciently developed to be readily implemented, further research and development is required. In addition, an overall strategy for the adoption and commercial exploitation of PLF systems needs to be developed in collaboration with private companies. This article outlines the potential role PLF can play in ensuring that existing and new knowledge is implemented effectively on farms to improve returns to livestock producers, quality of products, welfare of animals and sustainability of the farm environment.
Content may be subject to copyright.
1
© Institution of Engineers Australia, 2009 Australian Journal of Multi-disciplinary Engineering, Vol 7 No 1
* Reviewed and revised version of a paper
originally presented at the 2007 Society for
Engineering in Agriculture (SEAg) National
Conference, Adelaide, 23-26 September 2007.
Corresponding author Dr Thomas Banhazi can be
contacted at banhazi.thomas@saugov.sa.gov.au.
Precision livestock farming: A suite of
electronic systems to ensure the application of
best practice management on livestock farms *
TM Banhazi
Livestock Systems Alliance, South Australian Research and Development Institute,
Roseworthy Campus, Adelaide University, Roseworthy, South Australia
JL Black
John L. Black Consulting, Warrimoo, NSW
SUMMARY: The sophisticated global market place for livestock products demands safe, uniform,
cheap, and environmentally- and welfare-friendly products. However, best-practice management
procedures are not always implemented on livestock farms to ensure that these market requirements
are consistently satis ed. Therefore, improvements are needed in the way livestock farms are
managed. Information-based and electronically-controlled livestock production systems are needed to
ensure that the best of available knowledge can be readily implemented on farms. New technologies
introduced on farms as part of Precision Livestock Farming (PLF) systems will have the capacity to
activate livestock management methods that are more responsive to market signals. PLF technologies
encompass methods for measuring electronically the critical components of the system that indicate
ef ciency of resource use, software technologies aimed at interpreting the information captured,
and controlling processes to ensure optimum ef ciency of resource use and animal productivity.
These envisaged real-time monitoring and control systems should dramatically improve production
ef ciency of livestock enterprises. However, as some of the components of PLF systems are not yet
suf ciently developed to be readily implemented, further research and development is required. In
addition, an overall strategy for the adoption and commercial exploitation of PLF systems needs
to be developed in collaboration with private companies. This article outlines the potential role
PLF can play in ensuring that existing and new knowledge is implemented effectively on farms to
improve returns to livestock producers, quality of products, welfare of animals and sustainability
of the farm environment.
1 INTRODUCTION – THE IMPORTANCE
OF APPLYING EXISTING KNOWLEDGE
In most developed countries, a signi cant amount of
money is spent on agricultural research, development
and related extension efforts (RD&E) to generate and
communicate the research  ndings to the farming
community (Mullen, 2002; Alston et al, 2000). For
example, during 2005-06 approximately A$540
million was invested by the Rural Research and
Development Corporations on agricultural research
in Australia. Despite this considerable investment
into agricultural research, a comprehensive study by
Mullen (2002) suggested that in real terms the gross
value of agricultural production across Australia
has remained largely unchanged over the period
from 1953 to 2000 ( gure 1). However, Mullen (2002)
concluded that without productivity improvements
resulting from investment into RD&E, the real gross
value of agriculture production in Australia would
have fallen by approximately 65% over the period.
Two striking examples from Australian animal
industries show how the rate of productivity can
remain virtually unchanged over decades despite
large investments in RD&E. The first example
reveals that average reproductive performance of
pigs has hovered around 21 pigs weaned per sow
per year for the last two decades despite continuing
research funding and a realistic potential being 28-30
N08-AE06 Banhazi.indd 1N08-AE06 Banhazi.indd 1 3/07/09 10:08 AM3/07/09 10:08 AM
2
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
pigs weaned per sow per year (Black et al, 2002). A
study of 31 farms over at least 3 years showed that
the best farm produced 24.7 pigs and the poorest
produced only 16.1 weaned per sow per year. The
second example shows that on average grazing
beef enterprises in southern Australia utilise only
around 30-35% of the pasture grown, despite several
individual farms utilising over 85% of the pasture
grown (Black & Scott, 2002).
These two examples, plus many others, con rm that
a considerable portion of the funds spent by RD&E
providers in agriculture have resulted in limited return
to their respective industries. An enormous amount of
information has been generated around the world over
preceding decades, but many of the funds invested
by the research organisations have been on projects
that “reinvent the wheel” being largely repetition of
older research with, at best, marginal improvements
in industry productivity or pro tability.
The main reasons for this lack of progress is the
inef cient adoption by the farming community of
current knowledge, which means that the same
industry “problems” keep reappearing. The funding
organisations frequently react by continuing to invest
in these “problem” areas despite the application of
existing knowledge being the major issue, rather than
a lack of knowledge.
One main issue that needs resolving to bring about
more effective returns from RD&E expenditure
is a process that ensures effective adoption of
current knowledge. The greatest advances in farm
or enterprise productivity and pro tability in the
future will come from the application of a rigorous
procedure for ensuring that the most essential
processes are carried out correctly and consistently
using well known risk control methodology (Snijders
& van Knapen, 2002). There is currently an abundance
of information available to farm managers, but it is
generally not structured in a way that can be applied
readily. The information often appears piecemeal
in many places, including rural papers, radio,
television, commercial outlets, RD&E organisations,
universities, the internet and others. The information
is frequently “dumbed-down” or sensationalised and
is not structured in a way that promotes adoption.
Many managers suffer from “information-overload”
and thus cannot readily identify the practices that are
the most important to adopt or how to apply them
correctly. Frequently managers adopt procedures in
areas of most interest or in which they have the most
expertise, and neglect other important processes
that drive overall productivity and pro tability. A
system is needed to ensure that the most important
processes are identi ed and that all these processes
are carried out correctly and consistently, with none
being neglected.
Well-defined quality control and continuous
improvement methods based on (i) Total Quality
Management (TQM) systems (Landesberg, 1999) and
(ii) Hazard Analysis Critical Control Point (HACCP)
principles are used widely in the manufacturing
industries throughout the world (von-Borell et
al, 2001; Noordhuizen & Frankena, 1999). These
processes are engaged to ensure all products from
an industrial plant meet speci cations with little
tolerance for error. The HACCP principles were
developed originally by the United States Army and
the National Aeronautics and Space Administration
(NASA) to guarantee that food poisoning would not
occur in astronauts during early space  ights in the
1960s. Similar processes are now used widely for food
safety across the food processing and agricultural
industries (Petersen et al, 2002; Valdimarsson et al,
2004; Snijders & van Knapen, 2002), and have also
been applied in management of cropping systems
0
5000
10000
15000
20000
25000
30000
35000
1953
1956
1959
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
Year
GVP ( AU$ m)
Real GVP f rom product ivityReal GVP w ithout productivity
Figure 1: Gross value of Australian agricultural production (GVP) in the year 2000 in real dollar value
terms, showing the proportion due to productivity improvement (Mullen, 2002).
N08-AE06 Banhazi.indd 2N08-AE06 Banhazi.indd 2 3/07/09 10:08 AM3/07/09 10:08 AM
3
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
(Aubry et al, 2005). Indeed, the HACCP principles
can be applied to any sector of agriculture to control
risk and ensure high levels of productivity and
product quality at all stages along a production
chain (Snijders & van Knapen, 2002). The HACCP
principles provide the ideal structure for ensuring
that the most important processes determining
productivity and pro tability in an animal enterprise
are adopted and performed with least chance of
failure. The HACCP principles are now being used
by several sectors of the Australian animal industries
to aid adoption of existing knowledge.
The essential steps that need to be incorporated within
a well-designed and controlled production process
are (Beattie, 2001; Cumby & Phillips, 2001; Webster,
2001): (i) integration of automated data measurement
and acquisition systems into the production chain
(Frost, 2001; Banhazi et al, 2007b); (ii) establishment
of protocols for data-integration and automated
data analysis to identify inef ciencies in processes
and to facilitate decision making (Scho eld et al,
1994; 2002); (iii) transfer of the results from data
analysis as inputs into automated decision making
processes and trigger certain management actions
(Banhazi et al, 2002; 2003; Black, 2002); (iv) activate
control systems, which could be either automated
or appropriately documented in standard operating
procedures (SOPs) (Gates & Banhazi, 2002; Gates et
al, 2001); and (v) include procedures to monitor the
outcome of control actions and documentation for
quality assurance (QA) purposes (Black, 2001).
