Encyclopedia
Precision livestock farming (PLF) technologies have been developed with the intention to improve farm management and minimize aversive handling practices. Precision livestock farming systems could serve as useful support tools for the farmer’s decision making and improve the sustainability and competitiveness of dairy farms through the implementation of automated procedures that minimize the labour demand, animal disturbances and environmental impact.
- precision livestock farming systems
- animal welfare
- milk production
1. Introduction
Precision livestock farming (PLF) technologies have been developed with the intention to improve farm management and minimize aversive handling practices. According to Berckmans [1][28], a PLF system: (a) is a support tool that includes cameras, microphones and other sensors for tracking livestock, as well as computer software, and could improve the production efficiency through the adoption of electronic data collection, processing and application, but does not intend to replace the farmer; (b) is an animal-centric tool—the animal is the main part of the process; and (c) needs ideal conditions for the monitoring and control processes. For example, PLF systems have revolutionized the milking process through the introduction of automatic milking robots, leading to an improved quality and increased quantity of milk, while welfare status is maintained at high levels [2][29], since each individual animal can choose their preferred time of being milked [3][30]. PLF are real-time monitoring technologies with the core purpose to ‘manage even the slightest manageable production unit’s temporal variability (i.e., per animal approach)’ [4][5][31,32]. They are frequently integrated with other new technologies in order to improve the human–livestock interactions, productivity and economical sustainability of modern farms [6][33]. In early years, milking robot machines were mostly used for indoor farming. However, automatic milking systems could also be applied in semi-grazing and pasture-grazing farms [7][8][34,35].
2. Precision Livestock Farming (PLF) Advancements in Dairy Production
2.1. Dairy Cattle
23.1.1. Assessment of Health Status
The most common health issues in the dairy industry are mastitis and lameness. The presence of these diseases can have a serious impact on a unit’s everyday processes, damaging the health and welfare of the animals and the production quality and quantity—hence negatively affecting the economy of the unit [14][141]. The economic loss can be traced to a decreased milk yield [15][16][17][107,142,143], reduced reproductive performance [18][19][144,145] and increased culling risk [20][146]. A variety of PLF technologies have been introduced for the monitoring and early detection of such problems [21][4].Lameness Detection
Lameness is among the top three health-related causes of economic loss in the dairy industry. Animals suffering from lameness demonstrate a reduced mobility, milk yield, reproductive ability and loss of body condition and feel intense pain [18][144]. Several models have already been developed for lameness detection by monitoring individual animals’ locomotion and walking patterns and have been proven to be very promising for assessing the problem. Pastel and Kujala [22][104] developed a four-balance probabilistic neural network model based on weekly measurements of the leg load (i.e., leg weight pressure) of 73 cows over a 5 month period. They reported a score of 100% lameness validation and 96.2% classification success.Mastitis Detection
Milking dairy cows has been a common process since at least 3100 BC, and bovine mastitis disease has probably existed since the same era [23][149].2.2. Small Ruminants
Small ruminants are often managed as a flock/herd, allowing only average welfare states to be considered. Innovative technologies provide a unique opportunity to monitor and improve welfare management from the farm-level manual to automated or semi-automated assessment and management at an individual level, leading to a reduction along the value chain (https://techcare-project.eu, accessed on 7 September 2021) of on-farm labour and veterinary care costs [34][156]. However, PLF technologies and applications are not common in small ruminant farming, since most sheep and goat farmers’ acceptance and adaptation of the new technologies remains considerably low, and most technologies and PLF applications are therefore mainly used for research purposes [6][33].Small ruminant farms are often located in remote and mountainous areas, where infrastructure and services are poorly developed. High equipment costs, a lack of demonstration and specific training and low education levels also contribute to the delay of PLF’s implementation. At the same time, there are also practical issues related to the productive cycle of small ruminants that present a barrier to the application of PLF advancements. For example, although oestrus detection is of vital importance in dairy cattle and the main driver of PLF applications for cows, it is not such a priority for small ruminant farmers [9][80]. According to Wishart [35][157], PLF technologies would be more attractive to sheep and goat farmers if researchers focused on demonstrating the beneficial impact these systems have on welfare and efficiency, leading to more sustainable and profitable units. Table 12 presents PLF applications used in research for minimizing or even solving a variety of different problems in contemporary sheep and goat farms.
