Will robots milk Wisconsin dairy cows?

Robotic milkers could someday free Wisconsin dairy farmers from the tyranny of the milking schedule. However, for the short term, the machines will be a risky investment, according to Doug Reinemann, a milking systems researcher at the College of Agricultural and Life Sciences.

Robotic milkers could improve the daily routine of both dairy farmers and cows. If they prove reliable and economically feasible, the milkers would lower the risk of accidents and repetitive-stress injuries. Milking labor consumes nearly 50 million person-hours per year in Wisconsin. Unshackled from the twice- or three-times-a-day milking routine, dairy farmers would find more free time and a social life — scarce commodities on many farms today.

“[Robotic milkers are] like when tractors were first introduced. They were expensive and not very reliable, but they were widely adopted as they became more reliable and less expensive.” Doug Reinemann Milking systems researcher

Farmers might also see bigger milk checks. The machines may allow them to milk more frequently, thus increasing milk yield per cow.

A robot in the barn may improve cow welfare by giving animals more control over their daily routine. Not held in a collecting yard, cows could choose to eat or lie down and ruminate. Less involuntary standing may reduce lameness, and some studies have shown reduced mastitis in herds using robotic milkers.

Four companies, all European, are building robotic milkers. About 700 units have been installed in the Netherlands, where the government subsidized development to help maintain small dairy farms. Those milkers are still under review, pending proof that they will reliably produce quality milk.

Reinemann thinks the machines will eventually pass the quality-milk test. “On average, on normal cows, the machines do better than people because they are consistent,” he says.

However, he thinks that at least for now, humans do a better job of detecting poor milk from cows with severe mastitis, or cows with damaged teats. So far, the robots have had some trouble monitoring udder health and detecting abnormal milk. Other studies have shown that the machines have problems effectively cleaning manure-crusted teats. On the plus side, since the machines monitor and control each teat cup separately, they can provide information on each quarter during milking, as well as reducing overmilking.

A stable milk price would go a long way toward making robotic milkers feasible in Wisconsin, Reinemann says. Europeans are willing to spend money on new technology because they know the milk price they will get. Prices of the milkers must also come down. Right now, a machine that can handle 50 to 80 cows is projected to cost about $150,000.

“In the near term, robotic milkers will require a high level of management skill, and they’ll be a high-risk venture because of the high cost of the technology and the uncertainties of milk pricing,” Reinemann says. “The best candidate is someone in a sound financial situation, such as a 55-year old farmer with money in the bank, who can’t or doesn’t want to hire milkers, and whose kids don’t want to go into dairying.”

When he saw early robotic milkers, Reinemann thought “These are never going to make it.” However, the machines have improved greatly over the past five years, he says. “This is like when tractors were first introduced. They were expensive and not very reliable, but they were widely adopted as they became more reliable and less expensive.”

Labor cost will influence adoption rates. If the supply of low-cost labor stays tight, Reinemann expects more rapid adoption. In the next five to 10 years, he thinks Wisconsin could put in 100 to 200 machines.

Before the machines can be sold in the United States, they must be approved by the U.S. Food and Drug Administration and other state and federal regulators. Since robotic milking is so different from current milking practices, he expects that new rules will have to be written.

Reinemann hopes to get one or two units at the College’s Arlington Agricultural Research Station this spring. He plans to work with the Wisconsin Department of Agriculture, Trade and Consumer Protection to get approval to sell the milk, then generate data to help with regulatory issues.

Denmark has about 50 machines installed, which are monitored by the Danish equivalent of the USDA to gather data on milk quality and system performance. Reinemann sees a similar approach in Wisconsin, where researchers would work with producers to identify key management strategies and head off problems.

At Arlington, Reinemann plans to look at a variety of management issues:

• How can we maximize the number of cows per machine? The Europeans average 45 to 50 cows per machine, but the manufacturers are trying to speed things up. Reinemann thinks that U.S. farmers will have to average 70 to 80 cows per machine to be profitable. Maximizing the number of cows per machine will be a function of barn design, feeding strategies and control of the milking machine, he says.
• Is it better to milk fewer cows more often or more cows less often? In other words, is it more profitable to milk 60 higher producing cows 3 times a day or 90 lower producing cows twice a day?
• How do the robots perform in cold weather?
• What is the best floor layout for the barn?
• What feeding strategies will ensure good cow flow?
• How can we make sure that all the cows come to the machine consistently?

Current robotic milking technology is best suited to confinement dairy systems with 50 to 250 cows, making the Midwest the preferred market for introducing this technology, Reinemann notes. He believes that in 10 to 15 years, there will still be more than 10,000 farms in Wisconsin milking fewer than 100 cows.