Precision livestock farming (PLF) can be defined as real-time monitoring technologies aimed at managing the smallest manageable production unit’s temporal variability, known as ‘the per animal approach’. The first massive application of PLF, years before the term PLF was coined in 2004 (Berckmans, Reference Berckmans2004), was the individual electronic milk metre for cows. The first milk metres became commercially available in the 1970s (Peles, Reference Peles1978) and early 1980s (Brayer, Reference Brayer1982), followed by commercialised behaviour-based oestrus detection and later still, rumination tags and an online real-time milk analyser (Schmilovitch et al., Reference Schmilovitch, Katz, Maltz, Kutscher, Sarig, Halachmi, Hoffman, Egozi and Uner2007). In this special issue on PLF, Steensels et al. (Reference Steensels, Antler, Bahr, Berckmans, Maltz and Halachmi2016) make use of these milk and behaviour parameters to detect post-calving diseases. Further accuracy to predict individual cows’ feed intake is reached by employing different sensors (Pahl et al., Reference Pahl, Hartung, Grothmann, Mahlkow-Nerge and Haeussermann2016) and by adding feeding behaviour to a feed-intake model (Halachmi et al., Reference Halachmi, Meir, Miron and Maltz2016).
The milking robot is a classical PLF application: the smallest manageable production unit in this case is one single quarter of an udder. Simulation-optimisation based on animal behaviour was developed for robotic milking farms at the beginning of this century (Halachmi, Reference Halachmi2004). In this PLF issue, John et al. (Reference John, Clark, Freeman, Kerrisk, Garcia and Halachmi2016) review the utilisation of robots after a farm was designed and a robot operated. They also review the robot milking–pasture combination. Continuing with dairy in this issue, Salau et al. (Reference Salau, Haas, Thaller, Leisen and Junge2016) develop a Kinect-based system, and Kinect is applied by Van Hertem et al. (Reference Van Hertem, Bahr, Tello, Viazzi, Steensels, Romanini, Lokhorst, Maltz, Halachmi and Berckmans2016) for automatic lameness detection. Another approach for automatic lameness detection, gait behaviour and ground reaction forces, is proposed by van Nuffel et al. (Reference Van Nuffel, van De Gucht, Saeys, Sonck, Opsomer, Vangeyte, Mertens, De Ketelaere and Van Weyenberg2016). IR-thermography-based monitoring of body temperature of calves (Hoffmann et al., Reference Hoffmann, Schmidt and Ammon2016) and gene expression of calves undergoing gradual weaning (Johnston et al., Reference Johnston, Kenny, Kelly, McCabe, McGee, Waters and Earley2016) are presented in this special issue.
In meat production, one of the first PLF applications was related to individual feed distribution in a group housing (Marcon et al., Reference Marcon, Brossard and Quiniou2015). In this special issue, we see monitoring of drinking behaviour of individual pigs housed in a group using radio frequency identification (Maselyne et al., Reference Maselyne, Adriaens, Huybrechts, De Ketelaere, Millet, Vangeyte, Van Nuffel and Saeys2016). We also see the development of automatic surveillance of animal behaviour and welfare using image analysis and machine-learned segmentation technique (Nilsson et al., Reference Nilsson, Herlin, Ardö, Guzhva, Åström and Bergsten2015).
Broiler chickens are usually fast-growing, bred intensively with up to 40 000 conspecifics. The farmers’ challenge is to reach high final BW in a short time while maintaining an efficient feed-conversion rate. Therefore, one of the first PLF applications in broilers was to automatically monitor the animals’ weight (Fontana et al., Reference Fontana, Tullo, Butterworth and Guarino2015). Another application in this issue offers vocalisation sound pattern identification in young broiler chickens (Fontana et al., Reference Fontana, Tullo, Scrase and Butterworth2016).
Beyond these achievements, the integration of PLF into scientific communities is moving forward: the EAAP (European Federation of Animal Science), with its annual meeting in Copenhagen in August 2014 was, to the best of our knowledge, the first animal science federation to host an international symposium on PLF. The EAAP facilitated ‘cross-disciplinary’ discussions focussing on interpretations of sensed animal responses and the associated management actions. Several livestock sectors participated in the discussions: (1) providers of animal-sensing technology such as start-up companies and sensor developers, (2) mature industries such as retailers, animal feed suppliers, farm equipment providers, farm designers and veterinarians, covering the animal and human food chains, and (3) animal geneticists, nutritionists, health experts, zoologists, biologists, environmental scientists, i.e. animal-focussed scientists and the farmer organisations that usually attend EAAP annual meetings. The ‘questions and answers’ debates inscribed during this joint session’s discussions are presented in a book (Halachmi, Reference Halachmi2015), and the selection of full-length papers are presented in this special issue of Animal. This special issue covers the next generation applications. Its content provides evidence of the initial integration of PLF into the community of animal scientists, with a widening and deepening of research, development and evaluation of underlying concepts of PLF for a vast and diverse world of livestock production. Potential applications include individual animal food intake, automatic early detection of illness or stress and monitoring of animal welfare. The prospects for further developments are manifest.
