NU Smart Farms

Quantify lying quality in sows

Background

The problem

Sows are the main cause of crushing their neonatal piglets.

It would be beneficial to capture sow traits influential for piglet survival, such as those associated with lying behaviour.

Results

The solution

An algorithm to detect sow posture and posture transitions (lying quality) can be developed. The data would be provided by tri-axial accelerometers attached to the sow.
This enables large-scale phenotyping.

Following on from the phenotyping, we can identify whether these traits are genetically heritable.

Potentially, this would allow for genetic selection against ‘risky’ (high incidence of crushing) posture transitions.

Additional applications

The technology could also be used to:

  • detect problems with the farrowing floor
  • highlight interactions between floor surfaces and incidences of piglet crushing
  • detect farrowing onset by the increase in activity levels in the accelerometer records