NZTE - An analysis of the economic advantages to New ...

An analysis of the economic advantages to New Zealand dairy farmers

of extensive electronic monitoring of dairy cows.

Prepared for

The Integrated Farm Management Systems Group

by

Colin Kingston, Director, Business Development Rod Claycomb, Chief Executive

Sensortec Ltd., Hamilton

December 2005

ii

Foreword All factual information (including but not limited to historical data) set out in this report is believed by Sensortec to be correct as stated, but it is not warranted as either complete or accurate. All other text or projections in the report must be treated as an expression of opinion, and cannot be warranted in any way. Sensortec shall have no liability for any reliance which may be placed by New Zealand Trade and Enterprise (“NZTE”) on any expression of opinion set out in the report nor for any consequences of NZTE acting on any recommendation made in the Report. The copyright in this report remains the property of Sensortec Ltd. No copyright is assigned to NZTE, the Integrated Farm Management Systems Group or any of its members.

iii

Abbreviations AI Artificial Insemination AIDC Automatic Identification and Data Capture AMS Automatic Milking System ASP Application Service Provider BCMS British Cattle Movement Service BSE Bovine Spongiform Encephalopathy CCIP Canadian Cattle Identification Programme CFIA Canadian Food Inspection Agency CIDR Controlled Intravaginal Drug Release CTS Cattle Tracing Scheme DEFRA Department for the Environment and Rural Affairs EBV Economic Breeding Value EIA Enzyme immunoassay EID Electronic Identification ELISA Enzyme-linked Immunosorbent Assay EU European Union FDA Food and Drug Administration FSA Food Standards Agency FMD Foot and Mouth Disease ICAR International Committee for Animal Recording IDF International Dairy Federation INGO International Non-Governmental Organisation ISO International Standards Organisation NCIMS National Conference on Interstate Milk Shipments NFIS National Flock Identification Scheme NLIS National Livestock Identification System NMC National Mastitis Council RFID Radio Frequency Identification RIA Radioimmunoassay SCC Somatic Cell Count USAIP United States Animal Identification Plan USDA United States Department of Agriculture USFDA United States Food and Drug Administration

iv

Table of contents

Foreword ............................................................................................................................ ii

Abbreviations..................................................................................................................... iii

1. Introduction.................................................................................................................... 1

1.1 Scope of review .....................................................................................................................1

1.2 International market context ..............................................................................................1 1.2.1 Current technology status..............................................................................................................2

1.3 The New Zealand context ....................................................................................................3

2. Background .................................................................................................................... 4

2.1 Pasture based dairy systems................................................................................................5

2.2 High yield dairy systems......................................................................................................5

2.4 Low yield dairy systems.......................................................................................................6

3. Market drivers ............................................................................................................ 7

3.1 Regulatory compliance ........................................................................................................7 3.1.1 Animal movements .......................................................................................................................7 3.1.2 Disease management.....................................................................................................................9 3.1.3 Food chain traceability..................................................................................................................9 3.1.4 Bio-terrorism...............................................................................................................................10 3.1.5 Veterinary compliance ................................................................................................................10 3.1.6 Access to export markets ............................................................................................................11 3.1.7 Animal welfare............................................................................................................................11 3.1.8 Consumer concerns .....................................................................................................................12

3.2 Productivity improvement.................................................................................................12 3.2.1 Labour cost..................................................................................................................................12 3.2.2 Scarcity of labour ........................................................................................................................13 3.2.3 Family labour ..............................................................................................................................13 3.2.4 Automation .................................................................................................................................13 3.2.5 Management Systems integration ...............................................................................................14 3.2.6 Connectivity................................................................................................................................14

3.3 Genetic improvement.........................................................................................................15

4. Overview of Technology............................................................................................... 17

4.1 Discrete and systemic technologies ...................................................................................17

4.2 Electronic identification.....................................................................................................17 4.2.1 Management tags ........................................................................................................................18 4.2.2 Registration tags..........................................................................................................................18 4.2.3 FDX and HDX ............................................................................................................................19 4.2.4 Ear Tags ......................................................................................................................................20 4.2.5 Bolus technology.........................................................................................................................20 4.2.6 Subcutaneous implants................................................................................................................20 4.2.7 New technology developments ...................................................................................................21

4.3 Process control and information management systems..................................................21

4.4 Milking systems ..................................................................................................................23

v

4.4.1Conventional milking systems .....................................................................................................23 4.3.2 Automation of conventional parlours..........................................................................................23 4.3.3 Automatic milking systems.........................................................................................................24

4.4 Feeding systems ..................................................................................................................25

4.5 Fertility management.........................................................................................................26 4.5.1 Pedometers..................................................................................................................................27 4.5.2 Progesterone sensing...................................................................................................................27

4.6 Mastitis management .........................................................................................................28 4.6.1 Conductivity measurement..........................................................................................................29 4.6.2 Somatic cell count.......................................................................................................................29 4.6.3 Milk Amyloid A..........................................................................................................................30 4.6.4 L-lactate ......................................................................................................................................30

4.7 Milk Quality Management ................................................................................................30 4.7.1 Contaminated milk......................................................................................................................30 4.7.2 Abnormal milk ............................................................................................................................31 4.7.3 Undesirable milk .........................................................................................................................32

4.8 Animal Health Management .............................................................................................32 4.8.1 Body condition scoring ...............................................................................................................32 4.8.2 Locomotion monitoring ..............................................................................................................33

4.9 Overall Farm Management ...............................................................................................33 4.8.1 Pasture Management ...................................................................................................................33 4.8.2 Infrastructure Support .................................................................................................................33 4.8.2 Business Management.................................................................................................................33

5. Conclusion................................................................................................................ 34

1

1. Introduction

1.1 Scope of review This report was commissioned and funded by New Zealand Trade and Enterprise (‘NZTE’) within the framework of research work being undertaken by the Integrated Farm Management Systems Group (“IFMS”). IFMS is collaborative research group set up and managed by Innovation Waikato Limited, MediaLab South Pacific Limited and funded by private sector interests and New Zealand Trade and Enterprise. It encompasses wide ranging rural interests including agritech equipment manufacturers, research and academic organizations, Information Technology (‘IT’) and telecommunications providers. The Group’s objective is to investigate and commercialise integrated technologies to improve the productivity and marketing of the New Zealand primary sector. The report provides an analysis of the international business environment for electronic animal monitoring technologies and considers the specific circumstances of New Zealand and the economic advantages that New Zealand dairy farmers could derive through the widespread adoption of such technologies. The report is focused on the users of electronic animal monitoring technologies rather than the owners of the IP, such as manufacturers and marketers. For the purpose of this report, electronic animal monitoring technologies are defined as those electronic technologies, comprising both hardware and software, which monitor animal criteria that are economically significant and that facilitate the animal management process. In addition, although not always directly associated with the users of the technologies, the technologies may also serve to address animal welfare issues and associated requirements and the potential to use monitoring to demonstrate requirements are being met or within compliance. Finally, a particular emphasis has been placed on monitoring technologies that, once implemented, actually result in a decision that can be acted upon in a continuous fashion, as opposed to more isolated or one-off type monitoring opportunities.

1.2 International market context It is important to recognize that the electronic animal monitoring technologies business environment is set within the wider context of an increasingly competitive international market for agricultural and dairy produce. As agricultural subsidies and market protection mechanisms in the developed world are progressively dismantled in line with the international trend towards more liberal trade agreements between countries and trading blocs, agricultural produce markets are becoming exposed to increasingly international competition. The overriding challenge that these markets now face is to supply increasingly discerning and well informed consumers with food products that are of high quality and that are price competitive, whilst at the same time guaranteeing farmers a sustainable income. As the mechanisation of livestock farming, which was ongoing throughout most of the twentieth century, begins to reach the limits of incremental productivity improvement in

2

many parts of the developed world, so this process is giving way to increasing use of automation and information management technologies as a means of further improving efficiency and reducing production costs. Electronic animal monitoring technologies have a pivotal role to play in facilitating this automation and information management process.

1.2.1 Current technology status The advent of the animal monitoring technologies market can be traced back several centuries to the time when animal identification techniques were first used to prove ownership. Whilst early systems such as ear punch marks, horn brands and tattoos are still in use today, even in the developed agricultural markets, these have been widely replaced by alternative technologies, such as freeze branding, colour marking and ear tags. In more recent years there has also been a strong uptake in a variety of electronic identification and animal monitoring technologies. Not only are these technologies more reliable, they also offer numerous additional benefits, which have seen their utilisation extend beyond proof of ownership to fulfill a variety of farm management and regulatory compliance applications. Of greater significance is the fact that the value of these technologies is also being leveraged by their increasing integration with the automation of a variety of animal husbandry processes to create fully integrated and powerful process control and animal management information systems. A prime example of this is the development of robotic or automatic milking systems (AMS). Whilst the information available from such systems is of significant value in improving the efficiency of animal husbandry on the farm, such systems are also being used to add significant value beyond the farm gate. As the demand from consumers, retailers and regulators for better food traceability gathers momentum a key requirement of livestock farmers in the future will be the ability to collect, record and pass on information both cheaply and accurately to third party stakeholders down the food and distribution chains. As the most labour intensive livestock farming segment, the dairy farm equipment market has seen the strongest drive towards automation as a means of improving farm productivity and profitability. Automation has been embraced in particular in response to the trend towards increasing rural labour costs and growing scarcity of farm labour, combined with the added pressure of declining farm incomes as a result of reducing farmer prices for milk and dairy products. Another factor in the growth of this segment is that automation substantially assists large dairy farms in realising economies of scale. Automation is also being embraced in many countries to improve on-farm process control in response to milk buyers who, responding to consumer demand, are setting increasingly high milk quality standards for milk producers. Currently, automation is widely used in the labour intensive areas of milking and feed management. However, the focus is also now beginning to shift more towards improved process control and automation of the herd management processes in order to improve on-farm performance. In this regard, the two areas which are of greatest commercial interest are reproductive management and mastitis management, since mastitis and sub-optimal fertility constitute the two biggest herd management costs incurred by

3

dairying. Together they are estimated to cost the dairy industry globally at least US$8 billion per annum (Source: HortResearch, New Zealand).