The major bene ts from adopting a Precision Livestock
Farming (PLF) system as outlined are to ensure that
every process within a livestock enterprise, which
can have a large positive or large negative effect on
productivity and pro tability, is always controlled
and optimised within narrow limits. The system
enables near optimum use of all resources, even in
a highly variable environment, with animal growth
rates meeting predetermined speci cations through
optimal use of feed and water, and control of health
status. The system ensures consistent, high quality
products and the potential for real-time supply chain
management throughout an industry. An important
component of the system is for measurements,
interpretation of measurements and control of
processes to be conducted through electronically-
controlled technology, with limited need for human
intervention and the frequently associated inherent
human error. When fully implemented, the system
should ensure near-maximum pro tability through
continual optimisation of resource use in a highly
variable environment, as is normally encountered
in Australian agriculture.
The purpose of this paper is to describe the concepts
behind developing PLF systems for livestock
enterprises, and how they can be implemented
to ensure optimum use of variable resources and
maximisation of pro tability. Several dif culties in
implementing the systems on farms and the need for
development of new technologies are highlighted.
2 ADOPTION OF A RIGOROUS
PROCEDURE FOR THE APPLICATION
OF EXISTING KNOWLEDGE
2.1 Identi cation of important
processes on farms
The first step in developing the proposed PLF
system is to identify those processes, which if not
carried out correctly, will have a major impact on
either productivity or pro tability of an enterprise.
A large negative impact has been termed a hazard
(Landesberg, 1999). A critical control point (CCP) is
the last point in a process where the hazard can be
averted and product quality, level of productivity,
profitability and sustainability of the enterprise
can be maintained near to optimum/maximum.
The primary reason for identifying the CCP for
each process is that it helps determine the variables
that need to be measured and the frequency of
measurement required to avoid the hazard.
The most important aspect in developing the CCP
approach is to  rst identify and then ensure that only
those few processes or tasks that will have a major
impact on productivity, pro tability or sustainability
of an enterprise are carried out. These are described
as the “must do” or “big hit” processes, where the
consequences of them not being carried out correctly
could have a substantial impact on the success of an
enterprise (Black, 2002; Black et al, 2001). The aim is
to reduce to a minimum the number of tasks on which
a manager needs to concentrate, but ensure that these
tasks are carried out correctly. These essential tasks
are usually identi ed by considering the relative
magnitude of the “hazard” or loss in pro t that
would result if they were not applied correctly. For
example, only 24 processes were considered essential
in a grazing beef enterprise from strategic planning
to sale of product (Black & Scott, 2002).
Identifying the essential task for speci c enterprises
can be a major challenge, and needs a great deal of
lateral thinking and analysis. Methods that can be
used to identify areas where a change could result in a
marked increase in productivity and/or pro tability
are described.
One procedure is a formal analysis of historical
data within an industry sector or across speci c
enterprises to identify major determinants of
productivity and profitability. For example, a
statistical evaluation of the reasons for low and
variable reproduction performance of sows across
31 farms in Australia showed that 45% was due to
litter size and approximately 15% due to extended
weaning to mating intervals (Black et al, 2002).
Further assessment of the information resulted
in the conclusion that developing highly-ef cient
N08-AE06 Banhazi.indd 3N08-AE06 Banhazi.indd 3 3/07/09 10:08 AM3/07/09 10:08 AM
4
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
technologies for identifying oestrus and mating time
or reducing early embryonic loss should result in
major improvements in reproductive performance of
Australian pigs. Another example of this approach
could be the recently developed statistical models
that identi ed signi cant risk factors for sub-optimal
air quality in Australian livestock buildings (Banhazi
et al, 2008b; 2008c; 2008d; 2008e; 2008f). By managing
these risk factors, improvements in air quality can
be achieved.
Another approach is to undertake a workshop or
“think-tank” process, where experts across a whole
industry value chain identify potential areas where
change would substantially affect productivity
(Banhazi et al, 2007b; Banhazi, 2006). Some are
relatively clear, such as removing humans from the
dairy cow milking process (Bull et al, 1996; Ordolff,
2001). Others, such as several-fold increase in winter
fodder production, which commonly limits overall
grazing enterprise carrying capacity in southern
Australia, although recognised as important, are
often accepted as inevitable. However, low winter
fodder production should not be assumed to be
unchangeable.
Computer simulation models or spreadsheets can
also be used to quantify the likely effects of change on
productivity and/or pro tability. Table 1 shows the
predicted effect, using the AUSPIG simulation model,
of a range of scenarios suggested from a workshop
on the pro tability of a 500 sow piggery in Australia
(Black, 2006). The quantitative evaluation of likely
bene ts had a signi cant bearing on identifying
processes that must be carried out correctly and for
setting future research priorities.
One of the greatest opportunities for quantum
changes in productivity and profitability within
agricultural enterprises is to reduce the need for
high-cost and inef cient labour and/or increasing
its reliability.
2.2 Data collection systems
Once the main processes are identi ed that will have
a major impact on farm pro tability, the next step
is to identify for each process the farm or market
variable that must be measured, and the maximum
and minimum limits to the variable that will ensure
the process is always carried out correctly and the
potential hazard avoided. The desired frequency of
the measurements must also be set. Electronic capture
of the information is essential to improve reliability
of the information collected. Table 2 details the
areas of possible electronic data collection on farms
(Durack, 2002).
In the past, information collection on such wide
ranging parameters (especially in electronic format)
has been dif cult, costly and was often neglected
(Black et al, 2001). However, advancement in sensor,
computer technology and modern analytical tools
have made the collection and analysis of large
amounts of data highly feasible (Black et al, 2001).
A well-designed data collection system would
reduce the need for manual recording of farm
production data, and would present the producer
with management data in the most ef cient form
(Schon & Meiering, 1987). By using automated data
collection systems on farm, it would be possible to
transfer details of animal performance, labour input,
environmental performance and other essential farm
information to a central data warehouse, where
it could be stored, processed and sent to control
systems within the farm. Examples of relevant
information that can be captured electronically are
discussed below.
The weight of animals is one of the most important
measurements to obtain. Up-to-date monitoring of
the growth rate of pigs would provide producers
with valuable information on compliance with
projected growth pathways, health, and likely carcass
composition and yield (Korthals, 2001; Scho eld
Table 1: Predicted effect using the AUSPIG simulation model to examine a range of scenarios on the
pro tability of a 500 sow reference piggery (Black, 2006).
AUSPIG simulation Pro t ($/kg carcass)
Reference piggery – base simulation 0.13
Increase price paid for pig meat by $0.50/kg carcass 0.63
Decrease average cost of feed by $50/t 0.32
Increase average growth rate for male and female pigs by 50 g/day 0.19
Apply PSTa for 4 weeks at a cost of $5.00/pig 0.21
Decrease herd health and increase mortality from 3.5% to 7% –0.18
Change feed waste to 2% 0.22
Change feed waste to 22% 0.04
Halve labour costs 0.28
Increase pigs sold/sow/year to 23.9 0.34
Increase pigs sold/sow/year to 28.0 0.58
a Porcine somatotrophin
N08-AE06 Banhazi.indd 4N08-AE06 Banhazi.indd 4 3/07/09 10:08 AM3/07/09 10:08 AM
5
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
et al, 1999; Lokhorst & Lamaker, 1996). Acquiring
suf cient data points at acceptable time intervals
and with suf cient accuracy is impractical using
conventional scales because of the increase in labour
requirements and stress on both the farmer and
stock. An automated weighing system is essential.
Image analysis technologies are being developed
to acquire weight of animals automatically as they
visit a feeder, and also for identifying fatness and
carcass yield (Banhazi et al, 2007c; Kollis et al,
2007). The information on weights of animals can
be incorporated into computer simulation models,
such as AUSPIG, to calculate the daily ration needed
to meet speci ed growth trajectories based on the
genotype of animal and the current environmental
conditions. In addition, image analysis can be used
to assess behaviour and welfare of pigs (Xin & Shao,
2002; Hemsworth et al, 1995; Shao & Xin, 2008).
Detecting irregularities in feed (Sliva et al, 2007;
Banhazi et al, 2009), water intake, body temperature
and body weight could be used as an early warning
system for detecting diseases (Pedersen & Madsen,
2001; Mottram, 1997). Disease monitoring, as well
as the traceability of the  nal product, could be
dramatically enhanced by the wide spread adoption
of electronic animal identi cation tools and related
monitoring equipment (Jansen & Eradus, 1999;
Eradus & Rossing, 1994; Street, 1979; Schon &
Meiering, 1987). Some innovative approaches have
already been taken to automatically detect coughing
episodes in piggery buildings, and use the frequency
and intensity of these coughing episodes to detect
respiratory diseases (Chedad et al, 2001; Moshou et
al, 2001a; 2001b).
