Labour organisation on robotic milking dairy farms

B.R. Sonck

    Research output: Thesisexternal PhD, WU


    1. Research issues

    The research described in this dissertation is focused on the effects of the integration of the milking robot in a dairy farm on the labour organisation at operational and tactical level. Attention was paid to the future requirements concerning human labour and labour (re)organisation with respect to the complex interaction between the cows and an automatic milking system (AMS) on a robotic milking dairy farm. The study was divided in a number of research issues (Chapter 1) :
    (a) What is the capacity of a milking robot ? In determining the amount of human labour that can be replaced by the milking robot, the capacity of the milking robot forms an indispensable basis for calculations of possible labour requirement when an AMS is used.
    (b) What are the remaining "milking" operations and work elements of the farmer according to the chosen work method with the AMS ? Because cows will be kept closer to the milking system, other grazing systems than unrestricted grazing where cows are pasturing day and night will have to be applied with automatic milking. Therefore, the question requires an evaluation of the work methods with an AMS and grazing systems. A grazing system refers here to a specific time distribution of "keeping cows in- or outdoors" on daily and yearly basis.
    (c) What are the effects of different AMS management strategies on the daily labour requirement and labour organisation at operational level?
    (d) What are the effects of different AMS management strategies on the annual labour requirement and the labour organisation at tactical level ? What are the possible annual labour savings in comparison with conventional milking dairy farms?
    (e) The results of the above-mentioned research questions will have to give indications about labour quality and the quality of life of the farmer on robotic milking dairy farms.
    The two main AMS management strategies considered in the dissertation are : (1) the fullyautomatic strategy with computer-controlled cow traffic based on voluntary visits of the cows to the milking point during 24 h of the day, a cubicle house with restricted one-way cow traffic and individual concentrate supplementation and milking (AM-CCT) and (2) a semi-automatic strategy with humancontrolled cow traffic based on milkings at fixed moments of the day, under supervision of the farmer and either individual concentrate supplementation by means of computerized self-feeders or group feeding using a total mixed ration (AM-HCT). All studies discussed in this dissertation used an automatic milking system with a 'Prolion' milking robot with one robot arm serving one or more milking stalls. From an evaluation of the dairy technology, we conclude that technical solutions for the automation of each milking operation performed in conventional milking parlours are potentially available (Chapter 2). Automatic milking can replace the milker and the milking parlour to a large extent.

    2. The milking capacity of a milking robot

    The interaction between milking robot and cow was investigated by means of a simple formula of tuning which enables to calculate the milking capacity of any type of milking robot (Chapter 3 or research issue (a)). Using this static model, the main factors of robotic milking such as the times required for cow movements, milking processes and robot motions were investigated. It was shown that for a milking robot with one robot arm serving two stalls, the idle time of the robot arm was 54% and that the robot arm can serve up to four milking stalls in-line. The model showed also that the capacity of an AMS arrangement with two stalls in-line, can be increased from 11.7 to nearly 15.4 cows/h by increasing the robot speed, by simultaneous executing of some milking processes such as the simultaneous opening of the entrance doors of the milking stall and the milking parlour, and by changing the sequence of milking processes. If the robot is available for milking for 20 h, 308 milkings could theoretically be carried out. Thus, with a milking frequency of four milkings per cow per 24 h for the whole herd, the milking robot could serve nearly 80 cows.
    With the same modelling approach, formulae were developed to assess and evaluate fictitious AMS arrangements and to estimate capacities. A double or rotary tandem AMS arrangement can be a good alternative for an AMS with four or five milking stalls in-line. In a 2 x 2 tandem arrangement the capacity (cows/h) would be 8.5% higher than in a one-row arrangement. Cycle analysis showed that with a rotary tandem comprising five milking stalls a capacity of 29 cows/h can be reached. Arrangements with more than five milking stalls do not improve capacity, if the speed of the robot processes cannot be increased. If these processes could be carried out faster six milking stalls could be used and a capacity of nearly 39 cows/h could be reached.

