Precision livestock farming has become a buzzword in pig husbandry and nutrition, since swine farms can only be operated profitably and sustainably in the long-term if precision technologies are applied.
Theoretic background for precision swine farming
Precision swine farming (similar to precision livestock farming) means adopting a keeping, feeding and management system based on the use of IT technologies that makes it possible to care for the animals at an individual level, to recognise problems early and provide effective solutions even on large-scale farms (Halas, 2017). Real-time data collection is a key element of precision swine farming, as this provides information regarding the behaviour, health and feed intake of the individual animal as well as it’s environment (e.g. ambient temperature, humidity, air quality). This allows for the use of different resources based on the individual needs of the animals and optimisation that promotes sustainability. Data (numbers, images, sounds) recorded by precision technology are used to support production by first generating a dataset that is filtered and interpreted by different algorithms. Filtered and processed data are then integrated into the production process. This will support decision making either following individual assessment or based on an automated process. Accurate data (and the everyday experience of operation) can also support objective decision-making regarding costly investments that often also affect farming technology and long-term strategic planning.
Precision swine farming in Hungary
The development of precision tools (along with the developments in IT) is an extremely dynamic process, new solutions are appearing on the market continuously and the bottleneck of investments are on-farm facilities and the lack of motivation. Figure 1. Summarises the current willingness of Hungarian swine farms to invest.
Optimal environmental conditions must be insured to be able to fully use the genetic potential of the animals, making this area a focus point regardless of the age of the animals. Typically producers will invest in RFID chips for the individual identification of sows. Some effort has been made to use this technology in fatteners, since integration with identification at the slaughterhouse would allow for lowering the potential for errors and the number of complaints. Water intake is one of the basics of sustainable farming but still doesn’t receive enough attention, even though using drinking systems that are appropriate for the animals’ needs and the management system could significantly reduce water wastage characteristically seen in the farrowing and the fattening units. Willingness to invest in optimising feed intake is ranked second after willingness to ensure optimal environmental conditions, which is no surprise regarding the fact 70% of production costs can be attributed to feed cost. Investments in feeding technology are characteristically seen at the farrowing and pregnant sow unit, since this is where all the animals can be identified individually. We have to note that this would be highly relevant for fattening pigs also, but currently there are no tried and proven technologies for use at the individual level in fattening units in Hungary. Behaviour and health monitoring are closely related areas and their relevance has increased with regards to zinc oxide free feeding of weaned piglets and the measures aimed at reducing antibiotic use. Nevertheless willingness to invest in this field is still quite low in Hungary.
It is worth noting that the use of artificial intelligence is gaining importance in almost all of the above mentioned areas. There are some camera systems in place trained by machine learning methods that are capable of counting fattening pigs with 100% accuracy under farm and slaughterhouse conditions, reducing the number of potential human related errors. There are also some systems in use for real time prediction of weight for groups or individual animals. Bodyweight prediction has a 99% accuracy for groups, while the bodyweight of individual animals can be predicted with a relative variance of 2 to 4 kg. This solution supports decision-making in loading and the selection of the animals. There are also AI systems available for precision feeding that are capable of monitoring daily weight gain with an accuracy of 96 to 97%. This also provides an opportunity to monitor and optimise the feeding strategy.
Precision swine farming: the vision
It is clear from the review of the current situation that the use of precision tools is only effective and operational for a few resources in Hungarian practice due to the limited possibilities. The vision for the future includes the expansion of both resources and the tools utilised. Detailed calculations on the return on investments or commitment to innovation and sustainability might be the drivers for pig farmers in achieving this goal. Currently we have the least experience with AI based solutions, even though they will play an increasing role in sustainable agriculture as well as our every day lives in the future. Its significance in precision swine farming is not only in local data processing but also in providing integrated data services, allowing for a more efficient cooperation between integrations and their partners through more accurate planning of production and the monitoring of live animals. It is also worth mentioning that most Hungarian livestock farms are short of workers, which will increase the value of utilising AI solutions in the future.
In summary we can say that the market of precision tools is dynamically expanding with the IT developments but investments are limited due to a lack of appropriate capabilities and motivation. This means there is still a long road ahead of us but the future vision is to expand the range of precision tools (including artificial intelligence) used on swine farms to support long-term operations that fulfil both profitability and sustainability aspects.
Dr. Orsolya Vida
Senior product manager
Bonafarm-Bábolna Takarmány Ltd.
Gergely Boromisza
Sales executive
Pig Brother Ltd.