1.3 The New Zealand context In the context of this increasingly competitive global economy within which New Zealand has to compete to sell its dairy produce the New Zealand dairy industry faces the following challenges if it is to retain its position of cost leadership or identify some alternative source of sustainable competitive advantage:

• Increased global competition presents the challenge of reducing cost to compete in a climate of forthcoming reducing world market prices as New Zealand’s competitors reduce their costs.

• Growth in New Zealand, and thus economy of scale efficiencies, are restricted by

the fact that the base resource of land is limited. The estimated area suitable for dairying is about 23 million ha. of which 87% is currently utilized. For New Zealand, therefore, the challenge seems to one of both reducing production costs and increasing milk production from the same land resource base. This means intensification of milk production by increasing milk production off the existing land, but without having a negative impact on environmental issues, through better use of technology and better management of the production process.

• Improvement of returns per unit of milk is the third challenge which New Zealand

has set for itself with the industry target of increasing milk solids output by 50 per cent by 2010.The focus of the industry is moving progressively towards higher value added products based on milk. There is a role for identification of farms or even individual cows that have a higher milk solids production or higher production of specialised higher value milk component.

Electronic monitoring of dairy cows and integration of on-farm systems and processes have a pivotal role to play in facilitating the process of meeting each of these challenges.

4

2. Background The type of dairy farming practiced in the principal farming regions of the world varies considerably and the pattern of farming systems which has emerged is the result of a number of different interrelated geographical, economic and social factors. The adoption of electronic animal monitoring technologies is also strongly influenced by these same factors which explain the differing levels and types of automation that can be found in the various dairy farming regions.

Of critical importance to understanding the international dairy market is an appreciation of the pattern of consumption for livestock produce and the demands that this makes on the livestock sector, including the dairy sector. Historically the vast majority of livestock and dairy production is consumed in the country where it is produced. However, over the last 5-10 years there has been a significant growth in international trade in livestock and dairy production, fuelled by the trend towards more liberal trade agreements between countries and trading blocs. Consequently, trade in agricultural production now accounts for 10% of world production (Source: OECD) and several countries have specialised in areas of livestock production, driving the growth in productivity and performance to gain access to export markets.

There is thus a broad split between livestock production for local markets and livestock production for international markets. Furthermore this has led to the emergence of two pricing regimes that have profound implications for the methods of livestock farming employed in these markets.

On the one hand, there is a range of domestic market prices paid to local farmers, which may be considerably higher than the world market price, especially if tariffs or quotas are used to protect domestic prices. In some instances, such as parts of the EU, these pricing mechanisms are also used by governments to maintain smaller herd sizes and to keep rural people and services in place.

Such policies have also resulted in overproduction in many markets. For example, the EU has operated a quota system since 1984, the intention being to correct the imbalance between dairy output and consumption in the community by charging a superlevy on all milk produced above the prescribed quota level. The system has been extended numerous times beyond its initial 5 year period and the system is set to continue until 2008. However, it is clear that the world trend is towards freer trade in agricultural produce and the dismantling of protective barriers is exposing developed countries more and more to world market prices. This is driving the trend towards productivity improvement and automation. On the other hand, the laws of supply and demand determine the world market price for the internationally traded agricultural produce. These market forces, to which New Zealand is exposed, drive the quest for production efficiency to maximize the comparative advantage enjoyed in these areas of specialised livestock production that concentrate on competing in the international market. The span between the highest domestic market prices and the world price is considerable. The following table shows the price range for milk:

5

Below US$ 15/100kg Brazil US$ 15-20/100kg Estonia, Poland, Argentina, India, Australia, New

Zealand US$ 20-25/100kg Czech republic, Hungary, Colombia US$ 25-35/100kg EU countries and US Idaho US$ 35-45/100kg US New York, US Wisconsin, Israel US$ 45-50/100kg Switzerland

Source: IFCN (2002)

As can be seen milk prices range from US$ 13 to US$50/ 100 kg , the highest prices being paid in Switzerland and parts of the EU and the lowest prices in Brazil. Milk prices obtained by farms in Oceania and most South American farms can be regarded as the world market price for milk. EU prices are about double this price and prices in Switzerland are triple this figure. Within the framework of this pattern of consumption and pricing four broad categories of dairy farming can be identified as follows which have evolved in response to the geographical, economic and social factors at play.

2.1 Pasture based dairy systems Pasture based dairy systems are found predominantly in New Zealand, Australia, United Kingdom and Ireland. These farms produce milk mainly from grass, grass silage or hay with a small amount of concentrates. Management focus is much more on milk yield per hectare as a unit of production than it is on yield per cow, which tends to be the case in high yield dairy systems. This means that cows tend to have lower yields, typically in the range 4,000 – 6,000 kg per lactation and efficiency is sought by maximising grass production and using genetic selection. Herds tend to be large and milking systems focus on large throughputs of low to medium yielding cows with little or no automation in order to minimise capital investment. The temperate climate in these countries enables cows to graze on pasture for much of the year and requirements for cow housing are minimal. Production is also seasonal to match the period of grass growth, all of the herd calving within a short period of time each year. With grazing systems labour efficiency is achieved by having the cows go to the feed rather than having feed delivered to them. In essence, the cows are both the forage harvesters and the effluent spreaders! These farms focus on production of milk for manufactured dairy products such as butter and cheese and the low cost of production enables them to produce at world market prices to supply the commodity markets. These farms often use hired labour to supplement family labour.

2.2 High yield dairy systems High yield dairy systems are typical of the farms found in the countries of continental Europe such as France, Germany, Sweden and parts of the United Kingdom as well as areas like the mid-West of the United States. Milk yields are in the range of 6,000-8,000 kg, the higher yields being achieved by the use of supplementary feeding of concentrates. This means that the cows are housed in farm buildings for some of the

6

year. Unlike the pasture-based systems the focus is on yield per cow rather than yield per hectare, so technologies such as automated feeding and milk yield recording are widely used to support individual and herd management strategies to optimise production. The focus is very much on the cow as a unit of production. Milking systems are generally more automated to facilitate individual cow management and often include in-parlour feeding. Production is geared to supply local liquid milk markets as well as the local markets for manufactured milk products. These farms are predominantly found in high labour cost countries. They tend to be family farms and make significant use of family labour and automation in order to avoid hired labour. 2.3 Very high yield dairy systems Very high yield dairy systems are typical of the farms found in the US, the Netherlands, Sweden, Denmark, Finland, Israel and parts of Germany. Milk yields are in the range of 8,000 – 10,000 kg and there is a strong focus on maximising yield per cow by means of genetic selection and targeted feeding. The cattle are often housed inside almost all of the year and are milked through highly automated parlours that are linked to integrated herd management systems that focus on maximising farm productivity. Whilst the smaller herds are family farms in countries like Denmark and Sweden that avoid hired labour by automating, there are some large industrial farms in areas like California and Israel that make extensive use of hired labour.

2.4 Low yield dairy systems Other less important dairy farm systems include low yield (1,000kg – 4,000kg) small herd farms such as those found in Brazil, India and parts of Poland. There are also dual purpose cattle farms with low yielding cows in some Eastern European countries. Production is geared for local consumption and the farms tend to use mainly family labour.

7

3. Market drivers Operating within each of the dairy farming contexts described above a number of political, economic, social and technological influences can be identified, as illustrated in Appendix 1. From these influences a number of significant market drivers of change in response to the consequent needs of farmers can be identified that add momentum to the automation process and create demand for animal monitoring technologies. These drivers fall under three broad headings: regulatory compliance, productivity improvement and genetic improvement: These are explored in the following sections.

3.1 Regulatory compliance The farming industry has never been under so much pressure to complete ever increasing mounds of legislative paperwork and maintain accurate records of its livestock. All of this additional work has to be achieved in an environment of reducing produce prices, which means there is increasing pressure to reduce costs by making efficiency savings in this area. As an example, National Milk Records (NMR) plc in the United Kingdom has identified that a 100 cow dairy farm in England currently generates some 36,000 pieces of information annually and it predicts that this volume is bound to increase. One important aspect of regulatory compliance in some countries concerns the proper implementation of subsidy or premium schemes, which require clear identification of animals, reliable counting and varying degrees of traceability. In particular, eligibility for premium subsidy can depend on monitoring a number of parameters, such as the animal age, species, breeding type and number of animals on each farm. In this regard Breeders Associations have been emphasising for some time the need to introduce more reliable identification systems that are capable of providing high assurance of authentication of animal origin as well as production performances. Whilst New Zealand dairy farmers are not currently exposed to the same level of regulatory compliance as dairy farmers in other parts of the world such as the EU it is clear that the current international trading environment is likely to increase the demand on New Zealand for increasing levels of data collection for regulatory compliance purposes.

3.1.1 Animal movements Another area of growing importance with regard to regulatory compliance is the monitoring of animal movements. In fact, regulations now exist in most parts of the world in relation to mandatory identification of livestock and their movements.

• Under current EU legislation all cattle have to be identified with an ear tag in each ear. All calves must be double tagged with an approved ear tag within 30 days of birth, whilst all dairy calves must be double tagged within 36 hours of birth. This information is then used to monitor all animal movements from birth through to death. In the UK, the British Cattle Movement Service (BCMS) runs this mandatory cattle identification and registration scheme called the Cattle Tracing System (CTS). Similar organisations exist in the other EU member states.

8

• In September 2003 the US Department of Agriculture (USDA) introduced the U.S. Animal Identification Plan (USAIP), which called for the identification of all 30 million cattle in the US with RFID by July 2005, the goal being to have the capability of identifying animals within 48 hours.

• In Canada, the Canadian Cattle Identification Program (CCIP) arose as an industry-led initiative to protect Canada’s export market. Under the Program, which is regulated and enforced by the Canadian Food Inspection Agency (CFIA) cattle are individually identified with official ear tags to allow for tracking from the abattoir back to the farm of origin. As from January 2003, all animals must be tagged to qualify for registration using a three–read tag system whereby each animal’s unique national identification number can be read three ways – RFID, bar code and visual.