Airborne particles and gases are associated with sub-
optimal air quality in sheds predisposing both livestock
and staff to potential health problems (Banhazi et al,
2008d). Pollutant emissions from intensive livestock
production facilities are also associated with odour
transfer (Hartung, 1986; Bottcher, 2001). Therefore,
accurate, low-cost monitoring of these pollutants is
essential if appropriate reduction strategies are to be
implemented on farms (Banhazi, 2005; 2009).
2.3 Data analysis systems
Computer models should in the future become an
integral part of PLF management systems (Banhazi
et al, 2007b). The models would be capable in real
time of identifying individual animals that are of the
correct weight and body speci cations to maximise
payment from a wide range of alternative buyers.
Growth models would be capable of assessing
whether individual animals or groups of animals
Table 2: A summary of the range of production and environmental variables which could potentially
be measured, recorded and analysed (Durack, 2002).
Possible rate of collection
Animal parameters
Daily weight gain (DWG) Daily
Feed conversion ratio (FCR) Daily
Feed consumption (FC) Hourly
Body composition – eg. back fat (BC) Daily
Body conformation (BCo) Daily
Stress levels Daily
Antisocial/normal behaviours Daily
Oestrus (heat) detection Hourly
Environmental conditions
Ambient climatic conditions Hourly
Humidity and internal air and  oor temperature Hourly
Air speed Hourly
Floor and animal wetness Daily
Gas levels, CO2, NH3Hourly
Dust levels Hourly
Air born pathogen levels Daily
Transport and supply chain management
Electronic individual animal ID and trace back N/A
Transport environmental conditions log Hourly
Full individual animal carcass performance N/A
Input material ID, such as feed, medication, etc. N/A
N08-AE06 Banhazi.indd 5N08-AE06 Banhazi.indd 5 3/07/09 10:08 AM3/07/09 10:08 AM
6
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
have an excess or de ciency in amino acid supply for
every diet presented, whether ambient temperature
is above or below the zone of thermal comfort, and
whether the animals are under- or over-stocked
(Black, 2002). In addition, the models could assess
whether the livestock are being fed an energy intake
that is coincident with the minimum needed for
maximum protein deposition, where the ef ciency
of feed utilisation is maximised. The factors limiting
feed intake could be predicted and excessive feed
wastage identified. Diets could be reformulated
automatically at speci ed times to most economically
meet the nutrient requirements of each group
of livestock and feed intake modi ed to achieve
greatest economic efficiency of feed utilisation
(Wathes et al, 2001). Building temperatures, air
movement and humidity could be controlled to
ensure that the animals were in the zone of thermal
comfort for suf cient time each day to maximise
pro t in relation to constraints on growth and costs
of environmental control (Gates & Banhazi, 2002).
Similarly, stocking rates and diet composition could
be adjusted through the automatic movement of
animals and modi cations to diet energy density to
ensure maximum pro t.
A variety of approaches can be used for analysing
recorded information from simple graphing and
spreadsheet analysis to sophisticated statistical
analyses (such as data-mining) and computer
simulation modelling (Aerts et al, 2001; Bird et al,
2001; Durack, 2002; Schmoldt, 2001; Stafford, 2000).
Although the simpler techniques can provide useful
insights into production inef ciencies, they are often
limited and do not account for important interactions
between factors that affect livestock performance and
enterprise pro tability (Black, 2001). Computerised
models that account for total enterprise resource
use, such as AUSPIG (Black et al, 2001), could help
identify likely inef ciencies in production methods
and pro t generation.
2.4 Control systems
Each process identi ed to have an important affect
of enterprise pro tability should have an optimal
range. Keeping these processes within the optimal
range will ensure optimisation of farm pro tability.
Consequently, when these critical processes are
outside the limits, automated systems need to
be called into operation or managers/appointed
persons need to be alerted to rectify the situation.
An important outcome of either human or machine
intervention is to ensure that the key processes on
farms remain within the critical limits.
Control systems on livestock farms should be viewed
holistically and incorporate both technology-based
systems, such as automated sorting-gates, and
operator–based systems, such as moving animals.
The establishment of SOPs for individual tasks
is essential. The SOPs ensure that procedures are
carried out correctly and should keep all measured
variables within the maximum and minimum limits.
When measured variables fall outside these limits,
procedures are in place to undertake predetermined
corrective action. The SOPs in combination with
the HACCP process are designed to ensure that the
manager can be absent from the enterprise for an
extended period and be con dent that pro t will
be maximised and long-term sustainability of the
enterprise sustained (Black, 2002). There are both
high and low level SOPs. The high level SOPs de ne
each major task and when it should be undertaken,
whereas the low level SOPs describe how an
individual task is done or how to operate speci c
pieces of equipment or software designed to assist
with the data collection, data analysis or the control.
Development of SOPs for an individual enterprise
is a challenging task, but crucial for complete
uptake of the scheme proposed for improving
adoption of current knowledge. In addition, having
predetermined SOPs whenever the measurement
limits are breached, reduces substantially the stress
level for the manager because the plan of action and
when to apply it has already been established and
the consequences known.
3 IMPORTANCE OF AUTOMATION
For example, the HACCP and SOP based system
outlined is being implemented with enthusiasm
and some success in the Australian beef industry
(Black & Scott, 2002). The proposed system has also
been applied to speci c aspects of other enterprises,
including reproductive success in the pig industry
(Black, 2002). However, continuing implementation
of the system has proved difficult for many
managers. Although the system is logical, it must
have enterprise SOPs that are often time-consuming
to produce. In addition to the dif culty of preparing
precise SOPs, removal of mundane, repetitious
tasks is essential for all agricultural industries (Yen
& Radwin, 2000; Van Henten et al, 2006; Noguchi
et al, 2004; Belforte et al, 2006). Future advances in
compliance will come from automated measurement,
interpretation and control of most processes, which
can be overseen by a manager from the of ce.
The current revolution in agricultural engineering,
electronics and robotics has enormous potential
for agriculture, and is already being applied
in many areas (Ordolff, 1997; Artmann, 1997;
Noguchi et al, 2004; Zhang et al, 2006; Belforte
et al, 2006). Electronic, robotic and automatic
components have a particularly important role in
the measurement of critical variables, interpretation
of these measurements and control of processes.
A recent example from Bishop-Hurley et al (2007)
illustrated how radio technology can be modi ed to
develop virtual fencing for cattle. Narrow, straight
lines of radio signals, potentially controlled by
satellite positioning systems, form the boundaries to
N08-AE06 Banhazi.indd 6N08-AE06 Banhazi.indd 6 3/07/09 10:08 AM3/07/09 10:08 AM
7
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
the area designated for grazing. As an animal  tted
with a collar approaches the signal, a buzzer in the
collar starts to sound and becomes louder as the cow
nears the signal. An electric shock is delivered from
the collar if the line is breached. Trials have shown
that animals learn quickly to retreat once the buzzer
sounds (Bishop-Hurley et al, 2007).
3.1 Provision of integrated package
It is essential for the success of PLF systems to provide
all the tools necessary for making the essential
measurements, interpreting the measurements and
executing the most profitable corrective actions.
These tools are an essential component of the overall
“package” delivered to farm managers and must be
provided as part of the adoption process. The tools
may be physical instruments (for example, a meter
for measuring pasture height or mass), they may be
descriptions on how measurements are made, tables
of information, graphs, spreadsheet models, and
other forms of software or internet links to speci c
programs to retrieve essential rainfall, soil moisture
and market or satellite information. The process of
identifying these tools has proven to provide an
excellent framework for determining critical R&D
projects that need to be completed to ensure the
adoption “package” can be fully implemented. An
essential component of the “package” is that all
the information is readily available to the manager
whenever it is needed. One of the reasons for the
relatively slow adoption rate of PLF technologies
on farms is the fact that individual components of
the system are often independently developed by
different research groups without fully appreciating
the importance of integrating these technologies
into a fully functional package (Banhazi et al, 2007a;
Banhazi, 2006).
4 DEVELOPMENT OF PLF SYSTEMS
SPECIFICALLY WITHIN THE
AUSTRALIAN PIG INDUSTRY
The adoption of PLF technologies within the
Australian livestock industries should occur in
coordination of current hardware and software
components and systems to run on a standardised
data communication protocol. This approach
optimises the value of existing developments, but
also supports the utilisation of whole of industry
coordination (Banhazi et al, 2002). Figure 2 represents
a schematic presentation of the integrated system
proposed above, which has the most potential to
create a sustainable competitive advantage for the
Australian pig industry. In table 3, the likely research
programs needed to develop fully integrated PLF
systems are listed.
