Tips for Increasing Milk Per Box

Douglas Waterman, Ph.D

Strategies to increase milk yield with Automatic Milking Systems (AMS).

Douglas Waterman, Ph.D
Douglas Waterman, Ph.D

By: Douglas Waterman, Ph.D, (doug.waterman@trouwnutrition.com). Waterman is the Director, Technology Application – Dairy, for Trouw Nutrition Agresearch. He is responsible for developing training programs to enhance the on-farm competency and skill of dairy nutritionists and provides field support. In addition, he is responsible for the implementation of new technology and products developed by Trouw Nutrition Agresearch throughout Canada (Shur Gain and Landmark), USA and Mexico. Waterman is a native of New York State and holds degrees from SUNY Morrisville, Cornell University, and the University of Kentucky. Prior to Shur Gain, Waterman was the Director for Nutrition and Research for Milk Specialties Company.

 

-------------------------------------------------------------------------------------------------------------------------------

Waterman is the Dairy Profit Seminars speaker on Robotics and Maximizing Milk Per Box: Grouping and Feeding Strategies at 10:30 am Tuesday, August 8. Automated milking systems have skyrocketed in popularity. Many dairy farms have transitioned, are in the process of transitioning or are considering an automated system. If maximizing profit is a goal of your dairy, then maximizing milk per box in an automated system is critical. This seminar will focus on the factors affecting milk per box and the considerations for each. His article Tips for Increasing Milk Per Box provides strategies to increase milk yield with Automatic Milking Systems.

------------------------------------------------------------------------------------------------------------------------------

 Optimizing milk in an AMS has the same basic requirements as in a parlor:

  • High quality forage
  • Good cow comfort (deep bedded stalls)
  • Available feed
  • Adequate bunk space
  • Stocking density
  • Low DIM (days in milk)
  • Low lameness
  • Good quality heifers
  • Low metabolic disease
  • Few treated cows

Figure 1 shows the complexity involved in maximizing milk yield per AMS and the correlation between all the factors. Anything that increases the time available (usage) to milk cows, such as increased milk speed, increased milk/milking, decreased box time and treatment time, will allow more milking/day/AMS. The trick is to balance cow numbers and visits/cow with the number of cows you are willing to fetch.

Figure 1

Cow and Manager:

The cows themselves and the manager play a critical role in the level of milk achieved. Managers need to breed for good (wide) teat placement, teat length and milking speed. Cows with long box times (over 10 minutes) should probably be culled. Herds should average 150-165 Days in Milk (DIM). For every day past 165 DIM, a cow produces 0.18 lb. less milk. The lower DIM allows for more peak days and less cows >250 DIM. Late lactation cows with lower production do not visit the robot as often and their milk speed is slower. Udder pressure will assist in driving the cow to visit the robot.

Having a herd with severe lameness <2% on a 3 point scale is a goal for any herd, but high levels of severe and clinical lameness in robot herds is a huge problem, and negatively impacts milking frequency and production. King et al (2016) reported that AMS herds with lame cows produced 3.5 lb. less milk/cow compared to sound cows, were milked 0.3 times less, recorded more lying time and were fetched more often. Footbaths are a must and should be located on the robot exit, and allow enough room for the cow to comfortably exit the box prior to entering the footbath.

Facilities, Location and Grouping:

Facilities, location of robots and traffic flow are important. Data clearly shows that “free flow” systems produce more milk than “guided flow” systems. Robot location requires easy access with the ability to fetch four to five cows without disrupting the herd. Installing one-way gates allows cows to be fetched without restricting voluntary access to the unit. When there is more than one unit per pen, use one for fetching and the other all voluntary. If possible, the cows should exit the robot towards the feed bunk and water. Data is limited, but personal observation suggests, on average, two and three rows barns produce more milk per box than four row barns, due to greater bunk space and lower stocking density.