• Australia has a system for the identification and tracing of livestock called NLIS (National Livestock Identification Scheme). It is a permanent whole-of-life identification system that enables individual animals to be tracked from property of birth to slaughter for food safety, product integrity and market access purposes. NLIS uses RFID tags to track cattle. To date 35,000 out of 100,000 producers in Australia use RFID tags with some 7.5 million cattle out of 28 million in the country equipped with electronic tags.

• In Mexico the Confederación Nacional Ganadera (CNG), the national cattlemen’s association of Mexico, is currently conducting technology trials with the goal of both individually identifying cattle and labelling the beef products derived from them. CNG plans to develop a system that will both provide carcass quality data and protect and enhance its national and international markets. It is examining use of plastic or metal ear tags, microchip or transponder implants, and boluses. CNG would ultimately like to implement a system that would see a barcode on a package of beef identifying the animal from which it came.

• In 2003 the Japanese government announced a mandatory traceability scheme in the wake of mad cow disease scares in Japan.

• Brazil is also instituting a traceability program covering beef bound for the European Union, although implementation has run into some serious difficulties. The system is intended to make it possible to compile a comprehensive record of cattle movement and Brazilian companies are to use the new system to track beef destined for the EU. Tracking is to be accomplished via chips affixed to the animals that transmit data to the agriculture ministry via satellite. At ministry headquarters, specialised software will be used to track each animal throughout its lifetime.

In New Zealand the responsibility for tracing animal movements is currently split between Livestock Improvement Corporation (LIC), using their MINDA database, and Animal Health Board (AHB) for non-LIC members. The system is geared to eradication of tuberculosis in cattle rather than the establishment of a comprehensive food traceability program. As identified in 3.1.6 compliance with animal movement recording practices in key export markets is likely to become a requirement for entry to most if not all of New Zealand’s dairy export markets in the not too distant future.

9

3.1.2 Disease management. In addition to tracking animal movements the outbreak in recent years of livestock diseases such as BSE, Foot and Mouth Disease (FMD) and Blue Tongue in the European Union, followed by the outbreak of BSE in North America, have shown that current livestock identification systems are not efficient and reliable enough to provide adequate traceability and veterinary monitoring of livestock species when such epidemics occur. These emergencies, particularly the FMD crisis in the UK during which an estimated 4.2 million sheep and cows had to be destroyed, have underlined the need not only to improve animal traceability but also to implement appropriate animal health surveillance schemes. Estimates of the cost of the outbreak to the UK include direct compensation for slaughtered animals totaling in excess of £1 billion (NZ$ 2.6 billion) and associated expenditure on disease control (including 1600 veterinary surgeons, 2000 military personnel, slaughtermen, transport, disinfection procedures and excavation equipment) totaling in excess of £1 billion (NZ$ 2.6 billion). In addition there were significant losses incurred by the livestock industry through movement restrictions (e.g. feeding animals with no prospect of sale), and substantial losses incurred by other industries such as tourism and small rural businesses. The Institute of Directors has estimated that the likely cost to the wider economy is in the region of £20 billion (NZ$ 52 billion). Furthermore, massive eradication procedures in cases of diseases like Swine Fever and FMD are no longer in favour with the general public or the farming sector in many countries. Consequently the animal production sector in these countries is pushing for improved identification and registration methods to allow selective slaughtering and the use of marker vaccines. Of particular concern for the health authorities in the area of disease management is the monitoring of zoonotic diseases; that is diseases which can be transferred from animals to humans. New Zealand should learn the lessons of countries like the UK and the US in utilising electronic animal monitoring technologies as part of a cost effective yet comprehensive risk management program for disease management in order to protect its economically important dairy sector.

3.1.3 Food chain traceability Food safety has now become a high profile issue in livestock farming and the concept of traceability in the food chain has been gathering momentum for several years, particularly in the wake of the discovery in 1999 of dioxins in chicken meat and eggs in Belgium as a result of contaminated fats being supplied to an animal feed manufacturer, and also following the BSE crisis in the UK in 2001. There was also a similar case in the same year when German cows' milk was also found to be contaminated by dioxins, the source of which was eventually identified as feed made of Brazilian citrus pulp mixed with contaminated mineral lime. Amidst concerns for public health this issue has received considerable impetus from government agencies that are keen to see improved control measures that rely on better identification of livestock and integrated traceability systems throughout the food chain. In addition to political pressure for better food chain traceability the push for EID and traceability systems is also coming from the livestock industry itself, particularly the beef

10

industry in the wake of BSE scares, which have depressed beef consumption and cost the industry billions in reduced consumer demand and lost sales. In response to consumer demand, food retailers are also helping to stimulate the demand for such systems. The consequence of these various requirements for better traceability is that the simple visual identification of animals through tagging is progressively being replaced by electronic systems that provide full traceability from farm to consumer (now commonly referred to as ‘farm to fork’). There is now a growing need for real time animal identification and integration of data management systems throughout the supply chain and this is accelerating the replacement of visual identification systems by electronic identification technologies in many parts of the world.

3.1.4 Bio-terrorism In the United States the pressure for food chain traceability has been given added momentum by the recent introduction of the US Bioterrorism Act 2002. Following the events of September 11th, 2001, the US introduced a number of measures to improve the security of its citizens. This included the Bioterrorism Act 2002 which is aimed at meeting a bio-terrorist threat - the use of microbes and poisons - to attack the U.S. food supply. The purpose of the Act is to allow the Food and Drug Administration (FDA) and other authorities to quickly determine the source and cause of any deliberate or accidental contamination of food. The regulations are intended to allow the agency to track foods implicated in emergencies such as intentional contamination and means that food producers and other food handlers need to maintain records that identify the immediate previous source of all food received and the immediate recipient of all food released. An eight month implementation phase began in December 2003 during which violations did not necessarily lead to fines. However, in August 2004 an enforcement phase began when FDA inspectors have started to use the authority to impose stricter penalties for violations.

3.1.5 Veterinary compliance The regulatory environment relating to the administration of veterinary treatments is also making increasing demands on both veterinarians and farmers to maintain comprehensive records of animal treatments. In particular the use of antibiotics is subject to increasingly stringent controls and penalties are imposed on farmers who allow antibiotic residues to enter the food chain, for example by not adhering to the prescribed withdrawal times and allowing milk with residues to enter the bulk tank on the farm. Records from the USFDA show that in the 1960s the number of antibiotic contaminations exceeded 5% of all milk tanker loads. By 1991, owing to the imposition of new regulations the number of positive loads had decreased significantly to 0.10% even though the methods for residue detection had become more sensitive. By 1995 the level of contamination had decreased even further to 0.06%. However, even at these relatively low levels of contamination the absolute number of cases is significant. The National Milk Drug Residue Database maintained by the FDA and NCIMS reveals the

11

following statistics for the United States during the period 1st October 2002 to 30th September 2003:

• Total samples tested: 3.571.834 • Number of positive tests: 1,899 • Percentage positive: 0.053% • Contaminated milk disposed of: 70,106,000 pounds

At a market price of US$15 per cwt this amounts to a market value of disposed milk of some US$10.5 million per annum for the US market alone. Electronic monitoring technologies are relevant to veterinary compliance both in terms of facilitating compliance and improving the productivity of veterinarians who can, for example, gain remote access to farm records and monitor herds without visiting the farm.

3.1.6 Access to export markets Compliance with market access requirements imposed by markets such as the EU and the US are also proving to be strong drivers for those countries that are dependent upon these export markets for their agricultural production. Compliance has effectively exported the EU and US-led drive for better food chain traceability to all those exporting countries that are dependent upon the EU and US as markets for their livestock production. Compliance is becoming more and more the price to be paid for market access. For example, Brazil’s Agriculture Minister Marcus Pratini has described their traceability system as a key element in Brazil’s plan to expand beef exports aggressively over the next few years. Clearly the ability of New Zealand to comply with market access requirements is critical to its future success in key export markets such as the US and the EU.

3.1.7 Animal welfare In the current socio-economic environment of the developed world, particularly in many northern European countries, animal welfare is becoming an increasingly important issue for livestock farming and particularly for livestock produce marketing. Whilst many consumers still only have a somewhat vague understanding of the practical meaning of animal welfare in livestock production these considerations are beginning to influence consumer preferences. The growth in demand for organic food, particularly in the Scandinavian countries, provides clear evidence of this trend, as does the growth in demand in the UK for such products as free range eggs, organic milk and outdoor reared bacon. These consumer preferences are also often linked to a perception of enhanced food quality. More important still is the fact that animal welfare considerations are being driven by food buyers and distributors, particularly the increasingly powerful supermarket chains. They are beginning to make demands on farmers, not only in terms of farming practices, but also in terms of systems for collecting relevant data and maintaining accurate records about animal health, animal treatments and animal performance.

12

3.1.8 Consumer concerns In many markets the consumer is also becoming increasingly well informed about food products and therefore more discerning as a customer. This means that if the livestock farming sector is to sustain its markets it has to take account of the increased awareness by the consumer of health and dietary considerations. It also means that it has to ensure that it is in a position to collect and make available to its customers relevant information upon which they will base their purchase decisions.

3.2 Productivity improvement Whilst some electronic animal monitoring technologies are becoming essential for the purpose of regulatory compliance and others are helpful in terms of reducing the increasing financial and administrative burden of regulation, arguably their greatest economic benefit to the farmer lies in their potential to improve the efficiency of livestock operations on the farm by engineering costs out of the farm production process and improving on-farm productivity and profitability.