| Parameter of Interest | Applied Technology | Reference | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Grazing and ruminating behaviour | Animal-mounted accelerometer/gyroscope sensor | [36] | [158] | ||||||||
| Animal-mounted tri-axial accelerometer loggers | [37] | [159] | |||||||||
| Resting, grazing and searching behaviours | Animal-mounted tri-axial accelerometer loggers | [37] | [159] | ||||||||
| GIS systems | [38] | [160] | |||||||||
| Animal-mounted GPS sensors | [38][39][40] | [160 | [ | ,161 | 41 | ,162 | ] | ,163] | |||
| Animal tracking | Animal-mounted GPS sensors | [38][39][40][41] | [160,161,162,163] | ||||||||
| Animal-mounted tri-axial accelerometer loggers | [42] | [164] | |||||||||
| Sexual behaviour of rams | Animal mounted accelerometers | [43] | [165] | ||||||||
| Feeding behaviour | Camera-based analysis | [39] | [161] | ||||||||
| Microphones | [39] | [161] | |||||||||
| Animal-mounted gyroscopes | [39] | [161] | |||||||||
| Animal-mounted accelerometers | [39][44][45] | [161,166,167] | |||||||||
| GPS sensors | [46] | [168] | |||||||||
| Oestrus detection | Alpha-D detector | [47][48] | [169,170] | ||||||||
| Infrared thermography | [49][50] | [171,172] | |||||||||
| Lameness detection | Infrared thermography | [51] | [173] | ||||||||
| Hoof weigh crate with four load platforms | [52] | [174] | |||||||||
| Lambing detection | Animal-mounted temperature sensors | [53] | [175] | ||||||||
| Health status detection | Implanted sensors (heart rate and body temperature) | [54] | [176] | ||||||||
| Individual identification | Injectable transponders | [55][56] | [177,178] | ||||||||
| RFID sensors | [34][57][58] | [156,179,180] | |||||||||
| Endoruminal bolus | [58]59] | [180 | [ | ,181] | |||||||
| Drones—image analysis | [60] | [182] | |||||||||
| Age identification | Sound recorders analysis | [61] | [183] | ||||||||
| Flock monitoring | Drones—image analysis | [60][62] | [182,184] | ||||||||
| Weight monitoring | Automatic weigh-crates | [35] | [157] | ||||||||
| Standing/lying behaviour monitoring | Camera-based analysis | [63] | [185] | ||||||||
| Ultra-wide band real-time location | [63] | [185] | |||||||||
| Animal mounted accelerometers | [64] in goats | [186] in goats | |||||||||
| Flock management | Virtual fence (i.e., animal-mounted collars embedded with electromagnetic transmitters and ground-installed receivers and sound speakers) | [65][66[70]; [71] in goats | [187 | ][67][68] | ,188 | [ | ,189 | 69 | ,190 | ] | ,191,192]; [193] in goats |
- (a)
-
Automatic vacuum shut-off (AVSO), which is a mechanism, either time- or flow-based, that prevents overmilking and its negative impact on the animals’ health and welfare. Therefore, it improves the sanitary status of the milking parlour, while at the same time reducing the labour required. Bueso-Ródenas et al. [73][195] and Romero et al. [74][196] reported that the flow-based AVSO is better than manual milking, as the same amounts of milk were extracted in shorter time intervals. Furthermore, they proposed that the best combinations of the flow limit (i.e., the time interval during which the vacuum shut-off is activated) and delay time (i.e., the minimum flow set, such that below that point the vacuum shut-off is activated) for sheep are 150 g/min and 20 s or 200 g/min and 10 s, respectively, and that for goats it is 100–150 g/min and 10 s, respectively.
- (b)
-
Milk meter and flow indicators, which are sensors that allow the monitoring of the milk flow of every individual animal. Electronic milk meter measurements are based on the combined data of infrared and conductivity sensors and/or a volume measuring chamber [72][194]. The data is analysed and presented on the display of a personal computer. Furthermore, electronic milk meters have the ability to sample and analyse milk, providing information on the animals’ health status [72][75][194,197]. Therefore, milk and flow meter applications are an essential decision-making tool for the farmer.
- (c)
-
Electronic identification assessment, which is performed by various sensor-based applications such as injectable transponders [55][56][177,178], RFID [34][57][58][156,179,180], endoruminal boluses [58][59][180,181] and drones [60][182]. These systems carry individual information concerning the age, weight, gender, health status and milk flow of every animal. The producer can keep a catalogue of all individual animals in the flock, and thus, they are a very useful long-term decision-making tool. It should be noted that to date mostly the RFID technology is used, as the transmitters attached on ear tags or foot are read from the receiver installed in the milking parlour and therefore the data flow can be accessed remotely in real time.
- (d)
-
Automatically operated sorting gate and weighing scale systems are connected to a flock management software, which sorts the animals into groups or modifies existing groups and separates the animals in need of treatment. These systems minimise both the labour and time spent regrouping and relocating the animals [72][194], practices that are commonly applied to obtain uniformity within groups in terms of milk yield [76][198].
