Author:

Abstract:

The activity and distribution patterns of chickens are measured to check the performance and status of the flock using Precision Livestock Farming (PLF) technologies. The aim of this study is to illustrate how monitoring flock activity and distribution, provides retrievable infor-mation on broiler habits and behavioural patterns along one cycle and comparison between cycles. Four top view cameras are used in one commercial broiler farm in The Netherlands to visualize the floor area, continuously recording im-ages during the light periods of the cycle which are translated into ac-tivity and distribution indexes per minute. The monitoring of these time-series through successive light periods reveals how behavioral patterns of the flock, such as feeding, drinking or resting, evolve along the cycle. Forecasting this evolution allows to raise alerts when devia-tions from the predicted levels are higher than 20%. At the same time, the definition of a regular pattern for each behaviour using Single-Input Single-output (SISO) Transfer Function (TF) models for activity and distribution also allows to raise alerts when the model performance is showing a low fitting agreement, lower than 70%, in a light period. Linking these alerts to welfare quality assessment scores shows how variations in the activity or distribution patterns affect broiler health and welfare.