Robot Milking Systems

[Heritage Farm]

Hi to All,

 

We are Dairy Farmers in New Zealand and over the past couple of years have become very interested in the Robot Milking Systems. As we are looking to build a new cow shed possibly next year, we have joined the Future Dairy Group in the hope of gaining some first hand knowledge from other Dairy Farmers already utilising one of the systems on the market today.

 

We would be very interested to hear any comments from farmers with their experiences, be it good or otherwise.

Cheers

Dave & Cathy Yates

[Kendra Davis]

Dear Dave and Cathy

I guess we may be seeing some reluctance from farmers that have adopted AMS to expose themselves to the entire Google group website.  Also we should remember that in Australia we only have 5 AMS farmers up and running and most of these haven't had their systems installed for 6 months yet.  I apologise if you get no joy from our group with answers to your query.  It might be more useful to pose more specific questions that people could answer rather than a general request for comments on experiences.  We may also find some farmers may prefer to send you a personal email rather than share information across such a wide audience. 

 

I am not sure how much "homework" you have already done with regard to AMS but I would suggest that it is extremely important that your expectations of the technology and what it will deliver need to be fair and accurate.  I think with an investment this size it is really important to see commercial operations to extend your understanding and knowledge.  There are obviously two AMS farms in the South Island of NZ and the 5 farms in Australia are somewhat geographically spread out. 

 

I would suggest that you try to see a mixture of DeLaval and Lely farms operating (provided you haven't already made the decision which brand you would like to invest in).  I'm sure that DeLaval and Lely would be in a position to be able to get you onto farm that have invested in their machines so that you can assess the two brands. 

 

Kind Regards             Kendra   Dr Kendra Kerrisk
FutureDairy AMS Research Leader www.futuredairy.com.au			Faculty of Veterinary Science The University of Sydney
0428 101 372 (mobile) 02 4636 6327 (dairy office/fax) ken...@usyd.edu.au

 

 

 

 

 

 

 

 

 

 

 

---

From: futur...@googlegroups.com [mailto:futur...@googlegroups.com] On Behalf Of Heritage Farm
Sent: Thursday, 3 September 2009 8:51 AM
To: futur...@googlegroups.com
Subject: [FutureDairy] Robot Milking Systems

[Heritage Farm]

Hi Kendra,

 

Thank you for your reply. I appreciate your comments regarding replies to my email. I guess for a start I just wanted to introduce ourselves and hope to be able to get into some dialogue further along the track.

 

We are situated 35 minutes south of Auckland on the Manukau Harbour. We have split calving and milk approximately 200 Friesian cows in a 28 bale rotary that is well past its use by date. We have done a lot of research into the AMS mode of milking including several visits with Jenny Jago at the Greenfields research farm near Hamilton and are speaking at the moment with Lely, DeLaval and also Westfalia who are looking to have their AMS available in about 12 months. 

 

We hope to travel to the South Island late in November to view both the Lely operations there. DeLaval is keen for us to view one of their Australian operations as well, so hopefully we will be able to arrange that before the end of the year.

 

I would be interested to hear from anyone regarding the methods used for the training of cows to the AMS, what worked and what might be improved.

Cheers

Cathy

 

 

---

From: futur...@googlegroups.com [mailto:futur...@googlegroups.com] On Behalf Of Kendra Davis
Sent: Tuesday, 8 September 2009 11:51 a.m.
To: futur...@googlegroups.com
Subject: [FutureDairy] Re: Robot Milking Systems

[Kendra Davis]

Hi Cathy

The topic of training cows is fairly big topic and I will cover some aspects of it here.  There are a number of "ideal" times and processes but often these are just not practical for the individual situation.  It really depends on the calving spread/pattern and the timing of the new dairy coming on line and whether there are any plans to decommission the old dairy prior to the new dairy starting up or to run the two in parallel at all. 

 

Pay particular attention to training batch size and pasture allocation in this hectic period.

Training batch size

If the new dairy is commissioned during a ‘dry’ period (when all cows are dried off in a seasonal herd) then training will be dictated by calving spread but for training occurring at other stages, there are two main training batch size options. Initially train only a small group of animals - about 10 per milking unit. This allows full commissioning of the machines and downtime to modify machine function. You don’t need the pressure of having to achieve a high number of milkings per day in the early stages. These cows will also assist in training the subsequent batches/groups of cows that will join this group in approximately 2-6 days (pending successful machine commissioning).

 

If the majority of the herd is in milk, train large groups of animals after the initial small group. While it may be very daunting and depending on the herd size and the cow/machine ratio, it is possible to train the remainder of the herd in one batch. This approach reduces the length of time necessary for training of the entire herd and may minimise the time in which cows are being milked through two facilities (the conventional and the AMS). You can chose to train the remainder of the herd in one group or introduce 30-40 animals per machine first, then a further 10-20 cows per unit in a final group. The final group is the most difficult to work with as the competition for machine time is greatly increased. It is important that you minimise any negative impacts on the existing herd (e.g. long waiting times) to ensure that they do not reduce their willingness to visit the dairy.

 

You would expect for example, that if each group is introduced to the dairy on a Monday they should be well settled by the following weekend. This allows the majority of staff to take time off to revive and prepare for the next week.

 

Pasture allocation

In the first week of commissioning or when training a new batch of cows, ensure the amount of feed made available in any one location is kept to a minimum. You may need to reduce pasture allowances and increase access to feed on a post-milking feedpad to provide a strong incentive for cows to leave the paddock and voluntarily explore the system. In a less intensive system, you may need to offer three small pasture breaks per day to ensure that cows regularly deplete their pasture within a given paddock. Your aim is to reward cows that voluntarily move around the system with a fresh allocation of pasture or feed. To get to a fresh break, cows should have to traffic through an automated drafting gate and may require milking prior to being released to the fresh pasture.

 

Stage of lactation at change-over and likely impact on production

Late lactation

Late lactation Training cows in late lactation is ideal because while cows can be difficult to motivate to move, production is already low. Training at this time means they will have well established trafficking habits by the start of the next lactation. This will not be an option for those desiring a quick changeover from old to new dairy. Suitable only for seasonal calving herds.

Mid lactation

Mid lactation cows are more difficult to train than early lactation cows. With half of the current lactation in front of these cows, poor training will result in slow adaptation and quite a loss of production. In addition, many cows may be dried off early due to low production. Staff have to remain vigilant to ensure that they adapt to the system as quickly as possible. Expect cows trained in mid-lactation to have a lactation production level about 5-10% lower than previous and/or subsequent seasons.

Early lactation

Early lactation cows are the easiest cows to train because they have strong appetites and feed incentives work well. High levels of motivation mean cows learn faster as they are more inclined to explore in search of fresh feed. Note though that there is still about a 5% reduction in lactation performance for the season compared to the previous and subsequent seasons.

Dry period

Training cows when they are dry aims to achieve cow adaptation to the system during a period when milk production is unlikely to be negatively impacted. Dry cows have very low levels of motivation so feed incentives are less effective. This means labour is required to encourage cows to the dairy. Train dry cows for short periods only so that cows are not taught to expect just a ‘pass through’ and a feed of supplement without receiving a milking. Two weeks is the recommended period for training dry cows.

 

The biggest thing with training the herd is to be well prepared, and to ensure that staff have an understanding of what to expect.  Expect the best but prepare for the worst.  The above is based on our own experiences but it will be interesting to hear from others who have experience and learnings.

Kind Regards
Kendra Kerrisk (nee Davis)
Senior Scientist
FutureDairy

[Heritage Farm]

Thank you for that Kendra, it is very helpful.

Cheers

Cathy

 

---

From: futur...@googlegroups.com [mailto:futur...@googlegroups.com] On Behalf Of Kendra Davis

Sent: Wednesday, 9 September 2009 7:41 a.m.
To: futur...@googlegroups.com

[Vaughan Jones]

From http://www.grazinginfo.com the farmers¹ web site.

Automatic milking

This becomes a complete system covering the whole farm layout which is quite
a big job to set up. The farm layout and animal movement control is quite
complex.
The Waikato Greenfield one run from 2001 by Ruakura, now DairyNZ, can be
seen on http://www.dairynz.co.nz/page/pageid/2145836818/Greenfield_Farm
It was found that cows didn¹t like getting up in the morning before about 7
am, and they didn¹t like coming to be milked until about 9 am, resulting in
a queue of a quarter of the herd until about midday - not good in hot sun,
or for milk production and muck in lanes.
About 90% of the cows chose to be milked only once a day, so Friesian
production will suffer compared with smaller high milk solids cows.
Training cows is a job. Research has shown that Jerseys learn new systems
faster than others.
Repairs and maintenance was high with the technician having to come to
Greenfield almost every day. They had a log book that I checked a dozen
times over seven years when I took clients there. Newer machines may have
reduced this. I suggest that if the system suits you, buy an established
brand that has sold thousands, not a less unproven one.
Some have bad design faults, for example the floor should slope upwards
slightly because cows like standing that way. It is one reason why cows like
rotaries so much. Also standing slightly up discourages cow moving
backwards, which if done slows cluster application. The cow should not be
able to see anyone who enters the area and they should not talk. At
Greenfield if someone came in and spoke the cow in the bail would move back
and lower its head to see who was there.
The total cost of an automatic milker is excessively high with building, new
lane and fence layout, animal control and direction system, and can milk
only about 100 cows.
Before spending about a third of a million dollars, look into the
possibility of part time milkers.
Also, do a budget.
The cost to own and milk a cow normally is $750 per annum which is $650 in
an automatic milker.
The automatic system cost per cow is about $3,500.
Cost or value of cow $1,500
Total $5,000
Cost of money at 7% is $350 per annum, plus the $650 above equals $1,000 for
$1,700 worth of milk, so a gross income of $700. Land value is not included,
but at 7% can be $700 per cow in the Waikato, leaving no profit.
Greenfield fed grain to get the cows to come in. This cost is not included.
Production in the Waikato Greenfield system was lower than the area average.
Their figures are available.
Be aware that some AMS have been removed. One reason given was because it
was not installed in the centre of the farm. If this is done, power and
tanker track (and fencing) costs have to be included.
Large herds changing from a herringbone to an internal rotary produce about
a litre more milk per cow per day in the flush, and save one labour unit and
a house which need not be built or can be rented.
I would not install AMS. Think about everything; in season detection, AB and
mastitis treatment.

Vaughan Jones
78 year old ex dairy farmer and International Agricultural Consultant
Hamilton
New Zealand
Le...@GrazingInfo.com

[Kendra Davis]

Hi Vaughan

Thank you for your email, I think this is a very worthy discussion with many valid points that deserve comment.

 

We need to be clear that AMS is not for everyone and I am not interested in encouraging people to invest in such a technology if it is not right for them or their business.  I put a lot of emphasis in telling people to make sure that they have accurate expectations to ensure that the technology delivers for them.  I will take the time here to address all of the issues that Vaughan raises as I think some of his points are valid whilst others may be site specific.

 

We consistently see low throughput between about 3am and 6am (true, this is when cows like to sleep) so ensure that you are looking pasture-based numbers when you considering potential throughput rather than European generated numbers.  See our Google group website for an info sheet on throughput/machine utilisation.

 

We very rarely see queues of cows greater than about 30 animals (total herd size 165 cows) which equates to the last cow waiting 2 hours to be milked.  It is important that the system is managed in such a way as to achieve this otherwise the problem is compounding with cows not wanting to come to the dairy due to previous bad experiences of queuing. 

 

The Greenfield project always targeted a low milking frequency of 1.3 milkings/cow.day to allow more cows to be milked per machine thereby optimising the amount of milk harvested per machine and increasing the efficiency of each milking session.  The decision for this approach was somewhat driven by the once-a-day milking work that was carried out in NZ.  It was thought that a milking every 18 hours (not once-a-day) would minimise the reduction in milk production per cow.  To get an average of 1.3 milkings of course there will be cows that are achieving only one milking per day.  At Camden we have a split calving system and cows producing around 7,500 litres/year (550 kg MS/cow and 1500 kg MS/ha).  Throughout the year we hold a milking frequency across the herd of about 1.8 milkings/cow/day.  We generally hold the early lactation cows above 2xday and let the late lactation cows drop to around 1.5xday.  People need to be sure that they are targeting a system type that will suit their cows and their targets. 

 

I agree that with my time at Greenfields we found training cows to be a big job and quite time consuming.  However, I now personally believe that a lot of that was due to the type of machines we had in operation.  Whatever the reason certainly in the three years that I have been in Camden we have found that training cows is no longer a labour intensive task and that carried out with the right approach we have cows and heifers that learn the system extremely quickly.  Generally cows will be prepared to milk themselves within 3 days and the whole herd should be prepared to come to the dairy and milk themselves within 6 days.  This is no longer a task that we dread.

 

Repairs and maintenance of any AMS is likely to be somewhat higher than a conventional plant - compare maintenance of a car vs. an aeroplane.  AMS machines have improved a lot in the past 8 years and certainly we will see some brands far outperforming other brands.  The major manufacturers have put a lot of effort over recent years into improving the serviceability of the machines, minimising any down time of the machines, maximising performance and reliability of the machines.  We would be able to count on one hand how many times we have had to call out a technician to address a breakdown with our two robots over the past 3 years.  I would expect a similar situation with any major brand of AMS.  The costs of service and maintenance will be set by the manufacturer in a contract.  Farmers should ensure that they ensure that the contract will suit their needs and that the cost is easily calculated.

 

Brand design concerns - I agree buy a brand that is well known.  At the moment there are only two brands that are actively selling in Australasia and I see that both these (Lely and DeLaval) are reputable and will see you investing in a product that you can be sure is proven.  I don't think the floor needs to slope up - there may be some advantages of this but we have not concerns with our flat floored AMS units.  Also we have no concerns with people entering the milking area or talking etc.  The problem that the Greenfield team had was that the specific design of the machines is such that a cow with her head in the feedbin had to step back and look down to see things in the milk room when she heard people.  I can't comment on the Lely machines but certainly with the DeLaval machines a cow simply needs to lift her head out of the box to see what is happening.  I think it is important that cows be able to see out of the machines so that they don't become anxious about noises that are unfamiliar to them.

