Managing the change to a robotic milking system

The automatic milking system is not a substitute for good husbandry or good management. It is essential that all normal preventative measures are maintained in respect of mastitis control cow health and reproduction.

Introduction

Moving cows to a new milking facility, whether conventional or robotic, can be stressful and disruptive to the performance of any herd of cows. Unless the appropriate steps are taken to accustom cows to the new facility and train personnel to use it correctly, the result may be impaired milking time management and aggravation of underlying subclinical mastitis problems.

Where the change is from milking in a conventional parlour to an automated system, the challenge of change is even greater. In addition to the potentially stressful period of training, for both cows and personnel, milking no longer occurs at predetermined times and regular intervals. Milking frequency depends on management strategies relating to housing and feeding together with their effects on the willingness of cows to visit the milking station.

Studies in Denmark and the Netherlands indicate that milk quality and mastitis levels tend to deteriorate, at least for a time after the introduction of a robotic milking system. The extent of the deterioration is often related to the overall effectiveness of management and mastitis control in the herd prior to conversion.

Automatic Milking Systems offer the possibility of more frequent milking. Although traditional wisdom suggests that this should be associated with an increase in milk yield, practical experience indicates that on some farms it is achieved, on others it is not.

Clearly, management is a major factor in determining whether or not introduction of automatic milking is successful. This paper discusses the need for an effective management strategy. Advice is given on ways of recognising the potential problems and minimising the effects.

Milk Quality And Udder Health

A number of studies in various European countries all indicate deterioration in milk quality following the introduction of automated milking systems (Lind et al. 2000).

A more recent study carried out in Denmark (Rasmussen M D et al, 2001) of a population of 69 farms confirms that, on average, somatic cell count increased in the first few months after introduction of automatic milking. After this initial deterioration, somatic cell counts decreased. Rasmussen could not provide a conclusive reason for the increase, but suggested that more attention should be paid to the introductory period.

A further study in the Netherlands (van der Vorst, 2002), of data from a total of 394 farms in three countries, also indicates deterioration in somatic cell and total bacterial counts, freezing point depression and free fatty acid levels. This study indicates that, for all four milk quality parameters, the deterioration is worst with the oldest automatic milking systems and seems to have become less marked with more recently manufactured machines. After an initial increase, both somatic cell and total bacterial counts decreased significantly.

All of the studies already mentioned indicate that udder health, as measured by somatic cell count and new infection rate, deteriorates in the period after introduction of automated milking. After a period of a few months, the situation usually improves.

It is worth noting that the above observations are based on the average results from the respective studies. Whilst there seems to be a general and lasting deterioration in respect of both freezing point depression and free fatty acid levels, there is a great deal of variation between the extent of variation shown in individual farm records of somatic cell and total bacterial counts. This suggests that the extent to which both udder health, somatic cell and total bacterial counts deteriorate, or whether they do so at all, is influenced by circumstances and management on the farm.

Milk Production

In most cases, farmers installing automatic milking systems expect to increase milk yield by increasing the frequency of milking. Increasing from twice to three times daily milking can be expected to result in a yield increase of 5 to 25% (Hillerton & Winter, 1992).

A number of studies suggest that the actual increase following introduction of automatic milking is often less than that expected. In Sweden, milk yield increased by 6.75% following the introduction of automatic milking (Svennerstein-Sjaunja et al. 2000). On 69 Danish farms (Rasmussen M D et al, 2001) an increase was reported of 2% from 23.9 kg to 24.4 kg per day. NRS data in the Netherlands (de Koning, 2002) reports an increase of 11.4% associated with the change to automatic milking. It should be noted that this latter figure is not corrected for any change in the genetic potential of the cows milked in the automatic system compared with the group milked previously.

In addition to the above, practical experience shows that there is a great deal of difference in the actual performance achieved on individual farms.

In general, yield increase associated with more frequent milking with automated milking systems seems to fall short of that predicted by earlier experiments. One possible explanation is that the controlled trials investigating yield increase associated with more frequent milking are all based on a regular milking interval. With automatic milking, milking interval can vary considerably for individual animals. The frequency of milking can also vary between animals. Animals attending for milking after very long intervals may actually show a decrease in yield.

Management

It is very clear from the foregoing that management on farms choosing automatic milking is a major determinant in the success of the system and the extent to which problems may be experienced during the transition period.

