The increasing amount of data collected in livestock farms thanks to PLF devices represents a huge potential of information to optimise animal husbandry and barn design. With particular reference to dairy cattle farming, Automatic Milking Systems (AMS) and the related devices and control software represent a highly informative source of cow-specific data. Such data can be translated into useful indications for management and design through algorithms able to provide a clear characterisation of the animal within the herd. This study developed and tested a procedure for the comprehensive analysis of AMS-generated multi-variable time-series, with the following objectives: • to define clusters of cows on the basis of daily values of their descriptive parameters analysed over a selected time period; • to permanently monitor the time trend of the descriptive parameters of each cluster; • to identify animals with anomalous scores with respect to the herd. A commercial dairy cattle farm in the Po valley (Italy) was adopted as study case. A methodological approach based on hierarchical clustering has been formulated to best characterise each cow and identify cow clusters. The method proved suitable to identify clusters of cows clearly diversified in terms of parity, body mass, days in milk and daily milk yield, as well as individual outlier animals. The results lend support to cow monitoring and to herd management optimisation. Future developments are represented by the integration of processing algorithms in PLF software.

A methodology for daily analysis of AMS data providing herd characterisation and segmentation

Benni S.
;
Bonora F.;Tassinari P.;Torreggiani D.
2019

Abstract

The increasing amount of data collected in livestock farms thanks to PLF devices represents a huge potential of information to optimise animal husbandry and barn design. With particular reference to dairy cattle farming, Automatic Milking Systems (AMS) and the related devices and control software represent a highly informative source of cow-specific data. Such data can be translated into useful indications for management and design through algorithms able to provide a clear characterisation of the animal within the herd. This study developed and tested a procedure for the comprehensive analysis of AMS-generated multi-variable time-series, with the following objectives: • to define clusters of cows on the basis of daily values of their descriptive parameters analysed over a selected time period; • to permanently monitor the time trend of the descriptive parameters of each cluster; • to identify animals with anomalous scores with respect to the herd. A commercial dairy cattle farm in the Po valley (Italy) was adopted as study case. A methodological approach based on hierarchical clustering has been formulated to best characterise each cow and identify cow clusters. The method proved suitable to identify clusters of cows clearly diversified in terms of parity, body mass, days in milk and daily milk yield, as well as individual outlier animals. The results lend support to cow monitoring and to herd management optimisation. Future developments are represented by the integration of processing algorithms in PLF software.
2019
Precision Livestock Farming 2019 - Papers Presented at the 9th European Conference on Precision Livestock Farming, ECPLF 2019
53
59
Benni S.; Bonora F.; Tassinari P.; Torreggiani D.
File in questo prodotto:
Eventuali allegati, non sono esposti

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/740364
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact