Comparing the Performance of Automatic Milking Systems through Dynamic Testing Also Helps to Identify Potential Risk Factors for Mastitis.

Automatic milking systems (AMSs) are revolutionizing the dairy industry by boosting herd efficiency, primarily through an increased milk yield per cow and reduced labor costs. The performance of milking machines, whether traditional or automated, can be evaluated using advanced vacuum meters through dynamic testing. This process involves scrutinizing the system and milking routine to identify critical points, utilizing the VaDia™ logger (BioControl AS, Rakkestad, