Table 1.

(a) Studies on cattle behavior and activity budget in different livestock systems. (b) Studies on sheep behavior and activity budget in different livestock systems. (c) Studies on goat behavior and activity budget in different livestock systems.

(a)
Aim	Technology	Livestock System	Country	Main Result	Reference
Behavior	Accelerometer	Intensive	United Kingdom	Accuracy of 83% in classifying behavior	[24]
			Australia	Accuracy of 88% to 98% in monitoring licking behavior	[42]
			Australia	4-month-old calves suckled fewer times, but for longer	[73]
			United Kingdom	Classification of rumination, eating, and other behaviors with precision of 0.83	[74]
		Pasture-based	France	The accuracy of prediction of the main behaviors was 98%	[40]
		Semi-enclosed barn	United States	Accuracy of rumination detection was 86.2%	[41]
		Three dairy farms	Italy	Accuracy of behavior detection was 85.12%	[75]
		Dairy farm	Italy	Accuracy of classifying behavior was 96%	[76]
	GPS	Extensive	United States	Cattle followed water more than salt	[3]
			Hungary	Weather fronts affected the herd’s route	[64]
		Pasture-based	Malaysia	Observation of the grazing patterns was accurate	[63]
			England	Cattle tended to favor shorter material during the day and material of higher crude fiber in the evening	[66]
		Commercial farm	Spain	Sensor was able to detect hotspots of dung deposition	[77]
	GPS-GPRS	Extensive	Spain	Distance traveled daily was 3147 m	[65]
	Accelerometer, GPS	Pasture-based	Australia	Description of the animals’ movement and some behaviors was successful	[78]
			Spain	Accuracy of classification of behavior was 93%	[70]
	Accelerometer, RFID	Pasture-based	Australia	Accelerometer correlated highly with the observed duration of drinking events	[79]
	Accelerometer, magnetometer	Intensive	Tasmania	Grazing, ruminating, and resting were identified accurately	[80]
	Accelerometer, cameras	Intensive	China	Accuracy of 94.9% in recognizing behavior	[81]
Sensor evaluation	Accelerometer	Intensive	United States	The correlation was high between results of the sensor and visual observations in monitoring behavior	[7]
			Australia	Heavy breathing detected by the sensor correlated well with visual observations	[82]
			Japan	Precision of classifying behavior was 99.2%	[83]
			Germany	Accuracy was 70.8% in monitoring selected behaviors	[84]
			Germany	Accuracy was 96.2% in monitoring drinking behavior	[44]
			United States	Each sensor had high correlation with visual observations for a specific behavior	[43]
			United States	Accuracy was over 92.2% in monitoring sleep	[45]
			Netherlands	The sensor had a correlation of over 0.85 with the visual observation in monitoring behaviors	[47]
			Netherlands	Sensor’s results and visual observations correlated well for monitoring of behavior	[85]
			Netherlands	Sensitivity was over 96.1% for monitoring of behavior	[86]
		Extensive	Brazil	Over-sampling increased accuracy in prediction of grazing behavior	[87]
			Kenya	The harness was more accurate	[46]
		Pasture-based	United States	RumiWatch had accurate results for the studied behaviors	[20]
			Ireland	MooMonitor+, RumiWatch, and visual observation had high correlation for measurement of grazing behavior	[88]
			Australia	Accuracy was 95% to 98.8% in measuring suckling behavior	[18]
			Germany	Rumination and eating behavior were monitored accurately	[89]
			Australia	Grazing, resting, and ruminating were accurately detected	[90]
		Loose-house system	Denmark	The AfiTagII correlated very highly with direct observations and IceQube recordings in monitoring lying behavior	[91]
		Housed in an outdoor dirt floor pen	Canada	Sensitivity and specificity were 95% and 76% for feeding and 49% and 96% for rumination	[92]
	GPS	Pasture-based	United States	The Clark ATS provided real-time tracking	[68]
	Pedometer	A 0.2-ha sown pasture	Japan	Correlation coefficients between the pedometer values and the number of bites were all over 0.9	[8]
		Pasture-based	United States	Distance traveled increased with larger pasture	[6]
	Accelerometer, GPS	Intensive	United Kingdom	Accuracy was 80.8% to 94.2% in detecting variations in feeding behavior	[93]
		Pasture-based	United States	Patterns of behavior were accurately identified	[72]
			United States	Time spent grazing from 8.67 to 10.49 h daily	[94]
	Accelerometer, pedometer	Extensive	Italy	Accelerometer and direct observations for ruminating, feeding, standing, and lying correlated well	[95]
Health and welfare	Accelerometer	Intensive	New Zealand	Change in behaviors began 4 days before the diagnosis	[49]
			Denmark	Lying duration increased by 40 min but walking decreased for lame cows	[19]
		Intensive system with constant access to pasture	United States	The diseases had negative effects on ruminating and walking	[96]
		Rotational grazing system	Australia	24 h before the symptoms, heifers moved less	[12]
		Pasture-based	Germany	Associations found between sensor behavior traits and monitored cow behavior	[48]
	Pedometer	Individual pens (3 m2) in a calf barn	United States	Activity drop before the diagnosis	[2]
Estrus and calving	Accelerometer	Pasture-based	United States	100% sensitivity, 86.8% specificity in detecting changes in behavior	[51]
			New Zealand	Monitoring of behavior was successful	[97]
		Free-stall barn environment	Belgium	Performance increase with more sensors	[50]
		Lactating cows were housed in 2 free-stall pens	United States	Sensors were at least as successful as visual observation in detecting estrus	[98]
	Pedometer, accelerometer	Dairy cattle farms	Germany and Italy	Estrus detection was accurate	[99]
	GNSS	Commercial farms	Spain	Sensor provided indicators on the occurrence of calving	[100]
	Accelerometer, GNSS	32 ha paddock	Australia	Accuracy of 98.6% in calving detection	[101]
Bite rate	Accelerometer	Intensive	Australia	Semi-supervised linear regression model predicted bite rate well	[102]
(b)
Aim	Technology	Livestock System	Country	Main Result	Reference
Behavior	Accelerometer	Extensive	New Zealand	Accuracy of 89.6% for grazing, walking, and resting	[52]
			Wales	Accelerometers correlated perfectly with video observations for lying behavior	[9]
			Poland	Suckling episode detection rate of 95%	[53]
		Pasture-based	Australia	5 s time interval was best in identifying biting and chewing	[54]
		A rectangular field of 110 × 80 m	Denmark	Classification of behavior success was 74.8% for the entire herd	[55]
		Sheep alternating between intensive and extensive system	Italy	Accuracy of 93% in prediction of bite rate	[103]
	GPS	Extensive	Canada	Livestock’s presence had an effect on bighorn sheep’s behavior	[67]
	Accelerometer, gyroscope	Three pasture paddocks of 72 m2	Australia	Behavior classification had accuracy of 87.8%	[104]
Sensor evaluation	Accelerometer	Extensive	Italy	Collar attached was the best with accuracy of 90%	[57]
			Wales	100% accuracy for urination events	[58]
			Australia	Accuracy was best (87%) for the leg deployment	[13]
		Pasture-based	Australia	Ear-mounted sensor was the most accurate with 86% to 95%	[56]
		Semi-improved pasture for the 1st study and a small pen in the 2nd study	Australia	Grazing behavior was the easiest to detect	[105]
		Pasture-based but they were gradually removed from pasture	Italy	The device performed well and the number of bites was accurate	[106]
Parturition and sexual activity	Accelerometer	Intensive	United States	Accuracy of behaviors was 66.7%, and that for activity was 87.2%	[59]
		Extensive	New Zealand	Ewes were more restless around parturition	[60]
		Pasture-based	Spain	Sensitivity for mounting detection was 77.9% and for service detection was 94%	[17]
	GNSS logger, accelerometer	Extensive	New Zealand	Detection of parturition events and lambing activity was accurate	[15]
Effects of grazing on vegetation	GPS	Extensive	Spain	Grazing, depending on its intensity, may benefit or not the pastures	[25]
Health and welfare	Accelerometer	5.5 ha paddock	Australia	Accelerometer-based sensor can identify changes in sheep activity associated with H. contortus infections	[107]
(c)
Aim	Technology	Livestock System	Country	Main Result	Reference
Behavior and activity	GPS, accelerometer	Extensive	Morocco	Sensors monitored accurately the grazing activities of dairy goats	[1,108]
			Morocco	Sensors monitored accurately the grazing activities of meat goats	[23]
		Pasture-based	Germany and Oman	Recognition of eating 87% to 93%, 68% to 90% for resting, and 20% to 92% for walking	[109]
	Accelerometer, gyroscope	Extensive	Argentina	Prediction of behaviors had precision of 85% and recall rate of 84%	[110]

GPS, global positioning system; GPRS, general packet radio service; RFID, radio frequency identification; GNSS, global navigation satellite system; the number of animals (cattle) in the research varied between 3 and 348; the number of animals (sheep) in the research varied between 1 and 96; the number of animals (goats) in the research varied between 1 and 8; Intensive system, continuous supplementation of animals by cereal-based feed or industrial supplements is the standard; extensive system, mainly involves small ruminants and resource-constrained breeders, depending on rangeland; pasture-based, a system that relies significantly on pastures, which include grasses, legumes, and herbs [111].