by Tamara Scully

“We’ve seen tremendous growth and adoption of robotic milking within the industry,” Trevor DeVries, Dairy Cattle Behavior and Welfare, University of Guelph, said, estimating that 15% of dairy farms in Canada use robotic milking equipment.

The primary reason dairy farmers switch from conventional milking to robots are quality of life reasons, DeVries said. Robotic milking has an impact on the welfare of the dairy farmer, allowing more time to attend to non-milking aspects of the job and freeing them to leave the farm more readily.

But robotic milking systems also impact the welfare of the cows. Farmers who have adapted the technology believe that the robots equally improved their own quality of life and that of the cows, he said.

Some benefits seen include an increase in milk yield as well as in frequency of milking; better herd health and reproduction; improved herd management and increased data available to make management decisions; enhanced labor efficiency; and increased profitability. But while there are benefits to the cows when using robotic milking, there can also be risks.

Optimizing cow welfare

“We do have areas that may be potential areas for risk when it comes to good welfare of dairy cows on robotic milking farms,” DeVries said. “There may be impediments to cow welfare, but there are also solutions to those.”

The four areas of primary risk are cows cannot milk when they want to do so; cows don’t want to milk at the robot; poor udder management; and poor nutrition management. Understanding why these issues occur and knowing how to prevent or mitigate them is the key to success with automated milking.

Housing design is the major factor impacting voluntary milking. Barn design will influence how often the cows access the robot. Flow to and from the robot can either be free, meaning that access from the stalls and the feed bunk to the robots is not controlled via locked gates, or it can be managed in one of two ways.

Managed systems either force the cows to visit the feed bunk before they can gain access to the milking robots, or force the cows to visit the milking robots before they can visit the feed bunk. In both situations, one-way gates act to select where the cows can travel, and cows may be left standing for increased periods of time.

Timid cows can avoid going through the gates due to the presence of dominant cows. This effectively reduces laying time, as they are left standing and waiting, reduces their milking frequency and yield and compromises their nutrition. Sick or lame cows may also avoid selection gates and decrease their visits to the feed bunks and milking robots in guided systems.

A free flowing system offers cows the most freedom. Farmers may have to fetch more cows in a free flowing barn than in a guided system, but these systems are friendlier from the cows’ point of view.

“Both systems can work well with good management. Where we tend to run into problems is … when management within these systems is not optimal,” DeVries said.

Cow welfare risks increase when they are not able to express their desires freely, and guided systems can cause changes in eating patterns, decreased meal frequency and decreased feed intakes. Increased ruminal acidosis can occur when this happens, particularly if bunk management on the farm is not optimal.

Having too few robots also limits the visits a cow can make to the robot. If the stocking density per robot is too high, milking frequency will decline, decreasing individual milk production per cow per per day. Fifty to 55 cows per robot is the norm and seems to optimize robot use, DeVries said.

Cows may not want to milk if they are lame, no matter the cause of the lameness. There is not much difference in lameness rates between robotic milking farms and conventional milking farms, with an average of 26% of the cows on either type farm rating a 3 or higher on the lameness scale, with 4 or 5 being severe lameness, DeVries said. On robotic milking farms, cows with a lameness score of 3 or above were more than twice as likely to need fetching than other cows.

According to DeVries, lameness will decrease a cow’s daily milk production by 3.5 pounds on a robotic milking farm – a significant amount over time. Lameness prevention requires the proper stall width, length and lunge space. It also means reducing standing time and keeping areas clean. Lameness is more prevalent in barns that use mattresses than in barns using a deep-bedded system, no matter how much bedding is kept on the mattress.

“Deep bedding, in particular sand bedding, is protective, not only from an injury standpoint, but from a comfort standpoint,” he said.

Crowded barns will increase lameness as cows cannot get enough lying time and are forced to stand. It is not true that robotic milking systems will allow more stall stocking. There is a linear association between stall crowding and lameness, as cows prioritize resting time.

Udder & nutrition welfare

Udder health management can cause concern in robotic milking systems. Although robots are much improved in teat cleaning today than in earlier designs, they still are not able to remove dirt as effectively as humans do manually. Controlling environmental factors to increase cow cleanliness is of increased importance, in order to reduce risk of clinical mastitis and reduce somatic cell and bacterial counts in robotic milking barns.

One opportunity robots provide in relation to udder health management is the enhanced ability to dry cows off gradually, by programming the robot to reduce milking for individual cows on a gradual basis. A gradual dry-off decreases pain and prevents mastitis. Feed fed at the robot must be reduced in conjunction with the decrease in milking frequency.

Good nutrition and low stress environments promote good immune health and reduce risk. Nutrition management is different on robotic milking farms. “We’re still learning a lot,” DeVries said.

One primary nutritional challenge is the early lactation period. Subclinical ketoacidosis management in robotic systems involves programming the robots to feed the right amount of energy for each cow, based on stage of lactation and milk production. Cows with subclinical ketoacidosis were shown to produce more milk in robotic systems during the first two weeks of lactation than healthy cows.

Sick cows are “probably not consuming enough energy, whether that be at the robot or at the feed bunk, during that time period,” DeVries explained. “We don’t want cows that are producing more than their energy consumption is allowing for,” so robots need to be programmed accordingly.

Feed intake and milking frequency is linked in robotic systems. Feed bunk management on robotic milking farms requires frequent pushups and to allow cows to use their time more efficiently. Adding two feed pushups to the normal schedule was shown to increase lying time per cow by one hour in robotic milking barns.

Cow welfare in robotic milking systems can be optimized, and risks can be avoided, when cows are able to milk voluntarily, cleanliness of the environment is optimized and proper nutrition is maintained.