In addition to the changes of routine facing the cows, milkers and herd managers lose their regular opportunity for close contact with cows which occurs through twice or thrice daily milking. For many people, this means that they must learn to use the information provided by the robotic system. Instead of relying on visual perception of the cows welfare and condition, reliance must be placed on the production of adequate information by the system and its interpretation. In turn, a successful changeover is entirely dependent on milkers/herd managers making correct management decisions and taking the appropriate actions.

Automated milking systems, in their current state of development, cannot replace good husbandry and stockmanship. From the outset, it is important that users recognise that the system is unlikely to result in large labour savings, but merely changes the pattern and intensity of labour. To some extent manual labour is reduced, but replaced by more data handling and observation (Deutsche Tierärztliche Wochenschritt).

It is helpful to identify what the automated milking system can and cannot do in order to begin to shape the priorities for management and husbandry.

Automated Milking System Can:  Automated Milking System Cannot:
  • Admit cow to stall
  • Identify Cow
  • Decide whether cow is due to be milked, based on operator settings
  • Dispense Feed, as per operator settings
  • Clean teats
  • Attach teat cups
  • Remove teat cups when flow rate falls to pre-determined level
  • Spray post milking teat disinfection
  • Allow cow to leave stall
  • Reject milk, as per operator settings
  • Record milk yield
  • Raise alarm lists, as per operator settings
  • Page the operator in emergency, as per operator settings
  • Bring cows in to milking station
  • Clean up cows grossly soiled from housing
  • Distinguish teats from dirty hair on udder
  • Milk cows with unsuitable udder conformation
  • Determine the level of concentrate feeding for the cows
  • Determine the expected yield or milking frequency to decide when cow is due to be milked
  • Know whether or not the cow is completely milked
  • Treat sick cows
  • Call the veterinarian
  • Inseminate cows
  • Cure cows already infected with mastitis
  • Manage and deliver roughage feeding
  • Manage paddock grazing
  • Clean down the milking area
  • Order semen
  • Refill chemical containers
  • Order new supplies of feed and chemicals
  • Replace worn or damaged rubber components
  • Service itself
  • Become angry!

Having recognized the capabilities of the system, it is vital to begin preparations before the transition period. It should also be remembered that adrenaline release will inhibit milk let down. The focus of all preparations should be upon making the introductory period as stress free for both personnel and cows.

Preparation For The Transition – The Farm

The most obvious preparation is to ensure that building layout must be conducive to attracting cows to visit the automatic milker. Several studies (Lind et al. 2000) show that the most effective motivating force is feed. Whilst it is clear that there are numerous successful layouts, it is clear that consideration must be given as to how siting of the automatic milking system relative to concentrate and roughage feeding areas can be used to incentivise cow attendance.

It is not absolutely necessary for cows to receive concentrate feed in the automatic milking station. If the milking capacity of the machine is to be maximised, the most practical option may be to install a concentrate dispenser in the exit area (Prescott, 1996). To make the most effective use of this arrangement requires that control of the automatic milking station and the feed dispenser can be integrated in a single comprehensive herd management software package. This ensures that concentrates can be allocated, within the cow’s total ration, in either the automatic milking station or in the feed dispenser.

There are strong differences of opinion as to whether the most successful systems employ free access, where the cows can choose whether to attend the milking station, or forced access, where cows can only get to the main feed area via the milking station. The initial introductory period, and when new cows are added in the future are likely to be much easier when provision is made for forced access.

It may also be necessary to give consideration to the farm’s cooling system to ensure that milk is cooled as quickly as possible, but without risk of freezing the milk. There also needs to be provision for milking to continue when the main storage tank is being emptied, e.g. a buffer tank facility. Otherwise, the capacity of the automatic milking system can be seriously limited. It is also necessary to ensure that the cooling system controls can be integrated with those of the milking system.

Preparation For The Transition – The Cows

It often pays to eliminate obvious problems before the transition period.

Recognising that udder health may be adversely affected by the transition period, cows that have high somatic cell counts should be identified and appropriate action taken, ideally on the advice of the farm’s veterinary surgeon. This may include early drying off, treatment or culling.

The supplier of the equipment should be asked to provide details of unacceptable udder and teat characteristics. The herd should be inspected to identify cows with udder and/or teats that are unsuitable for being milked by the automatic system. Whilst it may be worth persevering with marginal cases, it may be worth culling, or at least drying off early, those cows which are likely to present serious problems.