3.2.1 Labour cost Internationally, there has been a long term trend of labour reduction in livestock farming. Statistics for the Netherlands, for example, show that in 1950 about 300 hours of labour per cow were required annually, but by 1980 this had been reduced to 50 hours per cow. Similar reductions in the labour required for milk production have been reported in the USA where milk was produced in 1985 with only 42% of the labour per cow compared with 1975 and 33% of the labour per unit of milk produced. In general terms labour is one of the main cost components for dairy farming, accounting for between 20% and 40% of total farm costs (Source: IFCN, 2002). However, labour costs and labour productivity vary widely from country to country. The highest wages are found in the Netherlands, Denmark and Sweden, at around US$16 per hour (Source IFCN, 2002) where farming has to compete with industry in a high labour cost environment. Improved mobility of rural labour has also accelerated the rural exodus, even in countries like France where distances to urban centres of employment are significant. These countries see the most efficient use of labour on farms as a result of the higher opportunity cost of mechanisation and automation. The lowest wages, by contrast, are found in countries like Brazil and Colombia where they are around 10% the level of the Netherlands at under US$2 per hour. In addition to the cost of labour there is in some countries, such as those in the EU, the added burden of a growing regulatory environment that surrounds employment, which adds to the cost of employment. These requirements have become a strong disincentive for many farms to employ labour and this has also driven the trend towards automation on farms as a substitute for labour. In New Zealand the cost of hired farm labour is becoming an issue of increasing importance. Farmers are demanding individuals who are motivated, have a higher level of education, and have a good practical knowledge to ensure the smooth running of their farming operation. With a shortage of such people in the market, and farm workers understandably looking for improved remuneration and better working conditions, wage

13

costs are increasing as the dairy farm sector has to compete with other sectors that offer alternative employment opportunities. It seems reasonable to assume therefore that automation in the dairy farm sector will have to be used increasingly to offset the rise in farm labour.

3.2.2 Scarcity of labour Added to the rising cost of labour is the more pressing problem of scarcity of farm labour, which is driven by the availability of alternative employment opportunities for the rural population. These alternatives appeal to a growing number of the rural population who are attracted to a generally less harsh urban working environment with more sociable hours of work and what are perceived by many as more attractive working conditions and better rates of pay. In other words, the pool of employed labour that is prepared to get up at the crack of dawn seven days a week in all winds and weathers to milk cows is now in decline.

3.2.3 Family labour Faced with both scarcity and rising cost of employed labour family farms, which are the norm in many countries, particularly the high labour cost countries, find there is growing pressure on family members to provide more and more of the on-farm labour. On many farms this has started to have a serious negative impact on the quality of family life. Consequently lifestyle considerations have become a strong driver for automation on some farms. Farmers are now putting a much higher economic value on lifestyle, particularly on the availability of more free time for their family and this is being used more and more to justify capital investment in labour saving technologies. The decision to invest in automation is therefore not always a purely economic one.

3.2.4 Automation In addition to providing a substitute for farm labour automation has the potential to deliver a range of significant additional benefits that can improve farm productivity and profitability, including significant improvements in process control and better quality control. In short, rather than having to work harder it enables farmers to ‘work smarter’. Automation provides scope, for example, to improve process control where it can replace repetitive manual processes that farm labour is unable to apply with consistency. An example is automated teat spraying in automatic milking systems. Automating this process within controlled parameters yields much greater repeatability and consistency than manual teat dipping. Udder health benefits significantly from this improved process control, which reduces veterinary costs and increases milk yield. Automation of body condition scoring or locomotion scoring are also current areas of research that could improve the consistency of these manual and intuitive processes. Automation also provides scope for improved quality control. For example, sensor technologies are now being implemented in automatic milking systems to detect and divert abnormal milk, preventing it from entering the food chain. The precision of these optical technologies far exceeds the ability of visual inspection of milk to deliver the same levels of sensitivity and specificity. Milk quality improves as a consequence.

14

3.2.5 Management Systems integration Of even greater economic value to the farmer is the potential to leverage these various automation technologies by integrating them through on-farm process control and herd management information systems. Integrated herd management systems are now widely used on many farms in Europe and the US, which are capable of integrating a wide range of process control functions, such as animal identification and selection, feeding, weighing, recording of milk yield, fertility monitoring and health monitoring. These various functions provide a valuable source of information that can be combined to build powerful on-farm databases the utility of which can then be significantly leveraged by driving the process control systems and automating and integrating the functions of production management, feed management, fertility management and health management.

3.2.6 Connectivity The communications revolution is also spreading to agriculture, which means that the use of farm information systems now extends beyond on–farm applications to embrace a growing number of off-farm applications. In addition to data being held on on-farm information systems for farm management purposes, an increasing amount of data is also being held or shared on off-farm databases using browser and other Application Service Provider (ASP) technologies to communicate between the two. In addition to providing information and decision support to farmers these off farm databases also enable other relevant stakeholders to log on and access or upload data, thus providing access, for example, to veterinarians and feed consultants to assist them in their consultancy and decision support roles. Examples include NMR’s Interherd herd management information system in the UK and Canada’s DairyCheq system, which is a new web-based method of monitoring and managing bulk milk tank data. In some instances data processing and data management specialists collect this data electronically to provide specialist information management services to farms and breed societies, such as the herd improvement organisations like CR Delta in the Netherlands. In other instances regulatory bodies maintain the database for compliance purposes, which farmers can access on-line, often using software links which avoid duplication of data entry. Food traceability issues are also driving the trend towards greater connectivity and better sharing of data between stakeholders in the food chain. This was confirmed in March 2002 when the Food Standards Agency (FSA) in the UK published a study entitled ‘Traceability in the Food Chain’. This identified the need for what it described as ‘robust mechanisms to facilitate the collection and authentification of information, to enable it to be updated and shared through the food chain’. Their vision was of a food chain traceability system characterized by ‘a backbone of connectivity between robust databases with lateral connections to national/local users. Subsequent initiatives include the Livestock Data Programme, which merges the British Cattle Movement Service with the Rural Payments Agency and the development of their new RADAR veterinary surveillance database.

15

3.3 Genetic improvement Genetic improvement in dairy farming has been a fundamental driver of change for a long time and still remains so today. Indeed, artificial selection of livestock for genetic improvement purposes has been responsible for significant improvement in livestock dairy productivity. Much of the improvement in dairy cow yields over the past 25 years has been as a result of genetic improvement. This has taken milk yields averaging only 3,600 kg milk per cow per annum in the EU in the early 70s to the existence of high genetic cows that are capable of producing more than 12,000 kg of milk in a 305 day lactation period. The foundation of all genetic improvement programs is the accurate collection of animal based information, including performance data, and pedigree data. Most livestock producing countries therefore operate livestock improvement schemes which manage large animal databases. Data recording includes details such as calving date, ease of calving, insemination details, recorded weights throughout the life of the animal, milk yield performance, health events, pregnancy diagnosis, weaning details, culling information and disposal or death details. Data is generally held in a national database to which farmers and other data providers have access for data input purposes. Livestock improvement organisations and other relevant stakeholders, such as Herd Societies and semen companies then access this data for their work in genetic improvement. In some countries, such as Ireland, the animal breeding database (Cattle Breeding Database) is also linked to the animal movement database (National identification and Registration Database), which offers the farmer the advantage of single data entry. Livestock Improvement Corporation (LIC) operates one of the largest bovine databases in the world with records of more than 16 million animals. It considers this knowledge base to be one of the primary sources of New Zealand’s competitive advantage in dairy production. Herd recording is recognised as one of the significant activities for increasing the efficiency of dairy herds though improved genetics and better dairy farm management. In addition to providing the breed societies with essential data for their genetic improvement work it also provides the farmer with valuable data on milk yield and milk quality for herd management purposes. The livestock improvement organisations in all the major dairying countries, such as LIC in New Zealand, CR Delta in the Netherlands, NMR in the UK and the National DHIAs in the US, operate herd improvement schemes that follow ICAR guidelines for collecting data on milk yield and milk composition. This is based on a system of monthly recording on farm of milk yields and the collection of milk samples from each cow for laboratory analysis of the content in terms of such components as fat, protein and lactose. The systems for collection of data vary from country to country in that some organisations provide the milk meters for milk recording and others require the farmer to invest in his own equipment. Some countries have a predominance of technicians going on farm to collect the milk samples, whilst other countries operate DIY schemes that reduce the cost to the farmer. The data collected form the basis for the compilation of lactation certificates that record the genetic performance of individual cows for farmers to monitor the performance of

16

their herd and value their animals for resale. These data are also of particular value to the breed societies and semen companies that rely on these data for the building of their databases that support their work in genetic improvement and the compilation of breeding indexes. In addition to herd recording data the livestock improvement organisations also provide valuable herd management data to dairy farmers. Of greatest value among these is information on somatic cell count for mastitis management purposes. Other milk components that they measure include urea (MUN), milk progesterone and bacterial analysis of milk samples. The advent of on-line technologies that can provide this information on-farm therefore represents a competitive threat to these organisations. Whilst numerous initiatives have been undertaken by herd recording organisations to reduce the labour intensive process of herd recording new electronic animal monitoring technologies offer substantial scope to engineer further cost out of this process.

17

4. Overview of Technology Electronic animal monitoring technologies consist of a number of different types of technology. It is important to recognize that whilst a few are discrete technologies that operate independently as stand alone applications, the majority are systemic technologies that have little utility on their own and derive their functionality from being combined with other technologies to form fully integrated systems.

4.1 Discrete and systemic technologies Discrete technologies are capable of operating as a ‘closed system’ and are not dependent upon integration with other technologies to give them their functionality and utility. They are thus capable of adding value as stand alone applications. Systemic technologies, by contrast, can only operate as part of a larger system or can leverage their performance significantly by means of integration with other technologies. They are thus dependent upon systems integration to add functionality and utility. However, most hi-tech animal monitoring technologies fall into the category of systemic technologies in that their utility is dependent upon them being exploited as part of a larger system. Examples of systemic technologies include on-line milk sensing technologies that rely on milking machine process control to control the operating cycle, automated animal weighing systems that rely on EID and herd management software for data management and data interpretation purposes. One of the major problems associated with systemic technologies is that they can only diffuse as fast as their slowest diffusing component, which means that systemic technologies do not diffuse as fast as discrete technologies. For example the market for integrated herd management systems is dependent upon the market for milking systems. A fundamental technology common to all these systems, is electronic identification (EID), which forms the ‘glue’ that binds all of these component parts together.

