ICT monitoring and mathematical modelling of dairy cows performances in hot climate conditions: a study case in Po valley (Italy)

Authors

  • Filippo Bonora University of Bologna
  • Matti Pastell Natural Resources Institute Finland (Luke)
  • Stefano Benni University of Bologna
  • Patrizia Tassinari University of Bologna
  • Daniele Torreggiani University of Bologna

Keywords:

Precision Livestock Farming, THI (Temperature Humidity Index), Productivity, Dairy Cow

Abstract

Automatic Milking Systems (AMS) measure and record specific data about milk production and cow behaviour, providing farmers with useful real-time information for each animal. At the same time, indoor climatic conditions in terms of temperature and humidity within a dairy livestock barn represent a well-known crucial issue in farm building design and management, since these parameters can remarkably influence cows behaviour, milk yield and animal welfare.

The goal of the study is to develop and test an innovative procedure for the comprehensive analysis of AMS-generated multi-variable time-series, with a focus on the analysis of the relationship between milk production and indoor climatic conditions. The specific purpose of the study is to develop and test a mathematical computer procedure using AMS-generated data and environmental parameters, designed to provide a forecasting model based on the integration of milking data and temperature and humidity levels surveyed from local sensor grids, designed to model milk production scenarios and, specifically, yield trends depending on the expected environmental conditions.

For this purpose, a typical Italian farm with AMS has been adopted as a study case and internal climatic data of the barn have been analysed to understand the influence of high values of the Temperature Humidity Index (THI) on milk production in time. Then the correlation between yield variations and THI has been computed and characterized. Finally, external climatic data have been used to forecast the milk production in summertime. Once the model was validated, tests has led to predict milk yield with a relative error smaller than 2%.

This study represents a step of a research aimed to define integrated systems for cow monitoring and to develop guidelines for the optimization of barn layouts.

Downloads

Published

2018-09-18

Issue

Section

Special Issue: Animal housing in hot climate