 

It is very important that people put a considerable amount of time into developing a budget for the purchase and installation of AMS.  I personally think that many farms will not require too many changes with regard to laneway and paddock configuration - it depends what they have to start with.  The area outside the dairy obviously will need some work to allow cow traffic control systems to be implemented (one-way gates and pre-milking drafting gates).  The dairy itself need not be particularly large or flashy.  The original dairy at Greenfields was very modest, in addition the yards associated concrete is generally a small area (needing capacity to hold only about 30 % of the cows in the herd at any one time).  

 

If you are putting a new dairy in - either a new conversion block or replacing an old dairy that needs decommissioning then it is possible that AMS is an option.  The cost of the technology is higher than a conventional dairy so you would not necessarily do this to make more money.  There may well be some labour savings - depending on how you implement the system and capture the benefits.  There should definitely be lifestyle savings but these are difficult to put a dollar value on.  One thing is sure - make sure you do the economics for your situation, don't use generic results put out by our scientists and consultants - the investment is too big to use guesswork.

 

Greenfields showed that whilst there was some benefits of feeding grain in the milking station they successfully had half of the herd unfed in the stations for 12 months.  This ~80 cows performed just as well as the ~80 cows that were fed with regard to milking performance.  Obviously the extra energy of the fed cows resulted in extra milk but be clear that the system is not reliant on an incentive in the bail to make it work.  At Camden we do use a pelleted concentrate but certainly we have not problems with cow traffic if we run out of concentrate for whatever reason.

 

I would suggest that the lower than district average of the Greenfield cows was mostly attributed to the lower than district average milking frequency (remember they were targeting this).

 

I disagree that the dairy needs to be in the middle of the farm.  Our farm layout at Camden is such that the dairy is right at one end of the milking area and we have one central laneway leading off from the dairy.  Ours is not an ideal design but it works and we achieve good milking frequency and production levels.  As far as I am aware there is one farm in Australasia that has removed its robots and the reasons were justified and were not due to the dairy not being located in the centre of the farm. 

 

Oestrus cows, AI and mastitis treatment are all just part of this new way of farming.  The system would not be feasible if we didn't have a way of coping with such aspects of dairy farming.  The key is making sure the dairy and surrounding layouts allow you to generate a routine and system that works for you. 

 

People need to be sure that they do their homework prior to investing in such technology.  Ensure it is right for you,

AMS is not for everyone, I don't sell machines so I have no interest in people buying any brand of machine if it is not right for them.  There is nothing more damaging than for the industry to witness people pulling out machines as quick as they are putting them in.  Rest assured that the manufacturers also are not interested in selling machines to people that will not make them work for obvious reason.

 

Kind Regards
Kendra Kerrisk (nee Davis)
Senior Scientist
FutureDairy

University of Sydney
Private Mailbag 3
Camden, 2570
NSW
AUSTRALIA

Mob:      0428 101 372
Dairy:    +61 2 4636 6327
ken...@usyd.edu.au
www.futuredairy.com.au

 

[Kendra Davis]

You're very welcome.

[Vaughan Jones]

Thank you Kendra.

> We very rarely see queues of cows greater than about 30 animals (total herd
> size 165 cows)

Is that through one unit?

Is that 3.5 hours waiting if milking time is 7 minutes.

> Also we have no concerns with people entering the
> milking area or talking etc. The problem that the Greenfield team had was
> that the specific design of the machines is such that a cow with her head in
> the feedbin had to step back and look down to see things in the milk room when
> she heard people. I can't comment on the Lely machines but certainly with the
> DeLaval machines a cow simply needs to lift her head out of the box to see
> what is happening. I think it is important that cows be able to see out of
> the machines so that they don't become anxious about noises that are
> unfamiliar to them.

A radio in all dairies is a good idea proved by trials and if on all the
time discourages birds from entering. A dropped bucket etc., doesn¹t worry
them.

When we put a radio in ours in 1957, cows became more relaxed.

Best wishes.

Vaughan Jones
Hamilton
New Zealand
sup...@grazinginfo.com

http://www.grazinginfo.com

[Kendra Davis]

Hi Vaughan (and others)
Sorry, I should have specified.  We have 2 machines so a 30 cow queue is just under two hours for the last cow to be milked. 

We go through phases of radio or no radio.  For us it doesn't seem to make a whole lot of difference.  The cows are very relaxed and laid back with the system anyway.  I think the radio often does good things for "people temperament".

Kind Regards
Kendra Kerrisk (nee Davis)
Senior Scientist
FutureDairy
University of Sydney

Private Mailbag 4003
Narellan, 2567