    3. Labour organisation at operational level

    In general the farmer can allocate his time to milking job, non-milking jobs, personal care and social activities. Two kinds of operations can be distinguished in the milking job using automatic milking viz. planned and unplanned milking operations.
    Times for planned milking operations were derived from observations on commercial farms where automatic milking was combined with a human-controlled cow traffic and on an experimental farm where automatic milking was combined with computer-controlled cow traffic (Chapter 4 or research issue (b)). Based on these work studies, the 'planned' milking operations of the milking job were derived for automatic milking methods combined with five grassland strategies. Seventeen variants were quantified by means of a case-study. Calculations with a developed task time program show that the automatic milking method with human-controlled cow traffic applied during the whole year and with a milking frequency of three times a day results in important physical labour savings for milking (37.9%). This method allows to apply grazing systems where cows are pasturing even day and night. However, automatic milking with computer-controlled cow traffic with cows kept indoors the whole year results in the largest labour reduction (66.1%).
    The unplanned milking operations include (1) repair of robot failures, (2) bringing cows that exceed a maximum milking interval to the milking point and (3) interventions for cows which fail automatic teatcup attachment. Malfunctioning of the AMS will determine the occurrence of unplanned milking operations which can disturb the daily labour planning. On the other hand, the daily planned tasks during which the farmer cannot be disturbed and is unavailable to the AMS, will delay unplanned milking operations and therefore negatively affect AMS functioning. To study the interdependency of automatic milking and labour planning at operational level, a dynamic stochastic simulation model (Chapter 5) and a program for labour planning quality (Chapter 6) were developed.
    For bringing cows to the AMS, it was found that bringing cows during three fixed periods to the AMS is preferable with respect to the low labour requirement, the low impact on labour planning and the negligible negative effect on the average milking interval and the milk production. The farmer will have to learn how the cows behave in the cubicle house and depending on their visiting pattern and production level he has to choose the maximum allowable milking interval. The choice of the maximum milking interval has a marked influence on the number of cows that need to be brought to the AMS and consequently, on the labour requirement of this operation.
    One of the most important concerns of potential robotic milking dairy farmers is how to deal with robot failures. Robot failures and repair are defined as unplanned milking operations. The simulation model described in Chapter 5 makes it possible to study the effects of robot failures on the quality of the milking process for different degrees of availability of the farmer or the maintenance service to the system. The results show that a permanently available maintenance service is very important to guarantee the quality of the milking process, especially on those farms where the AMS already operates at the limit of its capacity. We learned that the milking process will benefit more from a farmer who is able to repair most of the robot failures himself and without delay than from one who always immediately calls in the maintenance service of the robot manufacturer.
    Cows unsuitable for automatic teatcup attachment require additional work from the farmer. The farmer will have to set a maximum time that he wants to spend for this operation. When the farmer aims to spend a maximum of 0.5 h per group (three groups per day) for milking the separated cows, the herd may consist of 6 to 7% cows that are unsuitable for automatic attachment. Culling of these cows can be considered to reduce the labour input. If the cow is a high producing one, the decision may be hard. It is up to the fanner to set out the pros and cons before taking a decision. Failing teatcup attachment can also be caused by the system itself. We can derive that the milking robot has to achieve an attachment score of 93 to 94% to limit the additional work of the farmer to 1.5 h a day. This additional work consists mainly of supervision. Only 12 to 20% of the time is used for physical work.

    4. Labour organisation at tactical level

    The effects of the integration of an automatic milking system on the labour organisation of a dairy farm at tactical level will depend on the characteristics of the farm at the moment of the introduction of the AMS and on the automatic milking management strategy applied once the AMS is integrated in the farm. By combining two existing programs, namely the IMAG-ARBGRO labour budgeting program, extended with task time modules for automatic milking and the program 'Standards for Fodder Supply', it was possible to calculate the labour budget of robotic milking dairy farms with various grazing systems applied and to compare these farms with conventional milking dairy farms (Chapter 7). In several experiments we studied the following grazing systems combined with automatic milking : unrestricted grazing with or without supplementary feeding of maize silage, restricted grazing with or without supplementary feeding of maize silage, zerograzing (cows indoors the whole year and feeding fresh cut grass) and summerfeeding (cows indoors the whole year and feeding grass or maize silage). Fully automatic milking based on a 24 hours attendance of cows to the AMS supposes summerfeeding or zerograzing.
    Comparing various grazing systems for farming plans with grassland only we found that all cases result in a labour reduction of at least 15%, with a maximum of 22.2% (approx. 1150 h) for restricted grazing or overnight housing with supplementary feeding of 6 kg DM maize silage per day. For farming plans with grass- and maizeland the introduction of an AMS results in labour savings of 800 to 1000 h per year for all grazing systems compared. Farmers using contract workers for grass silage production will profit more from automatic milking in terms of labour reduction. For these farms, the labour reduction ranges from 923 h (20.0%) to 1371 h (29.6%) for farming plans with only grassland and 816 h (17.9%) to 1361 h (29.9%) for farming plans with grassland and maizeland. If we compare different grazing systems for this case, summerfeeding appears to be the best alternative.
    From the discussion in Chapter 8, we summarize that automatic milking is a working tool for the farmer that will lighten the mental and physical load and as such will lead to a higher work quality. The farmer will need to become acquainted with this technically and electronically sophisticated device. The farmer will become more a brain worker than a manual labourer. The effects of automatic milking on his family and social life will depend on the AMS management strategy chosen by the farmer. Some cases where automatic milking may lead to a higher stress are discussed. In this context, labour psychological studies are needed to learn how the farmer and his family deal with stress situations and how they solve the related problems.