Note that the main barrier to this complete solution
is currently data compatibility and transfer. In
Figure 2: Australian PLF Vision (integration
of herd management tools with the
AUSPIG computer model, resulting in
a holistic decision support system).
delivering on the vision expressed in  gure 2, the
industry must recognise the need for investment
in an effectively targeted R&D program to link the
different components together and optimise the
value of existing technologies (table 2). In summary,
the following key developments will be needed:
(i) Establishment of “fully instrumented piggery
buildings” to collect data from these production
sites for analysis.
(ii) Application of in-depth analysis, modelling
and investigation of the available data (using
a variety of methods) with the clear focus of
demonstrating direct financial benefits for
producers through the identification of key
production indicators (KPI).
(iii) Improvement of the “control functions” of
AUSPIG to facilitate outputs from the system
to be linked with automatic control tools.
(iv) Application of economical analysis to
demonstrate the  nancial advantages of using
PLF systems and commercialisation of PLF
systems developed.
5 RESEARCH NEEDS IDENTIFIED
The main objectives of the farm implementation
project are to (i) demonstrate the economic bene ts
of the PLF system on farms, and (ii) identify
the shortcomings of the system and implement
improvements. The establishment of the so-called
“fully instrumented piggery buildings” will be the
first step in the project implementation process.
These facilities will make the real time analysis and
modelling of the available data possible with the
N08-AE06 Banhazi.indd 7N08-AE06 Banhazi.indd 7 3/07/09 10:08 AM3/07/09 10:08 AM
8
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
clear focus of demonstrating direct  nancial bene ts
for producers. The planned study will facilitate
interaction between producers, manufacturers of the
hardware devices needed for making measurements,
and developers of software that capture and process
the electronic outputs into a form that can be analysed
using traditional statistical and other approaches.
Recommendations for changes in management
practices to improve the biological ef ciency and
pro tability of the enterprises will then be made in
consultation between advisers and producers.
Figure 3 shows an overview of data  ow throughout
the system. Measurement data are collected and
accumulated on farms. This temporary storage is
automatically accessed daily. The data warehouse
will access the on-farm data collection system and
request data upload.
The accumulated data will be sent to the web server
where it is compressed and stored. The data are
erased from the on-farm loggers and the system will
be ready to continue gathering data. Any authorised
owner of data can log onto a secure website and
download information. The requested data are
uncompressed and sent to consultants in a standard
format. The data received is then processed and
uploaded into speci c software such as PrimePulse,
AUSPIG and statistical analysis packages. Once
the data have been analysed, the results can be
interpreted by the project team and communicated
to producers.
To transform the information management systems
currently used on farms to a more sophisticated
system, data collection and more importantly data
analysis need to be automated (Schon & Meiering,
1987). Automated data collection, management and
analysis, together with accessible data-warehouses,
would transform the currently used segregated
systems into a powerful information based PLF
system (Enting et al, 1999). Centralised data
collection sites (data warehouses) will be an essential
component of a well-functioning PLF system. Such
centralised data management will enable uniform
data analysis and appropriate interpretation of
available data. On-farm data processing could
provide valuable support for farm managers in
everyday management, but the real gains would
come from using in-depth data analysis provided
by remote data warehouses via the internet (Geers,
1994; Petersen et al, 2002).
6 CONCLUSIONS
Funds spent by RD&E providers in agriculture
have resulted frequently in limited return to their
industries in comparison to their potential. However,
large opportunities exist to increase the returns
on investment in RD&E in the animal industries,
and to substantially improve productivity and
profitability in the 21st Century. Investment of
funds into the rigorous application of well-de ned
HACCP principles and individual enterprise SOPs,
along with essential associated tools, should ensure
that relevant existing knowledge is focused on
maximising pro tability and sustainability of animal
Table 3: Potential R&D topics within the major program areas (Durack, 2002).
Development areas
Environmental management
On-farm measurement and documentation tools (Banhazi, 2005; Sliva et al, 2007)
Housing management
Advanced climate control tools (Banhazi et al, 2008a; 2008f)
Animal welfare and behaviour assessment tools (Shao & Xin, 2008)
Production management
Real time individual pig weighing system (Kollis et al, 2007)
Real time feed and water consumption recording (Madsen & Kristensen, 2005; Madsen et al, 2005)
Disease monitoring tools and systems (Maatje et al, 1997; Eradus & Jansen, 1999)
Feed disappearance measurements
Market intelligence systems
Integrated performance analysis of units (Heinonen et al, 2001; Pomar & Pomar, 2005)
Online KPIs monitoring and comparison with modelled performance norms (Tukey, 1997)
AUSPIG integration program
Supply chain management
• Information ow from slaughter houses (Petersen et al, 2002)
Individual animal ID (Naas, 2001; 2002)
Automated record keeping (Holst, 1999)
Real time supply management programs (Dobos et al, 2004; Guerrin, 2004)
N08-AE06 Banhazi.indd 8N08-AE06 Banhazi.indd 8 3/07/09 10:08 AM3/07/09 10:08 AM
9
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
enterprises. Further investment is essential into
technologies that automatically measure, interpret
and control these crucial systems, because many of
the system breakdowns are due to human failings.
The replacement of mundane, repetitive tasks
within many animal enterprises with automated
systems should substantially improve quality
control and job satisfaction, thereby reducing risks
commonly associated with intensi cation. Additional
investment will generally be required to develop the
tools needed to ensure effective application of all
aspects of the HACCP-SOPs systems. Agricultural
engineers will have a major role in providing these
essential hardware and software components for
animal industry PLF systems. In addition, more
critical thought is frequently needed when setting
research priorities. An important challenge is to
identify those changes in an industry sector that
are likely to result in “quantum leap” changes in
productivity and pro tability. The greatest progress
is usually made by developing new technological
tools, rather than  ne-tuning existing methods of
animal production.
ACKNOWLEDGMENTS
The authors wish to acknowledge the professional
assistance of professors D. Berckmans, S. Pedersen,
C. Wathes and R. Gates, and the  nancial support of
Australian Pork Limited.
REFERENCES
Aerts, J. M., Wathes, C. M. & Berckmans, D. 2001,
“Applications of process control techniques in
poultry production”, in Integrated Management Systems
for Livestock, Wathes, C. (editor), Selwyn College,
Cambridge, UK, BSAS, Edinburgh, pp. 147-154.
Alston, J. M., Marra, M. C., Pardey, P. G. & Wyatt, T. J.
2000, “Research returns redux: a meta-analysis of the
returns to agricultural R&D”, The Australian Journal of
Agriculture and Resource Economics, Vol. 44, pp. 185-215.
Artmann, R. 1997, “Sensor systems for milking
robots”, Computers and Electronics in Agriculture, Vol.
17, No. 1, pp. 19-40.
Aubry, C., Galan, M. B. & Maze, A. 2005, “HACCP
methodology and quality/environmental
speci cations for crop farms: Implications for the
design of good agricultural practices guidelines”,
Cashiers Agricultures, Vol. 14, pp. 313-322.
Banhazi, T. 2005, “Improved air quality measurement
procedure – BASE-Q system”, in AAPV Conference,
Gold Coast, QLD, Australia, Fahy, T. (editor), Vol. 1,
pp. 71-75.
Banhazi, T. 2006, “Potential precision livestock
farming technologies for the egg industry”, in
Advances in agricultural technologies and their economic
and ecological impacts, Gan-Mor, S. (editor), ISAE, Tel
Aviv, Israel, Vol. 1, pp. 45-47.
Data
Warehouse
Figure 3: PLF system implementation, adapted from Banhazi et al (2007b).
N08-AE06 Banhazi.indd 9N08-AE06 Banhazi.indd 9 3/07/09 10:08 AM3/07/09 10:08 AM
10
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
Banhazi, T. M. 2009, “User friendly air quality
monitoring system”, Applied Engineering in
Agriculture, Vol. 25, No. 2, pp. 281-290.
Banhazi, T., Black, J. L., Durack, M., Cargill, C.,
King, A. & Hughes, P. 2002, “Precision Livestock
Farming”, in Proceedings of the Australian Association
of Pig Veterinarians Conference, Adelaide, Australia,
Australian Association of Pig Veterinarians, Vol. 1,
pp. 103-110.
Banhazi, T., Black, J. L. & Durack, M. 2003, “Australian
Precision Livestock Farming workshops”, in Joint
Conference of ECPA – ECPLF, Werner, A. & Jarfe, A.
(editor), Berlin, Germany, Wageningen Academic
Publisher, Vol. 1, pp. 675-684.
Banhazi, T., Dunn, M., Black, J. & Ludvigsen, G.