4.2 Electronic identification EID was first introduced to the livestock sector in the 1970s for the purpose of dairy automation and is at the heart of all systems of livestock automation in that it links an identity to individual animals. It found a substantial market in the 1980s for use in conjunction with computer-controlled concentrate feed dispensers, or out-of-parlour feeders. These early systems were all based on relatively expensive collar transponders. Since then the technology has evolved significantly and there is a wide range of so-called Automatic Identification and Data Capture technologies (AIDC) available. The various formats available include ear tags, boluses, subcutaneous implants, neck transponders and ankle straps. The amount of data held by these devices has also evolved over the years. In its simplest form the data may be a simple numeric or alphanumeric string in read-only format which gives access to data stored elsewhere on a system such as a computer or embedded software application. At the other extreme the amount of information that can be carried within the identification system has expanded such that electronic

18

identification systems are now capable of storing in excess of 64K of information. As data storage capabilities in the computer world improve and costs come down so the capacity continues to grow. This means that, in theory, full animal ancestry histories can now be stored in an animal tag. Originally all EID products were passive, which meant that they only had the ability to store fixed data and had limited power that was derived from the magnetic field generated by the reader that captured this data. However, the development of active EID products, normally powered by an internal battery, has lead to the development of read/write devices, allowing information to be updated and tags to operate in essence as a mobile database.

The EID technologies used in livestock farming are all based on Radio Frequency Identification (RFID), which is a form of non contact AIDC technology for which transfer of data from an ‘identifier’ to a ‘reader’ is achieved by a radio-frequency link. RFID is highly efficient in hostile environments like agriculture where alternatives like barcode labels are not durable. It also has the advantage of being able to track moving objects, which has meant that RFID has established itself as the leading technology in the livestock world in addition to a wide range of other markets including automated vehicle identification (AVI) and cargo identification.

A basic RFID system consists of three components: an antenna, a reader and a tag or chip that is electronically programmed with fixed (read only) or variable (read/write) data. When the transponder comes into the magnetic field of the antenna it is activated and sends a radio signal with a preset frequency together with its unique identification code to the reader. The reader decodes the message, or data stored in the tag and transmits it to any data processing units like a PC or a process control device. The market for EID tags can be divided into two distinct segments which serve different applications: management tags and registration tags.

4.2.1 Management tags Electronic management tags are used on farm to monitor, feed and automate day to day management tasks on the farm. They operate in conjunction with such equipment as out of parlour feeders, calf feeders, electronic milk meters, sort gates and weighing devices and purely serve an internal on-farm herd management function. Management tags come in a variety of formats including ear tags, neck collar transponders, boluses, and ankle straps. One important difference between management tags and registration tags is that management tags normally need to identify animals in a moving position, such as when they come into a feeding station or milking parlour. By contrast registration tagging primarily requires ID systems and readers that are capable of reading tags from stationary animals such as animals restrained in a truck or standing in a sale yard.

4.2.2 Registration tags Registration tags are used to permanently identify an animal over its entire life for regulatory purposes and are subject to a rigid regulatory environment set by each country. In particular they have to be tamper proof and a number of manufacturers have developed patented systems to fulfil this requirement. They also have to comply with national standards in terms of official identification numbers, markings, size and colour etc. to identify them as registration tags. They are used to track ownership and transfers,

19

track and monitor disease, and prove identity and progeny for genetic improvement programs. In recent years progressive introduction of legislation in numerous countries around the world has made the use of registration tags a legal requirement. In the EU there is a requirement for double tagging of livestock to minimise the risk of loss of identity. This means that registration tags are composed of two groups. There is a Primary tag, which is normally a relatively large double ear tag which has all the relevant regulatory markings and an official identification number. Historically these have been predominantly visual tags, but there is now progressive use of electronic identification. The Secondary tags are a smaller tag which is used as insurance in case the primary tag is lost. Sometimes the management tag can effectively ‘double’ as the secondary registration tag. Most countries have an approved list of Registration tag suppliers which is issued by the competent regulatory authority and confirms the suppliers whose products comply with the relevant regulations. Some countries also have approval lists for ancillary equipment like readers and antennae. Some approval lists are drawn up on the basis of compliance with ISO standards, whilst others, such as Australia, for example, relate to compliance with technology standards as their system has adopted a sole HDX standard. Some authorities evaluate equipment and issue an approval list, as is the case for the UK British Cattle Movement Service which issues an approval list of officially tried and tested tags. In addition to the prime market for new tags there is also a sizeable replacement market for manufacturers to service which is also subject to a stringent regulatory regime.

4.2.3 FDX and HDX In the wake of the introduction of EID in agriculture ISO was asked in the early 90s to develop a standard for electronic tags. A working group was formed by the Agricultural Subcommittee of the Electronics Technical committee and three companies were considered in these trials: Destron (now Digital Angel), Nedap and Texas Instruments.

ISO categorised the basic requirements of tags into 6 groups: uniqueness, permanence, reading frequency, read range, mounting location, and electronics. These can vary depending upon the application. Tag size frequently affects read range, with bigger tags having longer range. Longer range generally increases the cost of the tag. However, there are applications where short range is not a drawback.

The working group opted to adopt a dual technology approach that combined the NEDAP Full Duplex (FDX-B) and Texas Instruments Half Duplex (HDX) technologies. This was intended to encourage competition and avoid a monopoly, although it also took account of the fact that both technologies have advantages and disadvantages. However, the technologies had enough in common that a single reader was capable of reading both and there is also an ISO standard for readers. The EID standard was finally approved after 4 years in 1996.

Subsequently Texas Instruments HDX technology, marketed as the TIRIS system, has more or less become the world standard for EID, particularly for regulatory EID as it has a superior read range. However, FDX-B is still quite common in management tag

20

applications, where high levels of accuracies in single reads are not as critical as in registration applications.

Nedap’s patent on FDX-B technology expired some years ago, but the patent on HDX technology held by Texas Instruments does not expire for some years yet.

4.2.4 Ear Tags EID tags come in a variety of different formats. The neck transponder was the first technology to market. Whilst it has the advantage of a high read range- typically 60-80cm, cost has been an obstacle and the market has shifted progressively to ear tags, which are significantly cheaper. They have a read range typically of 40-60 cm. This trend has been resisted by some OEMs, partly because they claim that their accuracy and read range is inferior to neck collar transponders, but partly also because the sale of the more expensive neck collar transponders has been a lucrative source of income for them. A typical example is an Allflex electronic ear tag, with a transponder containing a unique electronic number integrated into a plastic housing. Allflex claim to be the number 1 in animal ear tags, producing some 500 million tags in the year 2000. (Source: Derwent Intellectual Property Services). The electronic ear tag is a smaller but fast growing market. Research suggests that Allflex currently sells some 400,000 electronic tags per year in the UK alone and four million in Australia (Source: Derwent Intellectual Property Services).

4.2.5 Bolus technology The bolus is an ingestible device comprising a capsule containing an electronic transmitter, which is an alternative to ear tag technology. The bolus sits in the rumen. Some are also able to provide temperature data in response to electromagnetic interrogation, which can provide useful animal health information. One advantage of the bolus is that it has been found to be the most fraudproof method of EID. This was confirmed in a recent European funded trial called IDEA (Identification Eléctronique des Animaux) which was a project undertaken to test electronic identification systems in practical farming situations, reporting its findings to the EU Commission at the end of 2001. These trials also demonstrated that the bolus had the best retention rate in trials conducted in a number of EU countries. However, they have proved less popular with abattoirs owing to problems in recovering the bolus post-slaughter. They are also less popular with farmers who tend to prefer an ear tag which combines visual identification with EID.

4.2.6 Subcutaneous implants In the late 1980s technology was perfected to produce EID chips using integrated circuit technology that reduced their size to such an extent that they could be implanted in the animal by means of a subcutaneous implant that sits just under the skin. It is implanted by a vet using a syringe, and can contain both a chip and power supply. Implants have undergone extensive trials to prove their lack of migration into other areas of the body, and were also one of the devices involved in the EU IDEA project evaluating the feasibility of EID in livestock. Whilst commonly used in the domestic pet market they are not widely used in the livestock industry due mostly to their very short read ranges.

21

4.2.7 New technology developments Possible new technologies under development include identification of the vascular pattern of the retina which is unique to each animal. An image of this pattern can be captured using a suitably configured digital camera. Iris scans are also unique and could be used for identification. It is claimed that such technology may not be more costly than implementing RFID and has the advantage that it is permanent, secure and, obviously, tamper-proof. However, achievement of commercial systems is hindered by the difficulty in making these systems fit in a profitable business model for the technology supplier, since there is no ‘per animal’ cost involved. Other possible new technologies include DNA analysis and chemical signatures. These technologies are in the early stages of development and appear to be very high cost at present. One example includes an electronic ear tag that is designed in such a way that a sample of the animal’s ear is taken for DNA purposes by the ear tag applicator and stored in a bar coded vial that matches the ear tag.

4.3 Process control and information management systems In addition to EID the other common feature of all electronic animal monitoring systems is some sort of process control or information management technology that links the EID and various other technologies together and thereby leverages them to lend them their full functionality. The origins of process control in livestock farming date back to the early 80s when automated feeding systems were pioneered and introduced to the market by companies like Nedap and Hoko Farm (now Insentec) in the Netherlands. These early systems were ‘stand-alone’ discrete technologies consisting of an EID system and an embedded software system that contained the necessary basic process control to drive automatic feed dispensers. However, the obvious synergy with automating functions in the milking parlour meant that the technology soon extended to the milking system to automate gates for applications such as cow traffic management. In particular, the development of the electronic milk meter in the 80s and the ability to link the two via EID brought about a paradigm shift in the industry and heralded the advent of substantial growth in dairy automation. The ability to assign a cow ID to a milk meter and then to collect and transmit milk yield data to a feeding system meant that there was now considerable synergy between milking system automation and herd management automation: on the one hand the electronic milk meter provided a source of valuable data for automated herd management purposes and on the other, animal data could be used to automate the milking process. Milking machine manufacturers identified a strategic need to add process control and information management systems to their product lines in order to enhance the performance of their milking systems. They therefore collaborated with companies like Nedap and Afikim to integrate the two, lacking the relevant expertise themselves. They had identified the potential of these technologies to generate substantial incremental income by adding value and functionality to their equipment as well as the potential for it to differentiate their product in the market. Ultimately this led to the decision by most OEMs to bring herd management system competences in-house rather than outsourcing them. They saw these capabilities as a core competence that was strategically too pivotal to their future to outsource.

