Feeding trial with robotic milking systems using combined feeding management

Feeding trial with robotic milking systems using combined feeding management

Before speaking about the numbers related to multiphase feeding management in AMS systems, it is important to consider an important factor. Why do we consider this factor important? Because increasing wages are relevant for all agri businesses, although in different aspects depending on their type and volume of output.

The monthly average gross wage in Hungary was HUF 557.900 in September 2023 according to the flash report of KSH (Hungarian Central Statistical Office), 14.1% higher than the previous year. The report also shows that agriculture is ranked 17/19 in the ranking of industries for this index, although with a similar gross increase of 14.2% (up from 385.400 HUF/month in Sept. 2022 to 440.200 HUF/month in Sept. 2023) despite the ranking.

How businesses have evolved in the past few years according to their production volumes:

  • on family farms there is an increasing need for a better lifestyle (quality time with the family)
  • industrial farms are finding it increasingly challenging to find and retain skilled employees (absorption by other sectors)

Further to the above aspects, robotic milking systems also represent a significant investment cost, so it is essential to plan for their financing and return on investment. One of the major and most decisive cost factors in automated milking systems is feeding. Feeding management in robotic milking is a puzzle-like combination of two different elements. With one element being the PMR (Partial Mixed Ration), the other a pelleted lactation concentrate, fed in the milking box 24 hours per day according to stage of lactation, production and gestation, with the cow usually staying in the same production unit from calving until drying off.

We teamed up with a key partner, Állért Ltd. to perform a feeding trial in 2023 to evaluate the feeding management options of two-phase feeding in a robotic system. We had the objective of feeding a higher quality pellet – with improved protein composition and palatable ingredients – and a so called finisher pellet – to address economic issues and prevent the animals from getting fat – parallel to the already used robot pellet. We wanted to assess if there is any unused production potential of the fresh cows, and if we can provide enough motivation in the second half of lactation by providing feed in a particularly economic form without compromising production.

Our partner has implemented the following investments towards a free traffic robotic milking system in a planned, stepwise manner:

  • 2021: construction of a new 2×60 stall house with 2 Lely Astronaut milking robots, with 2-phase feeding option in the milking box.
  • 2022: installation of 4 Lely Astronaut milking robots in an existing 200 stall house, with 1-phase feeding in the milking box.
  • 2023: installation of 2 Lely Astronaut milking robots in an existing 120 stall house, with 1-phase feeding in the milking box.

Our company has developed 3 categories of pellets to entice the cows to visit the robot: Premium, Medium and Basic. The aim of our trial was to assess how cows fed Medium feed – or a version of it adjusted to specific situations – react to being fed a combination of robot pellets of different quality and composition. We investigated how a reasonable combination of these feeds affects production and how feeding costs develop with the use of the 2-phase feeding management on the farm.

The duration of the trial was almost 100 days, starting at the end of summer and ending before the year was over (16th of August 2023 – 23rd of November 2023).

The average number of cows participating was 116, with cows designated to groups 4 and 5 according to the objectives of the trial. We have to note that these groups traditionally included 10-15% multiparous cows because of how groups are used and for farm management reasons. This might have affected mainly the comparison with the control group in relation to the absolute values for milk production.

The trial was composed of the following phases:
Phase I: Medium + Basic feed combination, 17.08.2023.- 21.09.2023. (36 days)
Phase II: Premium + Basic feed combination, 22.09.2023.- 18.10.2023. (27 days)
Phase III: Premium + Medium feed combination 19.10.2023.-23.11.2023. (36 days)

The parameters for the daily total volume offered in the milking box and the ratio of the robot feeds were set in the farm management system before all three phases, and a calibration of the shipped pellets was performed for each robot. We tried to assign cows to the treatment groups (group 4 and 5) and the control group in a way that no significant differences would arise regarding days in milk.

The following diagrams (Figure 1 and 2) demonstrate the amount consumed per day and per type of feed in each feeding phase, also in comparison to total daily feed intake in the control group. The control group was fed Medium pellets throughout the trial.



In phase I, when Medium and Basic pellets were combined, daily feed intake was 9% higher than feed intake in the control group. In phase II, when Premium and Basic pellets were combined, daily feed intake was 6.4% lower than feed intake in the control group. In phase III, when Premium and Medium pellets were combined, daily feed intake was 6% lower than feed intake in the control group.

Figure 3. demonstrates milk production and feed intake for the whole trial period. In phase I, pellet consumption decreased slightly accompanied by a decrease in milk production, but the weather was hotter in this phase resulting in heat stress for the cows. Pellet consumption increased slightly in phase II and milk production also increased slightly. Feed intake was highest in phase III, accompanied by increased milk production.


Figure 4. demonstrates the daily cost of the different pellet combinations for the 116 cows participating in the trial paired with the milk production achieved.

RoboticAverage purchase price of milk was around HUF 156 during the trial period (100 days). If we summarize data in a table (Table 1.) and calculate income and total cost of robot pellets for each feeding phase, we can draw further conclusions.

RoboticOur conclusions are summarized below:

  1. In cases when the combination of pellets included Premium pellets and thus provided a higher intake of nutrients compared to feeding Medium pellets only (control), total feed intake in the robots was lower in the treatment groups than the control groups. This was the case for phases II and III. The opposite was observed in phase I, where Basic feed was combined with Medium feed. Intake was increased for the Basic pellets that contained less energy. It has to be noted that we had warm summer weather at the beginning of the trial when phase I was conducted.
  2. We can conclude that phase I of the trial is less valuable because these two types of robot pellets (Medium + Basic) were fed during a period of significant heat stress. The period included 16 heatwave days, making up for 45% of this feeding phase. We also have to note that the decrease in production showed a similar pattern and rate for both the treatment and the control groups.
  3. We can draw the following conclusions by comparing the profitability numbers for the three phases. Robot feed intake was highest in phase III, the trial phase (Premium + Medium) where milk production was highest, both in daily volumes and per unit of feed. Robot feed intake was lowest both in volume and cost in phase II of the trial (Premium + Basic). Assessment of the ratio of robot feed cost and sales revenue for a given period revealed that robot feed cost per sales revenue was the best for the phase when Premium + Basic pellets were fed, with a value of 8.91%.

If we deduct the cost of robot feed from the sales revenues of milk for the given period, it is clear that phase III is the most favourable: with the highest milk production and considering the cost of robot feed, it produced the highest “net” income due to the high milk production.

This trial involved a lot of work for both the farm and the consultants but served as a good experience because of its novel approach. Using high quality ingredients to produce robot feed – to achieve better nutrient content and palatability – is a very important factor and an investment with good return. Using higher priced robot feed reasonably will result in higher milk production and better return. In order to improve return on investment for automated robotic systems it is important to implement a system that can handle multiple types of robot feed and to maximize profit by addressing both production and cost aspects.

Szilvia Molnár
Róbert Halász
Bonafarm-Bábolna Takarmány Ltd.

Related Posts