Although very obvious, where necessary, cows should be fitted with the automatic identification devices required by the automatic milking system. Care should be taken to perform this task at a time and in a place that will not result in undue stress to the animals. It should ideally be done some time before introducing the cows to the automatic milking station.

The hair on the udders of the cows should be clipped or singed. Excessive hairiness of udders can cause difficulties with teat locating and cup attachment systems. This should not be done when the cows first enter the milking station in order to minimise stress.

Preparation For The Transition – The People

It is essential that users of automatic milking systems are not allowed to assume that they can just “leave everything to the robot”. All personnel, both from the farm and the installer, must be fully briefed on their responsibilities during the introductory period and afterwards. In addition, the farm personnel should be fully familiar with the operation of the management software package. It is essential that up to date cow information is entered for the whole herd prior to introduction to the system.

Whilst all systems currently on the market have automatic cleaning systems, all have specific requirements for external cleaning, and areas not cleaned by the automatic system. In addition, there are a number of routine service checks to be made by the on farm personnel.

Introducing Cows To The Automatic Milking System

It is vital to try and reduce the stress to the cows caused by forcing them into unfamiliar situations. It does take time and patience to accustom cows to not only go into the automatic milking station, but also to become used to teatcups being attached. It should be remembered that people become tired and short-tempered. This invariably transmits to the cows. As a result, it is probably better to plan the introductory period over a number of manageable sessions.

Ideally, a group of around 30 cows should be walked into and through the milking station. A handful of feed should be given to the cows as an inducement. If this can be done a couple of times, the cows will become more relaxed. Some engineers comment that this process can be overdone, resulting in worse reaction when the cows are subjected to automatic teat cup attachment for the first time.

If it is not possible to familiarise the cows with the milker before the first milking, e.g. if the cows are moving to completely new premises and it is impractical to return to the old facility, it is more important than ever to proceed very gently and slowly on the first day of milking.

When the cows are to be milked automatically for the first time, a manageable group, around 30 cows, should be enticed into the stall using concentrate feed and the minimum amount of human coercion. On the first occasion, such a group size could take around 4 to 6 hours to milk.

Teats should be cleaned manually and foremilked to stimulate a sound oxytocin based let-down response. When this is achieved, the automatic attachment system should be initiated. Depending on the teat location technology in use, it may be necessary for the operator to “drive” the robot arm to the teats on this first occasion. Although the cows may be startled by the action of the automatic milker, they usually settle once the cups are attached.

Once the core group of cows has been milked the first time, it is sensible to initiate a hot clean of the milking equipment, before switching off the machine. It is unlikely that cows will attend voluntarily after that first milking.

It is likely that manual assistance of this type may be necessary for the second and third milkings. After the third milking, the robot should be left switched to allow cows to attend voluntarily. By this time, it is usual for cows to visit the milking station, attracted to the possibility of feed being dispensed.

It is essential that the installing engineers involve the farm personnel as soon as possible in this process. After all, the farm personnel will still be there when the engineers have left. The remainder of the group should be introduced over the following days.

Period Immediately Following Introduction

Following the initial introductory period, it is essential that the farm personnel concentrate on making sure that cows attend the automatic milking station, particularly where the cows are allowed free choice as to whether they attend. Whilst some cows adapt quickly, others will not visit the automatic milking station unless they are forced. Again, provision of one-way gates will assist farm personnel in ensuring that those cows are milked.

Farm personnel should resist the temptation to “help” the automatic milker. Helping during attachment prevents the system from “learning” the position of a cow’s teats, leading to a vicious circle of requiring assistance.

It is also advisable to check that cows are consuming their concentrate allocation.

Management software should be able to generate reports showing cows that have not attended in a prescribed period. Those cows should be brought to the station as soon as possible. Reports should also be used to identify cows that persistently cause failed attachments. These cows should be inspected for reasons. If necessary, they should be removed from the herd.

During this period, farm personnel should also begin to analyse the data collected by the management software.

The Next Few Months – Routine Management

Users of automated milking systems no longer have twice or thrice daily contact with each and every cow in the milking parlour. It is necessary to adapt to “management by exception” and to identify those exceptions from the information collected and processed by the herd management software which should be capable of generating standard management reports and alarm lists. Ideally, the management system should be in “real time” communication with the milking station(s).