22

As a means of differentiation and potential source of competitive advantage considerable investment has been made in the development of integrated herd management systems over the last 15 years, both in terms of adding additional process control to automate more and more manual operations and also in terms of adding additional information management functions that present the farmer with relevant herd management data and assist him in the herd management decision process. Process control automation includes applications such as automatic drafting, automatic weighing, automatic sampling, automatic diversion of contaminated or abnormal milk and ultimately automatic milking in the case of robotic milking. Information management functions include feeding management, fertility management, health management, milk quality management and milking management. The synergies between the various applications can be exploited to create fully integrated knowledge-based management systems. In these systems, information collected by the various sensors and devices for data management functions is then used to drive the process control system, automating some of the decision process. For example, automatically diverting cows in need of a particular treatment, or blocking the milking process for cows that have been treated with antibiotics. Process control also provides the integration of the various component parts of the system. Systems integration is the ‘glue’ that binds all the component parts together and determines the route to market. This step in the value chain adds significant value by lending utility and functionality to the components that, as individual components, add little or no value. The value of the component parts of the system is leveraged by virtue of their integration into a functional system. Central to this function is both process control which drives the whole system and data management, which collects data from the various peripherals and interprets it to present the farmer with useful information. The systems integrator is, in essence, the solutions provider. As already identified its strategic importance is such that all the major OEMs now have systems integration capabilities as an in-house core competence. Control of the data protocols gives overall control of the system and locks customers into a single supplier for system upgrades or enhancements. Systems integration is also a critical factor in the route to market. Various initiatives are in hand to challenge the closed nature of systems integration and open up the market to more choice and greater connectivity between applications. ISO is a key stakeholder in this process. For the moment, however, the route to market is dependent upon the support of a systems integrator. Whilst the focus of these systems has been on on-farm process control there has been a significant parallel development of alternative and complementary herd management systems which combine on-farm and off-farm applications, which have no process control, but are dedicated to information management and decision support. These dedicated herd management systems partly complement the process control technologies by offering additional data processing and herd management functions that the process control systems are unable to provide, including substantial data archiving, family histories and veterinary or medical records. It is important to note that these complementary systems are dependent upon the on-farm process control and information management systems to access relevant on-farm information.

23

In particular, these ‘dedicated’ herd management systems have the advantage of external connectivity. They include both on-farm and off-farm databases and provide scope for access to data by a variety of stakeholders such veterinarians, feed consultants and regulatory authorities. Examples include NMR’s Interherd, Uniform Agri in Europe and Dairy Comp 305 in the US market.

4.4 Milking systems Milking is a central function in the herd management process that extends beyond the milking process itself. The milking system also represents a central location for a number of herd management functions and for automated systems it often represents a critical point for data interchange between the animal and the management system. It therefore represents a critical component in electronic animal monitoring systems. The level of automation in milking systems and the degree of process control varies significantly but the systems fall into two broad categories: conventional milking systems and robotic or automatic milking systems.

4.4.1Conventional milking systems Conventional milking systems range from bucket milking systems which have the lowest level of milking mechanisation and milk one cow at a time, at one end of the spectrum, to highly automated rotary parlours that are capable of milking up to 60 or even 80 cows simultaneously, at the other end of the spectrum. In between there are a variety of milking systems that vary both in terms of configuration and automation that respond to the needs of the type of dairy farming system practiced.

4.3.2 Automation of conventional parlours A range of automation technologies are available for milking systems. The most automated milking systems also have a process control and information management system that integrates these technologies.

4.3.2.1 Automatic cluster removers The most significant development in automation was the advent of automatic cluster removers in the 80s. These devices monitor milk flow and automatically remove the milking cluster at the end of milking when pre-defined milk flow conditions are met. They reduce the workload on milkers and can increase throughput. They also avoid overmilking, which can be harmful to udder health. They are now common on most larger installations.

4.3.2.2 Automatic drafting Automatic drafting is used to separate individual cows form the herd for specific handling, such as insemination or veterinary treatments. It can also be used to organise cows into groups for feeding etc. It is not yet widely used in many markets, but its use is growing because of the labour savings and efficiency gains that can be achieved.

4.3.2.3 Automatic teat spraying Post-milking disinfection of teats is one of the most effective and common tools for prevention of teat infections which lead to mastitis. Teat dipping and teat spraying are

24

two common alternative manual technologies and teat spraying is automated in some parlours, although the efficiency of these systems is limited.

4.3.3 Automatic milking systems The Automatic Milking System (AMS) market is a relatively new market that emerged in the late 90s. The term Automatic Milking System refers to those milking systems that automate the functions of the milking process and achieve a high degree of automated cow management. The key feature of a milking robot is that it allows cows to walk into the milking box 24 hours a day and be milked without supervision. This transforms cow milking from a normally twice-a-day batch process operated by the farmer into a fully automated continuous process. All the current AMSs manage the milking process by milking each quarter separately, which facilitates the management of the milking process on a quarter by quarter basis rather than by reference to composite udder milk. This is particularly useful for udder health management purposes, where it is helpful to be able to milk each quarter individually and monitor and compare the health status of each individual quarter. The market is split between single box systems such as Lely and multi-box systems such as Galaxy and Prolion. Single box systems have one robotic arm per milking stall and have a capacity of about 60 cows under European conditions. Under European conditions cows are milked on average 2.7 times a day which means that a robot has a capacity of about 180 cow milkings per day. The fact that milking is spread over 24 hours makes for efficient use of the capital equipment. Single robot installations are common on some smaller European family farms with 60 cows, but multi-robot installations are also common for larger herds, especially in markets like Denmark. The US has the largest installation in California with 32 robots operating on one farm. Multi-box systems have one robotic arm for two or more milking points. They are designed for larger herds in order to reduce the capital cost, but have the disadvantage of reducing the number of cow milkings per milking point. Since the first commercial systems appeared on the market in 1992 the number of AMS installations has increased at a steady rate, albeit with lower rates of growth than anticipated in the last couple of years. The total number of milking robots is estimated at the end of 2003 to total some 3,000 machines (Source: PV Lelystad, 2004), which are in use on more than 2,200 farms located in over 20 countries around the world. Based on the figures for 2003 the annual market for AM systems is likely to be of the order of 500 farms per annum, which represents the equivalent of about 750 machines per annum. AMSs have evolved primarily in response to the demand for automation in the high labour cost countries of Europe. Their implementation has been based therefore on adapting the farming systems commonly practiced in Europe for the high yield and very high yield herds described in sections 2.2 and 2.3. The work being undertaken in the Greenfield Project in New Zealand provides positive evidence of the ability of AMS technology to be used as part of a total automated dairy farm system that responds to the needs of New Zealand’s pasture-based dairying environment, whereby the advantages of a pasture based approach to milk production are not compromised by implementation of automated milking.

25

4.4 Feeding systems Feeding was one of the first areas of livestock farming into which automation was introduced in the 80s when companies like Nedap introduced the first automated feeding systems. As one of the high component costs in production these technologies are designed to optimise feeding in order to maximise profitability. From early DOS based embedded software systems capable of dispensing simple feed rations modern feeding systems are equipped with powerful data processing functions that are capable of handling multiple feed types dispensed in multiple locations that balance the energy, protein and mineral requirements of individual animals in order to maximise the performance of the animal. Integrated systems use data such as weight gain, milk yield, milk composition and lactation stage to adjust rations to the individual requirements of the animal. Whilst supplementary feeding is not commonly practiced in New Zealand it is interesting to note the growth in automated feeding in Australia’s pasture based dairy farms. It may also grow in popularity in New Zealand as part of the quest to increase milk solids output. Milk production is determined by the genetic potential of the cow, her nutrition and her state of health. Of these three factors, nutrition is the most important for two reasons: it is within the direct control of the farmer and it has a profound influence on production in a healthy animal. Feeding management is therefore a critical herd management function. In pasture based dairy farm systems in countries like New Zealand and Australia, grass is the major part of the feed ration and often accounts for almost 100% of the feed ration. In these systems grass management is therefore the prime feed management concern. In farm systems with higher yielding cows grass feeding is supplemented with corn, silage and concentrates. The input of concentrate per kg of milk varies significantly in relation to farm production system as the following IFCN statistics show: Concentrate per kg of milk Country >440g US, Finland, Hungary, Spain, Israel, India 250g – 400 g Austria, Germany, Denmark, Sweden,

Czech Republic, Colombia, Argentina 100g - 250g Netherlands, France, United Kingdom,

Poland, Argentina, Brazil, Australia, Estonia

<100g Switzerland, New Zealand Source: IFCN (2002) The higher the level of concentrate fed the greater the incentive to optimize feeding to maximize productivity. On many farms concentrate feed costs represent the largest variable cost, accounting for up to 33% of total variable and fixed costs of milk production. Computer controlled feeders offer substantial benefits because they enable producers to:

• Feed adequate concentrates to selected cows during early lactation in order to achieve a higher peak daily yield and to have the cows maintain a high daily milk yield for a longer period of time.

26

• Improve digestive metabolism by spreading concentrate consumption through

the day for cows receiving large allowances of concentrates.

• Simplify the reduction of feed cost at the end of lactation by limiting the amount of concentrates allocated to cows when their daily milk yield declines.

• For the future, enable farmers to feed supplements that could either selectively

increase high-value desirable components or selectively decrease undesirable components in milk.

Feeding systems fall into two main categories of in-parlour and out of parlour feeding systems. In–parlour feeding systems are essentially used for supplementary feeding purposes and as an attractant to encourage cows into the milking parlour. Out of parlour feeding systems are accessible to cows throughout the day and night, unless they are being pasture fed. Combining pasture feeding with concentrate feeding is therefore a compromise in terms of access to low cost feed in the form of grass and optimised feeding of concentrate.