2007a, “Managing growth variability through the
implementation of Precision Livestock Farming
systems”, in The Bi-annual Conference of the Australian
Society of Engineering in Agriculture (SEAg 2007) -
Challenge Today, Technology Tomorrow, Banhazi, T. &
Saunders, C. (editors), Adelaide, South Australia,
Australian Society of Engineering in Agriculture,
Vol. 1, pp. 23-31.
Banhazi, T., Dunn, M., Cook, P., Black, J., Durack,
M. & Johnnson, I. 2007b, “Development of precision
livestock farming (PLF) technologies for the
Australian pig industry”, in 3rd European Precision
Livestock Farming Conference, Cox, S. (editor),
University of Thessaly, Skiathos, Greece, Vol. 1, pp.
219-228.
Banhazi, T., Rutley, D. & Dunn, M. 2007c, “Using
statistical modelling to improve the precision
of image analysis based weight estimation”, in
Manipulating Pig Production X, Paterson, J. (editor),
Brisbane, Australia, Australasian Pig Science
Association, Vol. 1, pp. 48.
Banhazi, T. M., Aarnink, A., Thuy, H., Pedersen, S.,
Hartung, J., Payne, H., Mullan, B. & Berckmans,
D. 2008a, “Review of issues related to heat stress
in intensively housed pigs”, in 8th International
Livestock Environment Symposium (ILES), Stowell, R.
R., Wheeler, E. F. & Yanagi, T. (editor), Iguassu Fall,
Brazil, ASABE, Vol. 1, pp. 737-744.
Banhazi, T. M., Rutley, D. L. & Pitchford, W. S. 2008b,
“Identi cation of risk factors for sub-optimal housing
conditions in Australian piggeries – Part IV: Emission
factors and study recommendations”, Journal of
Agricultural Safety and Health, Vol. 14, No. 1, pp. 53-69.
Banhazi, T. M., Seedorf, J., Laffrique, M. & Rutley,
D. L. 2008c, “Identi cation of risk factors for high
airborne particle concentrations in broiler buildings
using statistical modelling”, Biosystems Engineering,
Vol. 101, No. 1, pp. 100-110.
Banhazi, T. M., Seedorf, J., Rutley, D. L. & Pitchford,
W. S. 2008d, “Identi cation of risk factors for sub-
optimal housing conditions in Australian piggeries
– Part I: Study justi cation and design”, Journal of
Agricultural Safety and Health, Vol. 14, No. 1, pp. 5-20.
Banhazi, T. M., Seedorf, J., Rutley, D. L. & Pitchford,
W. S. 2008e, “Identi cation of risk factors for sub-
optimal housing conditions in Australian piggeries
– Part II: Airborne pollutants”, Journal of Agricultural
Safety and Health, Vol. 14, No. 1, pp. 21-39.
Banhazi, T. M., Seedorf, J., Rutley, D. L. & Pitchford,
W. S. 2008f, “Identi cation of risk factors for sub-
optimal housing conditions in Australian piggeries
– Part III: Environmental parameters”, Journal of
Agricultural Safety and Health, Vol. 14, No. 1, pp. 41-52.
Banhazi, T. M., Rutley, D. L., Parkin, B. J. & Lewis,
B. 2009, “Field evaluation of a prototype sensor
for measuring feed disappearance in livestock
buildings”, Australian Journal of Multi-disciplinary
Engineering, Vol. 7, No. 1, pp. 27-38.
Beattie, V. 2001, “Environmental design for pig
welfare”, in Integrated Management Systems for
Livestock, Wathes, C. M. (editor), Selwyn College,
Cambridge, UK, BSAS, Edinburgh, pp. 97-103.
Belforte, G., Deboli, R., Gay, P., Piccarolo, P. & Ricauda
Aimonino, D. 2006, “Robot Design and Testing for
Greenhouse Applications”, Biosystems Engineering,
Vol. 95, No. 3, pp. 309-321.
Bird, N., Crabtree, H. G. & Scho eld, C. P. 2001,
“Engineering Technologies Enable Real Time
Information Monitoring In Pig Production”, in
Integrated management systems for livestock, Wathes, C.
M., Frost, A. R., Gordon, F. & Wood, J. D. (editors),
British Society of Animal Science and Institution
of Agricultural Engineers, Edinburgh, Occasional
Publication, No. 28, pp. 105-112.
Bishop-Hurley, G. J., Swain, D. L., Anderson, D. M.,
Sikka, P., Crossman, C. & Corke, P. 2007, “Virtual
fencing applications: Implementing and testing an
automated cattle control system”, Computers and
Electronics in Agriculture, Vol. 56, No. 1, pp. 14-22.
Black, J. L. 2001, “Swine Production – Past, Present
and Future”, in Palestras XXXVII Reuniao annual
Sociedade Brasileira de ZootecniaBrazil, Sociedade
Brasileira do Zootecnia.
Black, J. L. 2002, “Experience in the successful adoption
of AUSPIG by Industry”, in Animal Production in
Australia, Taplin, D. & Revell, D. K. (editor), Adelaide,
South Australia, Vol. 24, pp. 442-448.
Black, J. L. 2006, RD&E Priority Setting Workshop,
Australian Pork Limited, Canberra, Australia.
N08-AE06 Banhazi.indd 10N08-AE06 Banhazi.indd 10 3/07/09 10:08 AM3/07/09 10:08 AM
11
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
Black, J. & Scott, L. 2002, More beef from pastures:
current knowledge, adoption and research opportunities,
Meat and Livestock Australia, Sydney, Australia.
Black, J. L., Giles, L. R., Wynn, P. C., Knowles,
A. G., Kerr, C. A., Jone, M. R., Gallagher, N. L. &
Eamens, G. J. 2001, “A Review – Factors Limiting
the Performance of Growing Pigs in Commercial
Environments”, in Manipulating Pig Production VIII,
Cranwell, P. D. (editor), Australasian Pig Science
Association, Adelaide, Australia, Victorian Institute
of Animal Science, Werribee, Victoria, Australia, Vol.
VIII, pp. 9-36.
Black, J., Cook, P., Marr, G., Skirrow, S., Hitchens, B.
& Frey, B. 2002, Improving herd reproductive performance
through recording, analysis and action, Australian Pork
Limited, Canberra, Australia.
Bottcher, R. W. 2001, “An Environmental Nuisance:
Odor Concentrated and Transported by Dust”,
Chemical Senses, Vol. 26, pp. 327-331.
Bull, C. R., McFarlane, N. J. B., Zwiggelaar, R., Allen,
C. J. & Mottram, T. T. 1996, “Inspection of teats
by colour image analysis for automatic milking
systems”, Computers and Electronics in Agriculture,
Vol. 15, No. 1, pp. 15-26.
Chedad, A., Moshou, D., Aerts, J. M., Van Hirtum,
A., Ramon, H. & Berckmans, D. 2001, “Recognition
System for Pig Cough based on Probabilistic Neural
Networks”, Journal of Agricultural Engineering
Research, Vol. 79, No. 4, pp. 449-457.
Cumby, T. R. & Phillips, V. R. 2001, “Environmental
impacts of livestock production”, in Integrated
Management Systems for Livestock, Wathes, C. M.,
Frost, A. R., Gordon, F. & Wood, J. D. (editor), Selwyn
College, Cambridge, UK, BSAS, Edinburgh, pp. 13-21.
Dobos, R. C., Ashwood, A. M., Moore, K. & Youman,
M. 2004, “A decision tool to help in feed planning on
dairy farms”, Environmental Modelling & Software, Vol.
19, No. 10, pp. 967-974.
Durack, M. 2002, Precision Pig Farming – Where Are
You Pigs And What Are They Up To?, National Centre
for Engineering in Agriculture, Toowoomba, pp. 13.
Enting, J., Huirne, R. B. M., Dijkhuizen,
A. A. & Tielen, M. J. M. 1999, “A knowledge
documentation methodology for knowledge-
based system development: an example in animal
health management”, Computers and Electronics in
Agriculture, Vol. 22, No. 2-3, pp. 117-129.
Eradus, W. J. & Jansen, M. B. 1999, “Animal
identification and monitoring”, Computers and
Electronics in Agriculture, Vol. 24, No. 1-2, pp. 91-98.
Eradus, W. J. & Rossing, W. 1994, “Animal
Identification, key to farm automation”, in 5th
International Conference on Computers in Agriculture,
ASME, Orlando, Florida, pp. 189-193.
Frost, A. R. 2001, “An overview of integrated
management systems for sustainable livestock
production”, in Integrated Management Systems for
Livestock, Wathes, C. M., Frost, A. R., Gordon, F. &
Wood, J. D. (editor), Selwyn College, Cambridge,
UK, BSAS, Edinburgh, pp. 45-50.
