Peter Edmonson is an Irish veterinary surgeon currently living in the UK. He has 35 years of experience in dairy practice in Ireland, Saudi Arabia, China and the UK.  He has set up his own consultancy, ‘UdderWise,’ which aims to deliver practical solutions to the dairy industry throughout the world.

In this webinar, Peter begins by defining a robot as a food factory. It has advantages in that milk quality should improve, and animal welfare also. There are some 38,000 robot units worldwide and this number is rapidly increasing. With new technologies such as this, new management skills are required, and these are dealt with during the webinar and he notes that there is a big change from milking in a parlour.

The management skills and training required can be summarised:-

• How to deal with warning signals

• How to diagnose disease

• Cow observation

• Variation between manufacturers

• Not all staff have the same skill levels and

• We are still learning about robots!

When discussing robot capacity, one robot should be able to milk between 50 and 55 cows with yield per robot of ideally 2000+litres. The all-important idle time is set at about 15%. The question is asked, ‘What is idle time?’ This is explained in the following way:-

• Time is when the robot is available for milking

• Two hours is set aside for washing the robot per day

• Robot is available therefore for milking approximately 22 hours

• Cows are milked for 19 hours

• Idle time = 3/22 =14%

Generally, the aim is for three milkings per day but some cows are bullies and push their way into the robot and it is heifers that are most likely to suffer bullying. If the ‘idle time’ is inadequate, cows that are bullied may only get milked twice daily. There are pros and cons of the ‘feed first robot after’ system as opposed to the ‘robot first feed after’ system as it affects cow flow. This is illustrated by on-farm pictures. There are many of these during the webinar reflecting Peter’s huge practical experience.

Free access means that cows can easily access the robot whenever they want. This helps avoid queuing and a figure of five metres around one robot or seven metres around two robots is suggested to avoid congestion. Even grooming is automated with large spinning brushes for cows to use-but these need to be sited away from robots to avoid congestion, as cows love the brushes!

Advice is now given about planning a new robotic system as there are many on the market. Many farmers choose the latest and more expensive as they want the best, but some second hand units may suffice for an individual farmer. There is also advice on cow selection, looking at temperament (heifers are quickest to learn), cell count, lameness, feet and age, for example. As one would expect from the title of this webinar there is a lot of time spent on teat preparation, which should be as good or better than normal milking. This is not always the case and one article published in 2011 talks about 67% success, although technology has moved on rapidly since then and this figure should now be higher. Nevertheless robots can’t differentiate between clean and dirty teats at present, although some robots use rotating brushes to wash teats. Teats are not always dried, although some newer systems use a wash and dry process and can pre-dip.

More on-farm pictures demonstrate the sub-optimal hygiene standards on some farms with cows lying down in dirty conditions. There is an amazing picture in this section of a very hairy udder-hardly conducive to hygienic milking-the farmer wanted to know why Peter was taking the photograph! There are also some pictures of heavily soiled tails, another predisposing factor for the development of mastitis.

In the robotic system mastitis detection is automatic in nature:-

• Quarter conductivity

• Light emission

• Milk yield

• Milking frequency

• Temperature

According to international standards auto mastitis detection should have a sensitivity of at least 70% and a specificity of at least 99% (1% false positives allowed). Some technical information follows on mastitis alerts messages with many farmers being unaware of false positives, which could lead to overtreatment. Mastitis warning signs are covered and its detection difficulties discussed. Much of this will be familiar to colleagues in farm practice but its relevance in robotic systems may be new information.

In robotic systems the key steps to reducing mastitis are suggested to be:-

• Clean udder and teats (environment)

• No overcrowding

• Adequate idle time (15%)

• Cow flow system

• Clean teat before milking

• Disinfecting units between cows

• Straw yards are not ideal-another excellent picture shows some contented looking, but very dirty, cows lying on straw.

Even though it has been many years since I diagnosed and treated a cow for mastitis, there was a lot to enjoy in this superb webinar. Peter said at the beginning his wish was to educate and entertain. He succeeded on both counts.