4.5 Fertility management One of major costs affecting dairy farmers is poor fertility, particularly since there has been a significant decrease in the fertility of dairy cows over the last twenty years. As the modern cow’s milk yield has increased, she has struggled to meet the metabolic and nutritional demands of her increased milk production. This means that poor reproductive performance has become one of the most costly and difficult problems for dairy producers. Consequently the market for herd fertility management products is vast. An indication of the potential market can be derived from the number of cattle in the developed world using AI which exceeds 100 million (FAO, 2000). Consequently considerable investment has been made in researching and developing technologies to improve fertility management. In countries like New Zealand with seasonal production it is essential for spring calving cows to get back in calf as quickly as possible if a tight calving pattern is to be maintained to maximise profitability. More than 90% of New Zealand dairy farms operate under a pasture based seasonal calving system. This requires the herd to calve every 365 days during a relatively short period, to match animal feed requirements with the availability of pasture. For these reasons a total calving spread of 8-10 weeks is favoured. It has been calculated that the cost of a delay of two months in attaining pregnancy costs the average New Zealand farm about NZ$13,000 which amounts to a total cost for the New Zealand dairy industry of approximately NZ$175 million per annum. The cost of sub-optimal fertility is even higher in other parts of the world, especially the United States where fertility levels have declined rapidly in the past 20 years. There was a particularly dramatic decline in reproductive performance between 1985 and 2000, both in terms of average number of days open (interval from calving to next conception), which were up from 130 in 1985 to over 160 in 2000, and in the number of services per conception, which increased substantially from 1.5 to over 3 over the same period.

27

Better oestrus detection and improved fertility management technologies provide significant potential to reverse the decline in reproductive performance and thus to save dairy farmers significant costs by reducing their average number of days open, whilst at the same time reducing the number of inseminations required to get cows back into calf.

4.5.1 Pedometers Early technologies were based on automating the traditional technique for detecting oestrus in cattle by observing changes in behaviour, the standard indicator of oestrus being the willingness of a cow to be mounted, attempting to mount other cows, restlessness and substantial increase in activity. Visual observation is time consuming and impractical for the larger farms, making oestrus detection very difficult. Several mechanical pedometer systems have been developed in an attempt to automate this process of activity monitoring. Systems typically comprise an automatic activity recording tag attached to the cow’s leg or neck, originally based on a mercury switch which records each step based on the movement of the leg. More recent systems avoid the use of mercury or are based on an accelerometer, which has greater sensitivity and can cope better with changes of environment such as mixed housing and grazing systems. An antenna receives data automatically, usually at milking, and transmits the data to interpretative software that is part of the herd management system. Some more recent systems have telemetric data transmitting capability to allow more frequent data transfer to cater for pasture based grazing and to improve the level of data analysis and others integrate this information with other herd management system data to distinguish true oestrus from increased activity as a result of other causes. However, all these systems have limitations in terms of predictive capability.

4.5.2 Progesterone sensing An alternative approach has been the development of tests capable of detecting the progesterone content of milk. Progesterone is a female sex hormone produced by the cow after ovulation that can be detected in milk. It is released into the blood by the corpus luteum that is formed on the ovary after the follicle has ovulated and is subsequently secreted into milk where it can be detected, particularly as it has an affinity for milk fat. Progesterone rises and falls during different periods of the cow’s reproductive cycle. For the majority of the cow’s cycle progesterone is at higher levels, but drops a few days before the cow comes into oestrus and starts bulling. The progesterone is at its lowest when the cow is actually bulling and remains low for around 2-4 days, after which it starts to rise again. Monitoring oestrus levels in dairy cows has three main benefits in fertility management:

• Advanced prediction of bulling -it is possible to predict when the next oestrus is likely and thus when it is best to inseminate.

• Early detection of pregnancy - Progesterone levels can be monitored after

insemination to determine pregnancy status.

28

• Early diagnosis of fertility problems - It is possible to detect abnormalities in

progesterone patterns and thus aid vets to identify fertility problems (such as follicular cysts, problems with the corpus luteum or anoestrous conditions) and then determine the right fertility treatment.

A number of manual tests are available which are based on Enzyme Linked Immunoassay (ELISA) technologies. These require the taking of a milk sample at intervals after calving and generally involve sending the samples away for laboratory analysis or the use of farm test kits. Interpretation of the progesterone levels is available in the form of a number of computer programs including MOIRA (Management of Insemination through Routine Analysis) developed by the University of Reading and Promise Fertility, developed by the University of Nottingham. A number of organisations are also working on automated on-line technologies capable of measuring milk progesterone. Some are based on automated sampling coupled with a centralised tester, whilst others are based on on-line sensor technologies located at each milking point. A system has also been developed in New Zealand that uses telemetry to control drug delivery to control oestrus in cattle. The Delivery and Monitoring Unit (DMU) uses several technologies including activity, temperature and detection of grouping of animals to detect oestrus. This information is then used to automate the release of prostaglandin from an intravaginal bolus. This device builds on the CIDR (Controlled Intravaginal Drug Release) technology. The CIDR is an intravaginal device which contains silicone impregnated with progesterone, which is released in low doses into the cow over a period of days. Used in association with other injected reproductive compounds it will induce the onset of heat in non-cycling cows. The CIDR is used early in the breeding cycle to assist in bringing non cycling cows into heat. It is also used in subsequent cycles of the season to stimulate oestrus in cows that have failed to become pregnant.

4.6 Mastitis management Mastitis is acknowledged as the most common and most costly disease encountered in the dairy industry. The National Mastitis Council (NMC) estimates that annual losses to the US dairy industry amount to between US$ 1.5 and 3 billion per annum or 11% of the total USA milk production. Losses in the UK are thought to be in the region of £80 million per annum (Source: Dairy Farmer, 2002) and in New Zealand comparable losses have been estimated at some NZ$200 million. There is therefore high demand for mastitis detection technologies. 70% of this cost of mastitis has been found to result from reduced milk yield as a consequence of the infection. A high proportion of this lost milk yield is attributed to sub-clinical mastitis, which has been found to be between 15 and 40 times more prevalent in herds than the clinical form. For example, a high cell count of 400,000 cells per ml has been found to result in a 5-8% loss in milk volume. Sub-clinical mastitis is also significant because it constitutes a reservoir of organisms that can often prompt infection of other animals in the herd. There is therefore a strong demand for technologies capable of earlier detection of the infection to assist farmers with the herd management process by enabling them to make better informed decisions about treatment of their cows.

29

It should be noted that, although there is a demand for technologies capable of earlier detection of mastitis, treatment options pose a problem for the farmer since the cow is capable of self-healing at any time. One of the benefits of on-line technologies, however, is the ability to collect data at each milking and thereby monitor mastitis as a process rather than simply looking at the condition of the udder at a single point in time. Monitoring the progression of the condition allows the farmer to ascertain whether the cow is capable of self-healing or in need of veterinary intervention.

4.6.1 Conductivity measurement Early systems focused on conductivity as an indicator of mastitis, based on the fact that conductivity in mastitic milk rises at the stage of clinical mastitis when the blood milk barrier in the udder is breached and sodium and chloride ions pass from blood to milk. These early systems had poor predictive capabilities since they were based on measurement of the absolute conductivity of composite milk; subsequent research has established that milk conductivity is subject to a number of variables unrelated to mastitis such as temperature, breed, stage of lactation and diet. Later systems have therefore focused on measurement of relative conductivity between quarters and have algorithms that are designed to eliminate many of the non-mastitic variables, resulting in much improved predictive capabilities. Robotic milking systems lend themselves well to quarter conductivity sensing since each quarter is milked individually. Conductivity is a measure of tissue damage which occurs quite late in the mastitis infection process at the point when clinical symptoms are present. Other technologies have attempted to detect the infection earlier in its development. Research has therefore been taking place to devise alternative sensor technologies capable of detecting mastitis earlier in the process.

4.6.2 Somatic cell count Measurement of the somatic cell concentration of raw milk is the most widely accepted indicator of mastitis and milk quality. Whilst conductivity measurement is indicative of clinical mastitis infections, somatic cell counts have the advantage of giving an earlier indication of mastitis at the sub-clinical stage. Somatic cells are made up of epithelial cells from the secretory tissue of the udder and white blood cells (leukocytes) that are constantly circulating in the blood stream. The level of leukocytes will vary in response to the state of health of the cow. When an infection occurs, the body sends high numbers of these cells to the affected site. An increase in the number of somatic cells indicates, therefore, that the immune system is in a high state of alert and provides an early indication of a sub-clinical infection. Somatic cell counts can be obtained by collecting milk samples and sending them away for laboratory analysis, but this involves an undesirable delay in obtaining results during which time the infection may progress. An imprecise but cheap alternative farm test also exists in the form of the California Mastitis Test which involves mixing a milk sample with a reagent. An on-line SCC sensor based on the chemistry that underlies the California Mastitis Test has been recently released onto the New Zealand market. Using a cheap reagent it offers for the first time low cost on-line SCC measurement. A manual SCC counter for on-farm use was also released onto the market in 2004 but this uses relatively expensive disposable cartridges and has mainly been purchased by veterinarians to date.

30

Not only is somatic cell count an important marker for mastitis it is also the most common method used to measure milk quality. For example EU Directive 92/46/EEC defines milk quality by reference to parameters of Somatic Cell Count. Bulk milk SCC (BSCC) is used as the main criterion and is intended as an indirect control of the number of infected cows delivering milk for consumption. The EU currently enforces a maximum bulk milk SCC of 400,000 cells/ml while the U.S. limit is 750,000 cells/ml. In addition to the above regulatory framework milk buyers use BSCC as a measure of milk quality. They pay bonuses and levy penalties to encourage quality improvement, generally based on three month geometric mean SCCs. For example, one UK milk buyer specifies a bonus of 0.3ppl for an SCC less than 150,000 per ml and a penalty of 15ppl for an SCC greater than 400,000 per ml.