Gates, R. S. & Banhazi, T. 2002, “Applicable
Technologies for Controlled Environment Systems
(CES) in Livestock Production”, in Animal Production
in Australia, Revell, D. K. & Taplin, D. (editors),
Adelaide, South Australia, Vol. 24, pp. 486-489.
Gates, R. S., Chao, K. & Sigrimis, N. 2001, “Identifying
design parameters for fuzzy control of staged
ventilation control systems”, Computers and Electronics
in Agriculture, Vol. 31, No. 1, pp. 61-74.
Geers, R. 1994, “Electronic monitoring of farm
animals: a review of research and development
requirements and expected bene ts”, Computers and
Electronics in Agriculture, Vol. 10, pp. 1-9.
Guerrin, F. 2004, “Simulation of stock control policies
in a two-stage production system: Application to
pig slurry management involving multiple farms”,
Computers and Electronics in Agriculture, Vol. 45, No.
1-3, pp. 27-50.
Hartung, J. 1986, “Dust in livestock buildings as a
carrier of odours”, in Odour prevention and control
of organic sludge and livestock farming, Nielsen, V.
C., Voorburg, J. H. & l’Hermite, P. (editor), Silsoe,
UK, Elsevier Applied Science, New York, Vol. 1, pp.
321-332.
Heinonen, M., Grohn, Y. T., Saloniemi, H., Eskola,
E. & Tuovinen, V. K. 2001, “The effects of health
classi cation and housing and management of feeder
pigs on performance and meat inspection  ndings
of all-in-all-out swine- nishing herds”, Preventive
Veterinary Medicine, Vol. 49, No. 1-2, pp. 41-54.
Hemsworth, P. H., Barnett, J. L., Beveridge, L. &
Matthews, L. R. 1995, “The welfare of extensively
managed dairy cattle: A review”, Applied Animal
Behaviour Science, Vol. 42, No. 3, pp. 161-182.
Holst, P. J. 1999, “Recording and on-farm evaluations
and monitoring: breeding and selection”, Small
Ruminant Research, Vol. 34, No. 3, pp. 197-202.
Jansen, M. B. & Eradus, W. 1999, “Future developments
on devices for animal radiofrequency identi cation”,
Computers and Electronics in Agriculture, Vol. 24, No.
1-2, pp. 109-117.
N08-AE06 Banhazi.indd 11N08-AE06 Banhazi.indd 11 3/07/09 10:08 AM3/07/09 10:08 AM
12
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
Kollis, K., Phang, C. S., Banhazi, T. M. & Searle, S. J.
2007, “Weight estimation using image analysis and
statistical modelling: a preliminary study”, Applied
Engineering in Agriculture, Vol. 23, No. 1, pp. 91-96.
Korthals, R. L. 2001, “Monitoring Growth and
Statistical Variation of Grow-Finish Swine”, in
Livestock Environment VI. Proceedings of the Sixth
International Symposium, Stowell, R. R., Bucklin, R. &
Bottcher, R. W. (editor), The Society for Engineering
in Agricultural, Food and Biological Systems,
Louisville, Kentucky, pp. 64-71.
Landesberg, P. 1999, “In the beginning, there were
Deming and Juran”, The Journal for Quality and
Participation, Vol. 11, pp. 59-61.
Lokhorst, C. & Lamaker, E. J. J. 1996, “An expert
system for monitoring the daily production process
in aviary systems for laying hens”, Computers and
Electronics in Agriculture, Vol. 15, No. 3, pp. 215-231.
Maatje, K., de Mol, R. M. & Rossing, W. 1997,
“Cow status monitoring (health and oestrus) using
detection sensors”, Computers and Electronics in
Agriculture, Vol. 16, No. 3, pp. 245-254.
Madsen, T. N. & Kristensen, A. R. 2005, “A model
for monitoring the condition of young pigs by their
drinking behaviour”, Computers and Electronics in
Agriculture, Vol. 48, No. 2, pp. 138-154.
Madsen, T. N., Andersen, S. & Kristensen, A. R. 2005,
“Modelling the drinking patterns of young pigs using
a state space model”, Computers and Electronics in
Agriculture, Vol. 48, No. 1, pp. 39-61.
Moshou, D., Chedad, A., Van Hirtum, A., De
Baerdemaeker, J., Berckmans, D. & Ramon, H. 2001a,
“An intelligent Alarm for Early Detection of Swine
Epidemics Based on Neural Networks”, Transactions
of the ASAE, Vol. 44, No. 1, pp. 167-174.
Moshou, D., Chedad, A., Van Hirtum, A., De
Baerdemaeker, J., Berckmans, D. & Ramon, H.
2001b, “Neural recognition system for swine cough”,
Mathematics and Computers in Simulation, Vol. 56, No.
4-5, pp. 475-487.
Mottram, T. T. 1997, “Automatic monitoring of the
health and metabolic status of dairy cows”, Livestock
Production Science, Vol. 48, No. 3, pp. 209-217.
Mullen, J. D. 2002, “Farm management in the 21st
century”, Agribusiness Review, Vol. 10, pp. 1-18.
Naas, I. A. 2001, “Precision Animal Production”,
Agricultural Engineering International: the CIGR Journal
of Scienti c Research and Development, Vol. 3.
Naas, I. A. 2002, “Application of Mechatronics
to Animal Production”, Agricultural Engineering
International: the CIGR Journal of Scienti c Research
and Development, Vol. 4.
Noguchi, N., Will, J., Reid, J. & Zhang, Q. 2004,
“Development of a master-slave robot system
for farm operations”, Computers and Electronics in
Agriculture, Vol. 44, No. 1, pp. 1-19.
Noordhuizen, J. P. T. M. & Frankena, K. 1999,
“Epidemiology and quality assurance: applications
at farm level”, Preventive Veterinary Medicine, Vol. 39,
No. 2, pp. 93-110.
Ordolff, D. 1997, “Experiments on automatic
preparation of milk samples in connection with
milking robots”, Computers and Electronics in
Agriculture, Vol. 17, No. 1, pp. 133-137.
Ordolff, D. 2001, “Introduction of electronics into
milking technology”, Computers and Electronics in
Agriculture, Vol. 30, No. 1-3, pp. 125-149.
Pedersen, B. K. & Madsen, T. N. 2001, “Monitoring
Water Intake in Pigs: Prediction of Disease and
Stressors”, in Livestock Environment VI, The Society
for Engineering in Agricultural, Food and Biological
Systems, Louisville, Kentucky, pp. 173-179.
Petersen, B., Knura-Deszczka, S., Ponsgen-Schmidt,
E. & Gymnich, S. 2002, “Computerised food
safety monitoring in animal production”, Livestock
Production Science, Vol. 76, No. 3, pp. 207-213.
Pomar, J. & Pomar, C. 2005, “A knowledge-based
decision support system to improve sow farm
productivity”, Expert Systems with Applications, Vol.
29, No. 1, pp. 33-40.
Schmoldt, D. L. 2001, “Precision agriculture and
information technology”, Computers and Electronics
in Agriculture, Vol. 30, No. 1-3, pp. 5-7.
Scho eld, C. P., Beaulah, S. A., Mottram, T. T., Lines,
J. A., Frost, A. R. & Wathes, C. M. 1994, “Integrated
Systems for Monitoring Livestock”, MAFF, Vol. 82.
Scho eld, C. P., Marchant, J. A., White, R. P., Brandl,
N. & Wilson, M. 1999, “Monitoring Pig Growth using
a Prototype Imaging System”, Journal of Agricultural
Engineering Research, Vol. 72, No. 3, pp. 205-210.
Scho eld, C. P., Wathes, C. M. & Frost, A. R. 2002,
“Integrated Management Systems for Pigs –
Increasing Production Ef ciency and Welfare”, in
Animal Production in Australia, Revell, D. K. & Taplin,
D. (editor), Adelaide, South Australia, Vol. 24, pp.
197-200.
N08-AE06 Banhazi.indd 12N08-AE06 Banhazi.indd 12 3/07/09 10:08 AM3/07/09 10:08 AM
13
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
Schon, H. & Meiering, A. G. 1987, “Computer-Aided
Control Improves Livestock Operations”, Agricultural
Engineering, Vol. 68, No. 7, pp. 15-18.
Shao, B. & Xin, H. 2008, “A real-time computer vision
assessment and control of thermal comfort for group-
housed pigs”, Computers and Electronics in Agriculture,
Vol. 62, No. 1, pp. 15-21.
