Application of passive RFID to monitor the feeding behaviour of broilers

Monitoring the feeding behaviour of broilers is important to study the effects of changes in the management. Different diet types (conventional and a more circular/sustainable diet) were compared in an experiment. During 6 rounds (1 test round and 5 experimental rounds), two groups of 500 broilers per round were reared on an experimental farm, one group with conventional diet and one with circular diet. In every round 30 animals per group were equipped with an LF RFID tag, attached to the leg with an elastic. The elastics were attached on the day of arrival and removed when the broilers left the farm. The elastics were widened when necessary to keep up with the increasing leg circumference. The RFID tags were read by circular antennas placed at ground level around the feeding bins. The readings of a broiler might then be indicative for the feeding behaviour. Video recordings were analysed to test this hypothesis: 18 observation periods in the test round and 48 periods of approx. 10 minutes in the third experimental round. The objective of this paper is to describe the analysis of the RFID readings and to present the results of the comparison of the calculated behaviour with the observed behaviour for validation of the use of RFID tags to detect feeding behaviour. RFID readings were combined into visits if the interval between two successive readings of the same antenna was less than the visit criterion (20 seconds). Subsequently, visits were combined into meals if the interval was less than the meal criterion (247 seconds). The resulting meals were compared with the recorded events manually observed in the selected periods. For the 48 periods in round 3, this resulted in a sensitivity of 89.0% (243 of the 272 events could be matched with meals) and a precision of 92.3% (215 of the 233 meals could be matched with an observed event). These results indicate that RFID readings can be used to monitor the feeding behaviour of broilers. Based on the RFID results, it was difficult to draw conclusions on the effect of diet as in this experiment the diet effect was confounded with the pen effect.