4.6.3 Milk Amyloid A Milk Amyloid A (MAA) is an acute phase protein present in milk that was first identified in 2001 by scientists at the University of Nebraska in collaboration with Tridelta, an Irish biotechnology company. Their research demonstrated that MAA could be useful as a diagnostic marker for the onset of mastitis from approximately 8 days after calving, and possibly as early as 3 days after calving, at the pre-clinical stage, which is earlier than SCC is capable of detecting. MAA tests are available as manual test kits for veterinarians and farmers, but are relatively expensive. Research is also being undertaken to develop an on-line biosensor capable of detecting raised levels of MAA during milking.

4.6.4 L-lactate Research undertaken by AgResearch in New Zealand has established that L-lactate (lactic acid) in milk is also a valuable early marker of mastitis. As a direct by-product of anaerobic metabolism, the presence of L-lactate in milk appears to be linked to the beginning of an infection. Bacteria entering the udder via the teat canal begin to multiply given optimal conditions for growth. This metabolic activity and subsequent immune response uses some of the available oxygen in the udder thereby inducing anaerobic conditions. The resulting lack of oxygen causes L-lactate to be produced that is proportional to the amount of metabolic activity. As the infection progresses, so the level of activity also increases leading to an elevation in the level of L-lactate. This technology has been patented and research is being undertaken to develop an on-line biosensor capable of measuring L-lactate levels in milk.

4.7 Milk Quality Management The demand from milk processors for higher quality milk and the imposition of penalties for milk that does not meet their milk quality criteria is also driving demand for technologies to monitor milk quality. Milk quality is measured using a variety of quality criteria that fall into three categories: contaminated milk, abnormal milk and undesirable milk.

4.7.1 Contaminated milk Contaminated milk is defined as milk that is unsaleable or unfit for human consumption following treatment of the animal with veterinary products e.g. antibiotics that have withholding requirements, or treatment with insecticides or pesticides not approved for

31

use on dairy animals by organisations such as the US Food and Drug Administration (FDA) Avoidance of antibiotic residues in milk is essential in the production of safe dairy products for human consumption since there is potential for allergic reaction in sensitive individuals. Also, increased resistance to antibiotics can develop from excessive exposure to these substances. This, in turn, renders them less effective when later prescribed by a physician. This has led to a growing body of legislation across the world to control the presence of these substances in milk. Legislation, such as the 'Maximum Residue Limits' regulations imposed by the EU since the 1990s, and the US 'Pasteurized Milk Ordinance' of 1991 require screening of all bulk milk for residues before entering the food chain. Combined with increasing consumer awareness, strong financial incentives have emerged for both farmers and processors to prevent milk contaminated with antibiotics from entering the food chain. Antibiotic contamination is also undesirable in that it can interfere with the processing of milk into manufactured products such as yoghurt and cheese. This financial obstacle reinforces the legislative pressure on processors to use only uncontaminated milk and has induced them to impose stringent financial penalties on dairy producers supplying contaminated milk. Dairy farmers have a strong financial incentive, therefore, to identify and divert any milk with antibiotic residues before it can enter the bulk milk tank, and certainly before it enters the tanker truck. Currently screening for antibiotic residues is carried out by milk processors using screening assays that have been developed to test milk from the tanker truck before it is transferred to the processing plant. There is a wide range of antibiotics, but the most prevalent are ß-lactam antibiotics. Currently all such tests are performed by hand. Some manual kits exist to enable farmers to check their milk for antibiotic contamination, but they are not widely used. Research is currently being conducted into biosensor technologies that can facilitate the rapid automated detection of antibiotic residues in milk both at bulk milk tank level, before it enters the processing plant, and at the point of milking so that it can be diverted before it even enters the bulk milk tank on the farm.

4.7.2 Abnormal milk Hygiene regulations in most countries stipulate that milk should be inspected for abnormalities before it enters the milk tank. For example EU Directive 89/362/EEC (1989) stipulates that ‘before the milking of the individual cow the milker must inspect the appearance of milk. If any physical abnormality is detected, milk from the cow must be withheld from delivery.’ The advent of robotic milking has created a demand for technologies capable of automating the detection of abnormal milk, since no one is generally present when cows are being milked in an AMS. A conference held in Denmark in 2002 addressed the issue of a scientific definition for abnormal milk and recommended a dual definition based on the criteria of colour and homogeneity. Colour was proposed as the criterion for identifying abnormal milk in respect of the presence of red blood cells. The presence of blood in milk is considered abnormal, making it unfit for human consumption, in the sense that it is indicative of damage to the secretory epithelia and of breaching of the blood-milk barrier which can cause leakage of blood-borne substances into the milk. The proposed reference method

32

for homogeneity was the presence of clots identifiable by filtration of the milk through a filter with a pore size of 0.1mm. Several colorimetric technologies are available for the detection of red blood cells in milk in robotic milking systems although they have variable levels of precision and research is being undertaken into technologies for automatic clot detection.

4.7.3 Undesirable milk Undesirable milk is defined as milk which, prior to the milking of the animal, is known to be unsuitable for human consumption. This includes the categories of colostrum and high somatic cell count milk, which is indicative of mastitis infection. Technologies for the detection of somatic cells have already been covered in section 4.6.2

4.7.3.1 Colostrum Some countries define colostrum by reference to calving, the most common being milk produced by a cow in the first 72 hours after calving. Integrated herd management systems are therefore capable of automatically diverting colostrum for the required withdrawal period. Other countries, including New Zealand define colostrum by reference to immunoglobulin (IgG) content rather than time span. This is because colostrum is undesirable for milk processors, its high immunoglobulin content stopping many of the processes used for milk processing like bacteria growth for yogurt and cheese manufacturing. The detection of colostrum is also desirable because some milk buyers collect colostrum separately to process the high-value components in milk such as Immunoglobulin and lactoferrin. Several colorimetric technologies are capable of detecting colostrum but with limited levels of precision to date.

4.8 Animal Health Management In recent years there has also been a focus on the development of technologies to automate some of the visual processes associated with animal health management. These include body condition scoring and locomotion scoring.

4.8.1 Body condition scoring Linear scoring of morphological characteristics or body condition scoring is an important source of animal health information for herd improvement and on farm herd health management applications. As a visual process undertaken by trained staff it is an expensive and time consuming business. The development of stereo vision technology, which is capable of monitoring a moving object is being used to develop technology capable of automating the body condition scoring process. Note that monitoring of body condition scoring is not synonomous with monitoring live weight. Rather, live weight is simply one of several parameters that could be monitored, all of which would be fed into algorithms to determine a body condition score.

33

4.8.2 Locomotion monitoring Lameness in cattle was estimated to cost the UK dairy industry in excess of £15 million per annum, according to research undertaken by the University of Reading in the late 80s. A more recent and comprehensive study of the costs of lameness in dairy herds in England and Wales was presented at the VIth International Symposium on Diseases of the Ruminant Digit in Liverpool in 1990. This estimated the total cost at £89.2 million per annum. The major cost items were identified as the costs of culling extra cows and the delay in conception. Lameness has assumed greater importance as cattle are managed more intensively. Lameness reduces performance in both beef and dairy cattle; it has a negative influence on feed intake and conversion, body weight, milk production, sexual activity, fertility, and in some instances, longevity. The incidence of mastitis has also been shown, in some cases, to be linked to a high incidence of lameness. In the US, the development of a step analysis technology capable of early detection of foot pain has been announced.

4.9 Overall Farm Management Efficiency gains can also be made in the area of overall farm management. Although the New Zealand farmer represents some of the world’s best in terms of cost-effective management of the overall farming system, gains in the areas of pasture management, infrastructure support and overall farm business management.

4.8.1 Pasture Management A number of opportunities exist to assist the farm in managing pasture growth. For example, parameters such as dry matter content, soil moisture content and nutrient monitoring can assist the user in optimising: water for areas of the country using irrigation, application of fertiliser and grass growth for planning pasture rotations.

4.8.2 Infrastructure Support Other areas on the farm could also be monitored and fed back to an integrated system. Power and water supplies could be continuously monitored, electrical gear on the farm could be managed to take advantage of low usage times of the day and technologies such as heat recovery mechanisms could be used to create a more energy-efficient farm.

4.8.2 Business Management Finally, the ultimate end-point of all of these on-farm technologies would be to seamlessly integrate them into an overall farm business management infrastructure. While this couldn’t ‘de-humanise’ the on-farm decisions, gains could be made by eliminating mistakes and providing the background for more fact-based decisions.

34

5. Conclusion In the rapidly changing global economy described in this report, within which New Zealand has to compete, the dairy farming sector faces new challenges which it will have to meet if it is to remain internationally competitive. Whilst genetic improvement and improved pasture management have provided the major source of New Zealand’s competitive advantage to date, incremental improvements in these technology areas are likely to prove increasingly elusive. For the future, therefore, New Zealand has to identify new sources of productivity and quality improvement, as well as to find novel ways of further increasing the added value component of its dairy output in order to sustain its competitive advantage. This trend points to an environmentally friendly intensification of dairy production through better use of technology and better management of the on-farm production processes. Electronic animal monitoring technologies and integration of on-farm systems and processes have a pivotal role to play in this process. Electronic animal monitoring and process control technologies developed for the high yield dairy systems of Europe and North America are relevant to New Zealand in that they form the basis for the development of customised process control and monitoring systems, which will be able to meet the specific requirements of New Zealand’s pasture based dairy environment and which will be capable of delivering the productivity and quality improvements needed to sustain New Zealand’s competitive advantage.

35

Appendices Appendix 6.1 PEST Analysis of electronic animal monitoring technologies market

Po

litic

al

Regulatory pressure Food traceability Access to export markets Bio-terrorism Trade liberalization

Eco

no

mic

Rising Labour cost Reducing agricultural commodity prices Rising food quality demands from buyers

So

cial

Scarcity of agricultural labour Farming lifestyle considerations Animal welfare Consumer awareness

Tec

hn

olo

gic

al

Automation Systems integration technologies Genetic improvement Communication technologies Connectivity