Sliva, E., Banhazi, T. & Tonkin, M. 2007,
“Development of a prototype sensor for measuring
feed disappearance in livestock buildings”, in SEAg
2007, Banhazi, T. & Saunders, C. (editors), Adelaide,
Australia, SEAg, Vol. 1, pp. 221-225.
Snijders, J. M. A. & van Knapen, F. 2002, “Prevention
of human diseases by an integrated quality control
system”, Livestock Production Science, Vol. 76, No. 3,
pp. 203-206.
Stafford, J. V. 2000, “Implementing Precision
Agriculture in the 21st Century”, Journal of Agricultural
Engineering Research, Vol. 76, No. 3, pp. 267-275.
Street, M. J. 1979, “A Pulse-code Modulation system
for Automatic Animal Identification”, Journal of
Agricultural Engineering Research, Vol. 24, pp. 249-258.
Tukey, J. 1997, “More honest foundations for data
analysis”, Journal of Statistical Planning and Inference,
Vol. 57, No. 1, pp. 21-28.
Valdimarsson, G., Cormier, R. & Ababouch, L.
2004, “Fish safety and quality from the perspective
of globalization”, Journal of Aquatic Food Product
Technology, Vol. 13, pp. 103-116.
Van Henten, E. J., Van Tuijl, B. A. J., Hoogakker, G.-
J., Van Der Weerd, M. J., Hemming, J., Kornet, J. G.
& Bontsema, J. 2006, “An Autonomous Robot for
De-lea ng Cucumber Plants grown in a High-wire
Cultivation System”, Biosystems Engineering, Vol. 94,
No. 3, pp. 317-323.
von-Borell, E., Bockisch, F.-J., Buscher, W., Hoy,
S., Krieter, J., Muller, C., Parvizi, N., Richter, T.,
Rudovsky, A., Sundrum, A. & Van den Weghe, H.
2001, “Critical control points for on-farm assessment
of pig housing”, Livestock Production Science, Vol. 72,
No. 1-2, pp. 177-184.
Wathes, C. M., Abeyesinghe, S. M. & Frost, A. R.
2001, “Environmental Design and Management for
Livestock in the 21st Century: Resolving Con icts by
Integrated Solutions”, in Livestock Environment VI.
Proceedings of the Sixth International Symposium, The
Society for Engineering in Agricultural, Food and
Biological Systems, Louisville, Kentucky, pp. 5-14.
Webster, A. J. F. 2001, “The future of livestock
production: planning for the unknown”, in Integrated
Management Systems for Livestock, Wathes, C. M.,
Gordon, F. & Wood, J. D. (editors), Selwyn College,
Cambridge, UK, BSAS, Edinburgh, pp. 113-118.
Xin, H. & Shao, B. 2002, “Real-time Assessment
of Swine Thermal Comfort by Computer Vision”,
in Proceedings of the World Congress of Computers in
Agriculture and Natural Resources, Zazueta, F. S. &
Xin, J. (editors), 13-15 March, Iguacu Falls, Brazil,
ASAE, pp. 362-369.
Yen, T. Y. & Radwin, R. G. 2000, “Automated job
analysis using upper extremity biomechanical data
and template matching”, International Journal of
Industrial Ergonomics, Vol. 25, No. 1 SU, pp. 19-28.
Zhang, G., Strom, J. S., Blanke, M. & Braithwaite, I.
2006, “Spectral Signatures of Surface Materials in
Pig Buildings”, Biosystems Engineering, Vol. 94, No.
4, pp. 495-504.
N08-AE06 Banhazi.indd 13N08-AE06 Banhazi.indd 13 3/07/09 10:08 AM3/07/09 10:08 AM
14
Australian Journal of Multi-disciplinary Engineering Vol 7 No 1
“Precision livestock farming: A suite of electronic systems to ensure ...” – Banhazi & Black
THOMAS BANHAZI
Dr Thomas Banhazi is a Senior Research Scientist at the South Australian
Research and Development Institute (Livestock System Alliance) and his
research interests are related to intensive livestock housing. He has undertaken
studies to investigate the relationship between environmental conditions and
management factors in livestock buildings. He has also investigated methods
for reducing the impact of poor air quality on the respiratory health of both
humans and animals. His recent research interest is related to the development
of data acquisition and data management systems for the livestock industries
to improve the precision of production management. He is the Vice-President
of the International Commission for Agricultural Engineering (CIGR) – Section
II group and the Chair of the Australian Society for Engineering in Agriculture.
JOHN BLACK
Prof John L Black (AM FTSE) received his PhD from the University of Melbourne
in 1970. He joined CSIRO Division of Animal Physiology at Prospect, Sydney,
in 1971, where he continued the study of amino acid and energy requirements
of sheep for body and wool growth. A major component of his research was
the integration of physiological and biochemical concepts into mechanistic
simulation models. He led the team that developed the AUSPIG Decision
Support Software for pigs and became Assistant Chief of the Division in 1992.
In 1996 he left CSIRO and started a consulting company specialising in research
management for government and commercial organisations.
N08-AE06 Banhazi.indd 14N08-AE06 Banhazi.indd 14 3/07/09 10:08 AM3/07/09 10:08 AM
... To achieve a whole-herd approach to population health optimization, the data available for analysis must integrate all aspects of production. Banhazi and Black (2009) demonstrated that swine producers gathered data on multiple aspects of production, but data streams are typically collected and stored independently -and are thus beyond the easy reach of comprehensive analysis. The issue commonly confronted in the data integration process is incompatibility among data streams and software (Lenzerini, 2002). ...
Article
Swine wean-to-finish (W2F) mortality is a multifactorial, dynamic process and a key performance indicator of commercial swine production. Although swine producers typically capture the relevant data, analysis of W2F mortality risk factors is often hindered by the fact that, even if data is available, they are typically in different formats, non-uniform, and dispersed among multiple unconnected databases. In this study, an automated framework was created to link multiple data streams to specific cohorts of market animals, including sow farm productivity parameters, sow farm and growing pig health factors, facilities, management factors, and closeout data from a Midwestern USA production system. The final dataset (master-table) contained breeding-to-market data for 1,316 cohorts of pigs marketed between July 2018 and June 2019. Following integration into a master-table, continuous explanatory variables were categorized into quartiles averages, and the W2F mortality was log-transformed, reporting geometric mean mortality of 8.69 % for the study population. Further, univariate analyses were performed to identify individual variables associated with W2F mortality (p < 0.10) for further inclusion in a multivariable model, where model selection was applied. The final multivariable model consisted of 13 risk factors and accounted for 68.2 % (R²) of the variability of the W2F mortality, demonstrating that sow farm health and performance are closely linked to downstream W2F mortality. Higher sow farm productivity was associated with lower subsequent W2F mortality and, conversely, lower sow farm productivity with higher W2F mortality e.g., groups weaned in the highest quartiles for pre-weaning mortality and abortion rate had 13.5 %, and 12.5 %, respectively, which was statistically lower than the lowest quartiles for the same variables (10.5 %, and 10.6 %). Moreover, better sow farm health status was also associated with lower subsequent W2F mortality. A significant difference was detected in W2F mortality between epidemic versus negative groups for porcine reproductive and respiratory syndrome virus (15.4 % vs 8.7 %), and Mycoplasma hyopneumoniae epidemic versus negative groups (13.7 % vs 9.9 %). Overall, this study demonstrated the application of a whole-herd analysis by aggregating information of the pre-weaning phase with the post-weaning phase (breeding-to-market) to identify and measure the major risk factors of W2F mortality.
... For these reasons, a specialised service sector is required to maintain technological components, interpret data collected by sensors, formulate and deliver easy, appropriate advice to farmers on a regular basis, and engage users in technological development. Furthermore, it would be desirable to build a fully integrated framework through which all system components could be supplied to end-users in order to promote the functional deployment of PLF systems on farms [205]. ...
Article
Full-text available
Precision livestock farming (PLF) has spread to various countries worldwide since its inception in 2003, though it has yet to be widely adopted. Additionally, the advent of Industry 4.0 and the Internet of Things (IoT) have enabled a continued advancement and development of PLF. This modern technological approach to animal farming and production encompasses ethical, economic and logistical aspects. The aim of this review is to provide an overview of PLF and Industry 4.0, to identify current applications of this rather novel approach in different farming systems for food producing animals, and to present up to date knowledge on the subject. Current scientific literature regarding the spread and application of PLF and IoT shows how efficient farm animal management systems are destined to become. Everyday farming practices (feeding and production performance) coupled with continuous and real-time monitoring of animal parameters can have significant impacts on welfare and health assessment, which are current themes of public interest. In the context of feeding a rising global population, the agri-food industry and industry 4.0 technologies may represent key features for successful and sustainable development.
