Smart farming: empowering livestock farmers with data

Bennie van der Fels, senior project manager in the department of Animal Health and Welfare at Wageningen University & Research, explains why smart farming has become increasingly important over the years. “The number of animals per farm has grown, so there’s a need for resources to help with monitoring the health and behaviour of those animals. The general public expect high standards of transparency too, which adds to the need for data.”

“There’s also been progress in sensor technology development, so affordable sensors and cameras are readily available to livestock producers,” says Van der Fels. “Sensors are a way of empowering the farmer. They’re a type of extra antennae for monitoring the animals,” he adds. But a fully robotic system wouldn’t work for the livestock sector, he says. Technology and robotics play an important role in farm management, but the expertise of the farmer is irreplaceable.

Existing technologies

Sensor technology is nothing new in the context of climate control in pig and poultry sheds. “The development of sensor technology over the past few years means it has become increasingly useful for monitoring animals’ living environment, including factors such as temperature, humidity and ammonia levels,” says Van der Fels. The emergence of smart cameras adds a new dimension to this, by enabling the monitoring of animal behaviour. These images can be combined with the climate control readings to generate new data which can lead to new insights. This kind of data mixing is already being applied to the poultry sector. Images showing the distribution of the animals in a shed can be used to promptly flag up any environmental abnormalities. This gives the livestock keeper the option of intervening immediately as a way of maintaining high standards of animal health and welfare.   

Algorithms provide insights into social behaviour

Cameras enable 24/7 monitoring of animals. This makes it easy to observe how cows and pigs stand, walk and lie down across a longer time frame, for example. Combining historic and real-time data on living environments and animal behaviour enables computer models to issue timely alerts if there are any abnormalities in animal health and welfare. “We definitely still need human input for the development of these algorithms,” says Van der Fels. “We’re participating in a number of research projects to develop reliable reference frameworks and threshold values for the development of the algorithms. From a scientific point of view, the challenge is really to analyse and interpret the social behaviour of the animals using images and sensor data for individual animals and herds as a whole, and to translate this into the necessary interventions."

Transition phase

Notwithstanding the rapid pace of development, many producers aren’t quite ready to adopt smart farming. “We’re in a transition phase at the moment,” says Van der Fels. “Pig farming is increasingly moving towards individual monitoring. This is made possible by combining sensor technology with electronic animal recognition (RFID: Radio Frequency Identification technology). A large group of pig farmers who are part of the Sustainable Pork Value Chain Association are already doing this. At birth, the pigs are given an ear tag containing a chip. The chip is tiny and doesn’t bother the animal, but it can collect a wealth of data about that individual pig. It’s a way of ensuring that antibiotics are kept out of the supply chain, for example, and it enables an upstream data flow with detailed information from the supply chain back to the pig farmer. This can help the farmer analyse outcomes from a technical and health perspective. The system can therefore help safeguard quality standards in the supply chain."

It’s a great example of what’s possible. More research is now needed on how to fine-tune and link up the data sets from different sources, bearing in mind the need to safeguard the privacy of the livestock keeper too. Fieldwork is currently being done on this, for example through the JoinData organisation. This independent data platform enables livestock farmers to establish agreements relating to data ownership and use.

The frequent use of cameras and the storage of the images taken is also an important issue. Supply chain partners will need to work together over the next few years to organise suitable infrastructure for this and to make agreements on the sharing of different data sets. And of course we can’t overlook the education aspect either. "It’s important that young producers are familiarised with the potential of smart farming and that it’s integrated into the learning curriculum,” he concludes.