AMS

In Norway, it has been common practice to collect large amounts of data for breeding purposes and health monitoring of dairy cows. The use of automatic milking systems (AMS) and related sensors has increased the amount of available data. Through innovative and advanced data analyses (AI), we aim to use this information to improve the health and productivity of cows, which also contribute to reduce the climate footprint of Norwegian milk production.

10 Jan 2024 - 09 Jan 2027

FFL- JA – Forskningsmidlene for jordbruk og matindustri, TINE, Geno og Mimiro

Project information

  • Background

    The iKu project puts forward our experiences from the AMS project, which demonstrated that effective utilization of data from farms with automatic milking systems (AMS) requires innovative approaches to data analysis. The iKu-project has a total budget of 19.3 million NOK (Research Council of Norway project no. 344157) and aims to establish bioindicators and phenotypes for health and fertility in Norwegian Red cattle using traditional mathematical / statistical methods along with artificial intelligence (AI).

    In most Norwegian dairy herds, milk samples are obtained monthly to undergo a spectroscopic analysis of the milk's composition. Previously, we have used parts of these spectra to calculate energy balance in cows and studied associations with subclinical ketosis and impaired reproductive performance (AMS project). The iKu project also aims to develop methods for early detection of subclinical mastitis using FTIR spectral data. We will calibrate the FTIR spectra against results from differentiated cell counting and bacteriological analysis of milk samples.

    With the iKu-project, we aim to develop new biological indicators to detect undesirable health and production events at an early stage. In addition to the spectroscopic analyses, we will include health and production data, as well as information from AMS and sensors in our prediction models. Furthermore, we aim to develop new phenotypes based on bioindicators for subclinical ketosis, udder health, and fertility. The genomic breeding value of these phenotypes will be evaluated using Geno's SNP-chip genotypes.

    We have assembled an interdisciplinary research team that includes experts in mathematical sciences, animal science, and veterinary medicine at NMBU. This team will play an important role in establishing a platform for the collection, analysis, and utilization of large datasets in collaboration with industry organizations Geno, TINE, and MIMIRO

  • Objectives

    The overall objective of the project is to to establishing new bioindicators for health and reproduction in Norwegian dairy cows by the use of innovative data analysis methods to analyse production records, milk analyses, robots, and sensors. These bioindicators will subsequently be evaluated as phenotypes for genomic selection.

  • Participants

Marit Smistad

TINE SA

Karoline Bakke

Geno

Torgeir Wiik

Mimiro

News

AMS-prosjektet