    5. Final conclusions

    - With a simple formula for tuning for robot and cow cycle duration it is possible to show that the capacity of current automatic milking systems can be improved by changing the sequence of certain activities and by programming the simultaneous execution of activities, like the simultaneous opening of the entrance doors of the milking stall and the milking parlour. These adjustments result in shorter idle times for the cows and the robot arm and consequently in shorter milking times.
    - Automatic milking with human-controlled cow traffic is a suitable way of milking. In comparison to conventional milking AM-HCT applied throughout the year and with a milking frequency of three times a day results in marked labour savings for the milking job (37.9%), The AM-HCT method can only be applied on farms with a small herd size (< 50 cows) if only two milking stalls are available. Application of the AM-HCT method on larger farms, requires an extension of the AMS with milking stalls and eventually with a (rotary) robot arm. Therefore, the installation of a milking robot in conventional milking parlours may provide an economically attractive solution for this capacity problem.
    - Automatic milking with computer-controlled cow traffic merely requires starting-up procedures, cleaning tasks and a regular inspection during the day. The AM-CCT method results in a labour reduction of 66.1% for the milking job in comparison to conventional milking. Unexpected failures or repairs were not included in these calculations.
    - The AMS management strategy will determine the absolute and relative importance of the labour requirement for planned and unplanned milking operations. With regard to the amount of labour for the different AMS management strategies under unfavourable circumstances, unplanned milking operations lead to marked reductions in labour savings with the AMS. Therefore, high demands have to be set to AMS functioning and to cow traffic to the AMS.
    - A permanently available maintenance service is very important to assure the quality of the milking process, especially on those farms where the AMS already operates at the limit of its capacity. The results indicate that a training of the farmer in which he is taught the basics to repair small robot failures is a worthwhile investment.
    - For labour organisation at tactical level when using an AMS, we stress that the labour requirement and labour savings will largely depend on the decisions taken by the farmer with respect to the use of contract work, the use of the available farmland and on the milk quota. In all experiments, the labour budget of a robotic milking dairy farm results in labour savings when compared to conventional milking dairy farms. We found labour savings of minimum 1.8% (91 h) for a farming plan with only grassland, summerfeeding and a herd of 80 cows and labour savings of maximum 29.9% (1361 h) for a farming plan with grassland and maizeland, summerfeeding a herd of 70 cows and with contract work for grass and maize silage production.
    - Family farms with up to 80 cows will benefit most of fully automatic milking in terms of labour reduction, especially when summerfeeding is applied and contract workers are hired for grass silage production. It will result in a low labour input throughout the year (slight labour peaks). Other grazing systems will result in more work for the milking job and lower labour savings. When fully automatic milking is applied on large farms, the herd will have to be divided into small groups (40 to 80 cows). summerfeeding can then be applied with or without the employment of contract workers for grass silage production. The size of the farmland will here determine which solution will be the most economical one. If one wants to apply a grazing system in which cows are pasturing, the AMS will need to have a high capacity in order to apply automatic milking with a human-controlled cow traffic. The grazing system of the conventional milking dairy farm can be continued.
    - Automatic milking will contribute to a lower physical and mental load of the farmer and his relatives if problems with cow traffic and technical problems can be kept to a minimum. The farmer will become more an intellectual worker than a manual labourer. More time will be available for animal care and farm management in general. Automatic milking can improve the farmer's social and family life. For certain persons automatic milking might lead to task enlargement and task enrichment, for others however, it might lead to stress situations. Therefore, a labour psychological study to investigate objectively the negative and positive psychological consequences of the robotization for the (potential) robotic milking dairy farmers and their family is recommended.

    Original languageEnglish
    QualificationDoctor of Philosophy
    Awarding Institution
    • Speelman, L., Promotor
    • Maton, A., Promotor, External person
    • Metz, J.H.M., Promotor
    Award date14 Jun 1996
    Place of PublicationWageningen
    Print ISBNs9789054855422
    Publication statusPublished - 14 Jun 1996


    • dairy cattle
    • dairy farming
    • robots
    • milking machines
    • work study
    • organization of work


    Dive into the research topics of 'Labour organisation on robotic milking dairy farms'. Together they form a unique fingerprint.

    Cite this