Connected Cow developed by Fujitsu for Japanese livestock farmers predicts oestrus in cattle by monitoring step count data. Agrimetrics is now working with the company to produce an economic model for the UK dairy market
The service works by attaching pedometers to cows which signal when their gait changes, providing an alert for when that animal is going into oestrus. An algorithm is applied to the cow's step data to detect indicative oestrus behavioural changes, such as a steep increase in movement. This helps farmers to help reduce calving intervals. The impact on profits can be significant: improving the detection oestrus in dairy cows by 10% above the national average can improve profitability by 0.97p/litre. A cow is typically only in heat for 8-12 hours and this often occurs at night, so targeting oestrus accurately normally requires continuous observation by the dairy farmer.
"The dairy market in Japan differs from the UK " says Karl Verhulst, IoT Solutions Lead at Fujitsu. "We're looking to validate that the product can be used with UK cows, adapting the solution to a dairy environment. After all, the economics are completely different – with profit at approximately .20p a litre – the margins are very thin."
As part of the project with Agrimetrics, the University of Reading is trialling Fujitsu's pedometers on 25 cows at the University farm. The devices are connected to an on-farm receiver, which transmits data to a cloud service.
Professor Richard Tiffin, Chief Scientific Officer at Agrimetrics, is investigating the accuracy of heat detection and also whether the service can be optimised to provide early detection of other well- being indicators.
Richard comments: "Data is collected on an hourly basis, providing alerts to when a cow is in oestrus. However, milk production can be impacted by mastitis and lameness, which can reduce fertility. Since these conditions are detectable by movement, we can develop an algorithm and utilise machine learning to further expand the benefits of this technology for dairy farmers." The best fertility performance is achieved with a calving to first service interval of 60 days; this requires a 70% heat detection rate. Missing the chance for successful conception carries a financial burden.
The trials of the pedometer at Reading have been promising. Data currently being gathered for analysis includes economic information such as the cost of feed and housing, along with that of missing an oestrus cycle or false positives. Armed with this intelligence Fujitsu will be in a strong position to create a UK business case for its technology.
If a dairy farmer has three cows to serve and does not get them in calf, it will be 21 days before the next fertile period. Delay in calving means a loss of £6.85 income per day for 21 days. For three cows, this is equivalent to £431.55 (Dawson, Farm Northwest).
The Agrimetrics Natural Capital Explorer tool uses a set of metrics that can be applied consistently to different farmed environments. The value of this tool is all the more important given the move by Government to reward farmers for delivering greater public and environmental benefits in the future.
Natural Capital Explorer from Agrimetrics and Natural England
Talk to our experienced team to help you access our data - and build your own in-house data capabilities
Our team of consultants help organisations of all sizes extract the hidden value in their data.