There is a definite benefit of grouping by lactation, forming at least heifer and mature groups, with a separate second and third lactation group, if possible. This, as well as a separate fresh group, allows management of the units for maximum efficiency. The results are less competition for heifers, and the ability to adjust robots to meet the group demographics. Further, this allows for different stocking densities (cows/robot) based on milking speed, box time and milk/visit. Heifers produce less milk and have shorter milking times, so a heifer group can handle more cows per box while mature cows produce more milk/visit and generally have longer box times, so less cows/box. When first lactation animals are mixed in with mature animals, the turns, and therefore milk production, are reduced for first lactation animals. With both a parlor and AMS, milk per box and efficiency is optimized by milking second and third lactation animals with good udders and minimal lameness in the AMS.

Feeding:

Formulating the PMR and determining how much pellet to feed at the robot is probably the most debatable topic and is driven more by wanting more visits/cow or more milk/AMS. A hard and palatable pellet with minimal fines is a must to encourage consumption. If pellets are not available, cow visits will decrease, so pellet feeding does entice cows to the unit. But how much is necessary? Many recommend including flavor to enhance pellet intake. Bach et al (2016) showed no benefit in the level of consumption by adding flavors. Others strongly feel that feeding more pellets through the robot increases visits. However, Prescott et al (1998), Halachmi et al (2006), Bach et al (2007) and Penner et al (2017) reported no increase in visits with higher levels of pellet, while Tremblay et al (2016) report a negative correlation between increased concentrate levels and milk production.

In addition to feeding more pellets, it is commonly recommended to lower the energy density of the PMR, to encourage visits. This may increase visits, but milk production is not optimized since cows that are unable to consume all of the pellet will be unable to compensate for the lower energy and protein from the PMR. This makes sense because to manufacture a high quality hard pellet that is palatable, the nutrient density will be low. Feed enough pellet to entice cows to the robot (8-16 lbs.), but have the PMR balanced to support the milk production level. In guided flow systems, increasing the energy density of the PMR may allow for less concentrate to be fed in the AMS, while increasing dry matter intake without negatively impacting visits (Penner et al, 2017).

Robot Efficiency:

To maximize robot efficiency, the robot should be used 90% of the available time (Berdell, 2016). Usage time is the amount of time available after maintenance, cleaning, and washing.  Castro et al (2012) evaluated 34 AMS units on 29 farms and found that cow numbers and milk flow rate had a greater influence on milk yield/box than milkings/cow and refusals. These factors explained 87% of the variation between herds. Milk flow rate is impacted by lactation number, DIM, peak flow, greater milk yield per milking and pre-milking delay time (Sundrucci et al, 2007).

An analysis of 635 AMS herds from North America (NA) showed an average yield/unit was 3587 + 875 lbs. (Tremblay et al, 2016). However, we know of herds averaging 5,000 plus lbs. of milk, and some averaging 6,000 lbs. The focus should be on total milkings and milk/box/day, and less on visits/cow and milk/cow. Achieving the appropriate balance between cow numbers and times milked can be challenging. So, what should the approach be to achieve the most milk per unit?  Between washing, cleaning the unit, and maintenance, there are an average of 22 hours of usable time/day. Completing preventative general maintenance is a plus and minimizes down time. One robot down for a period of one hour, versus all the robots down at the same time, reduces the risk of lost usage time. The next step is to maximize the number of milking visits/day. The goal should be 170-190 visits per day and is influenced by box time, milk speed and treatment time (stimulation to attachment). This goal can be achieved by milking more cows with less milking/cow or more visits with fewer cows (Table 1). More cows with less turns per cow means more milk per milking. Castro et al (2012) suggested the greatest influence on milk yield/AMS was maximizing cow numbers with turns of 2.4-2.6/cow, but milk yield was less than 4,000 lb.

Tables 1 illustrates the impact of box time, turns and cow numbers on potential milkings and milk yield/box. As box time increases from 6.5 to 8.5 minutes, the potential total milkings and turns/cow decreases with 60 cows or the number of cows decreases at 2.8 turns.

Table 1

Table 2 depicts milk/box based on milking/cow and the number of cows/box. As more cows are milked at the expense of visits, more milk per box should be produced. Less milk/cow is expected with less visits/cow, but total pounds of milk increases. However, too many cows may overcrowd stalls, limit bunk space and increase fetch cows.

Table 2

To maximize milk/box, the trick is to balance cow numbers and visits/cow and how many cows you are willing fetch.