... Scalability is another major issue that can damage the sustainability of the network. Load balance among CHs must also be considered for the network's life span [2], [4]. Communication of data is a resource-hungry process, especially in terms of energy. ...
Article
Full-text available
Wireless Body Area Networks (WBANs) are emerging in the livestock industry for remote monitoring of cattle using wireless body sensors (WBS). The random mobility of animals acting as nodes causes the network’s topology to change rapidly, originating from scalability and reliability issues. Stable transmission of acquired data to the base station requires an intelligent clustering mechanism that reduces the energy consumption and fulfills the network’s constraints. Several clustering techniques are available as a solution, but these techniques yield numerous cluster heads, resulting in more energy utilization. Higher energy utilization lessens the effective life of WBSs and increases monitoring costs. This paper presents a metaheuristic approach for selecting optimal clusters in WBANs to realize an energy-efficient routing protocol for livestock health and behavior monitoring. The proposed approach employs Ant Lion Optimizer (ALO) to select the optimal clusters for different pasturage sizes using sensors of different transmission ranges considering user’s preferences about cluster density. The proposed technique with ALO is compared with other recent techniques such as Ant Colony Optimization, Grasshopper Optimization, and Moth Flame Optimization. The comparison results show the proposed technique’s effectiveness in realizing energy-efficient protocols of WBANs for remote monitoring applications.
... On the contrary, a robot can be efficiently operated all day long without any fatigue. So, the labour problem and its associated costs can be greatly reduced by introducing robotics and automation in traditional farming [2], [3]. In fact, some phenomena are especially considerable and important for developing countries like Bangladesh. ...
Article
Agriculture falls vulnerable to the impacts of climate change, which implies that any change in the climate can significantly affect the quality of the crops produced. Moreover, issues like labour problems, labour costs, productivity problems, etc. are stumbling blocks of traditional cultivation systems. Also, farmers need to produce more, at a higher quality, and in a sustainable manner to feed the increasing population. All these complications necessitate an automated system in this sector. In this project, we have built an autonomous robot that can detect ripe fruits or vegetables using colour detection mechanism and successfully harvest those with a robotic hand. The system can be subcategorized into three units-fruit picker, watering pump and sensing unit. The function of fruit picker is the identification of ripe fruits or vegetables by their colour, cut them off of the tree and then store them into suitable storage. The watering unit will pump water and necessary elements from the source tank and spread it in the field. The sensing unit is for the indication of the instant state of surroundings to help a farmer to choose the right steps to be taken. Although there are some initial costs to implement this system, this robot can precisely detect the right fruit (e.g. tomato or pepper) to be harvested and hence it can be used to pace up harvesting speed and save the labour and other associated costs. In this study, we have described the detailed processes followed to build this robot; hardware used; software implemented and assembly of the whole system as a functional unit.
Chapter
Full-text available
Smart livestock farming systems may provide real-time on-farm scenarios enabling fast interventions that benefit the current herd or flock. Smart decision-making technologies refer to more precise control over livestock production processes, helping farmers improve their productivity and profitability. Livestock process parameters are often faced with inaccurate, incomplete, or even conflicting data, and a way of minimizing this effect when processing data is the use of non-classical logic. The use of conceptual non-classical logic might improve smart tools allowing for non-intrusive assessment of health status and welfare, where information can be collected without the stress of disturbing or handling animals. Continuous monitoring can also offer a more complete picture of the overall health and/or well-being of animals rather than a view in time, as provided by traditional assessment. Alerting farmers to problems as they arise in real-time allows for immediate and targeted interventions to benefit the current herds or flocks. This book chapter introduces the fundamentals of managerial processes using non-classic logic and data mining and offers several applications to improve the decision-making of smart livestock farming.
Article
Full-text available
The main objective of this review was to assess currently available technologies to be used in Australian Precision Livestock Farming (PLF) systems and the research and development (R&D) work required before these systems can be implemented. Commercially available hardware and software products were assessed and a literature review of current R&D was undertaken to capture emerging technologies. One of the main recommendations of the review was that the implementation of PLF systems within Australia should occur via a coordinated development and linking of existing hardware and software components. The main implementation barriers identified are data compatibility and transfer. In order to achieve these aims investment in (1) industry coordination, (2) awareness raising/training programs (3) effectively targeted R&D projects and in the (4) establishment of demonstration/research sites would be required.
Article
Full-text available
The sophisticated global market place for livestock products demands safe, uniform, cheap, and environmentally-and welfare-friendly products. However, best-practice management procedures are not always implemented on livestock farms to ensure that these market requirements are consistently satisfied. Therefore, improvements are needed in the way livestock farms are managed. Information-based and electronically-controlled livestock production systems are needed to ensure that the best of available knowledge can be readily implemented on farms. New technologies introduced on farms as part of Precision Livestock Farming (PLF) systems will have the capacity to activate livestock management methods that are more responsive to market signals. PLF technologies encompass methods for measuring electronically the critical components of the system that indicate efficiency of resource use, software technologies aimed at interpreting the information captured, and controlling processes to ensure optimum efficiency of resource use and animal productivity. These envisaged real-time monitoring and control systems should dramatically improve production efficiency of livestock enterprises. However, as some of the components of PLF systems are not yet sufficiently developed to be readily implemented, further research and development is required. In addition, an overall strategy for the adoption and commercial exploitation of PLF systems needs to be developed in collaboration with private companies. This article outlines the potential role PLF can play in ensuring that existing and new knowledge is implemented effectively on farms to improve returns to livestock producers, quality of products, welfare of animals and sustainability of the farm environment.
Article
Full-text available
A user-friendly version of the scientific environmental monitoring kit used in Australia during a large air quality (AQ) study was created to enable routine environmental assessment in commercial livestock buildings and thus improve building environments and reduce pollutant emissions. The objective of the study was to produce a reliable and cost-effective hardware and software system for measuring six key environmental variables. The main components of the "BASE-Q" system are the two measurement boxes containing the sensors; the internet-based PC and pocket PC-based softwares; and the Users' Manual detailing recommended monitoring procedures. The first BASE-Q box contains sensors for measuring air temperature (AT), relative humidity (RH), the concentrations of ammonia (NH3) and carbon dioxide (CO(2)) up to 10-30 days depending on the logging interval. The second contains a vacuum pump and Venturi tubes and is used to measure concentrations of inhalable and respirable particles gravimetrically over an 8- to 10-h period. Engineering characteristics of the buildings are recorded on site and the collected data stored and processed by the internet or PC-based BASE-Q program. The program automatically calculates the concentrations and emission rates of the different airborne pollutants from individual buildings by using prediction models developed during related studies as a pre-screening exercise before actual measurements are undertaken. The size and weight of the monitoring hardware have been markedly reduced to improve ease of installation and transport. The monitoring equipment has been simplified and waterproofed to improve ease of deployment and disinfection. The special software has greatly simplified data management and reporting. These improvements have reduced the labor input required for operating the system and thus minimized the cost of AQ monitoring. This will enable producers and consultants to measure AQ routinely on farms, reducing worker OH&S risks, improving environmental outcomes, and potentially improving production efficiency.
Article
The application of modern process control techniques to poultry production is outlined. Compact dynamic data-based models are proposed to describe and control the metabolic responses of broiler chickens to variations in the micro-environment. The dynamic response of heat production to step changes in air temperature and light intensity could be modelled with a , on average, of 0.83 and 0.93 respectively. Using recursive parameter estimation techniques, the time-variant response of animal growth to food supply could be predicted on-line with a prediction error of a maximum of 5%, three to seven days ahead depending on the type of feeding schedule. We argue that the potential conflicts between the environmental, financial and biological pressures on sustainable poultry production can be resolved through the development of integrated management systems using process control techniques.
Article
European agriculture is shifting toward greater process control in livestock production. The computer-based systems help managers to increase operation efficiency and enhance product quality. Considerable progress is now being made toward combining data acquisition, process optimization, and input/output control into an overall control system composed of sensors and signal-processing circuitry for process observation, computer-activated equipment, and control devices. The system also includes robotic elements and control dedicated microprocessors or microcomputers with interface to management computers.
Conference Paper
Further advances in the design and management of livestock environments require integration of (i) the scientific approaches to the study of an animal’s responses to its physical environment; and (ii) the management of both production and environmental processes. Approaches to assessing the impact of multiple factors in other fields of scientific research can be applied to the interaction between an animal and its environment, thereby allowing more realistic modelling. Process control techniques, that have been proven in other industries, can now be applied to livestock production in the form of integrated management systems. This will enable the industry to meet tight product specifications while satisfying society’s demands to reduce the environmental impact of intensive livestock production.