In addition, the management software should also be capable of generating customised reports to suit the particular needs of the farm. Most users eventually settle on their own preferred reports, but these should include:

Daily - Printed routinely On Demand - On- Screen and/or Printed
Action Required Action May Be Necessary
  • Activity/Insemination due
  • Cows due to calve
  • Cows due to be dried off
  • Conductivity/Milk Yield variation
  • Other Sensor based indications (future?)
  • Failed attachments
  • Cows attending infrequently
  • Milk yield
  • Conductivity/Udder Health
  • Rejected milk

It is essential that, where action is required, it is taken or instigated immediately. This may mean going and inspecting a cow, or bringing her immediately to the milking station. In less urgent cases, it may be sufficient to programme the system to direct her to the holding pen for inspection the next time she attends for milking, or when the veterinary surgeon is due to visit. Quite clearly, these decisions depend on good stockmanship and knowledge of the herd.

In addition to using the computer generated reports, the farm personnel should develop a routine approach to other tasks such as cleaning down the milking area, inspecting the equipment for damage, wear and tear and other malfunction.

Changing Circumstances

It is particularly important to monitor the frequency and regularity of cow attendance. This is likely to change depending on a number of external factors such as climate, forage-feeding regime, grazing management and even stage of lactation. Where significant proportions of cows’ total concentrate allocation is dispensed in the automatic milking station, decreasing visiting frequency will result in yield reduction and early drying off.

One source of problems can be farms feeding Total Mixed Ration diets where cows in late lactation receive very small quantities of feed in the automatic milking station and have a free choice as to whether they visit the milking station or the bulk feed area. In these circumstances, there is very little to attract the cows to the milking station. As a consequence, milking frequency can decline and lactation comes to a premature conclusion.

One solution is to re-introduce a one-way traffic regime, with a pre-milking diversion for cows not due to be milked. Alternatively, the ration can be reformulated to so that more concentrate is dispensed in the milking station or in a dispenser sited in the exit area.

Many users find that changing the pattern of forage feed delivery, or increasing the frequency, can stimulate cows to be more active and improve attendance rates.

Users must not only rely on information supplied by the farm’s management system, but also make use of data from other sources such as the herd recording organisation, e.g. individual cow somatic cell counts, and the milk buyer. It is essential that, whenever there is any indication of adverse results, that the causes are investigated immediately.

Concluding Remarks

  1. Introducing cows to any new milking facility can be stressful to both personnel and animals. This can be the cause of deteriorating milk quality, udder health and impaired performance.
  2. The extent of the effects of the transition period, or even whether or not such disruption occurs, is largely a consequence of the effectiveness of management on the farm before, during and after the introductory period.
  3. Transitory effects can be minimised by effective management.
  4. Users must quickly learn to rely on reports from the management system as a basis for management decisions and actions.
  5. Users must let the automatic have the chance to “learn” and have faith that the cows will adapt. Whilst leaving the machine to “get on with it” may be difficult, it is often the best tactic.
  6. Constant attention must be paid to all of the information available to the user, whether from the management system, or from off-farm sources, and appropriate action taken.
  7. The automatic milking system is not a substitute for good husbandry or good management. It is essential that all normal preventative measures are maintained in respect of mastitis control cow health and reproduction.

References:

Hillerton J E & Winter A (2000) The effects of frequent milking on udder physiology and health. Proceedings of the International Symposium on Prospects for Automatic Milking, EAPP No 65, Wageningen, the Netherlands.

De Koning, Personal Communication, Research Institute for Animal Husbandry, Netherlands.

Deutsche Tierärztliche Wochenschritt, Issue 108, p113.

Lind et al. (2000) Automatic Milking: Reality, Challenges and Opportunities. Proceedings of International Symposium on Automatic Milking, 19.

Prescott N B, (1996) Dairy Cow Behaviour and Automatic Milking. PhD Thesis. Bristol University, UK.

Rasmussen et al. (2001) Udder Health of cows milked automatically, unpublished draft of paper for Livestock Production Science.

Svennerstein-Sjaunja K et al. The Milking Process in an Automatic Milking System, Evaluation of Milk Yield, Teat Condition and Udder Health. Proceedings of International Symposium on Automatic Milking, 277.

Van der Vorst (2002) Automatic Milking and Milk Quality, Personal Communication, Research Institute for Animal Husbandry, Netherlands.

Author

John Baines

John Baines
1 articles

Fullwood Ltd

Fullwood Ltd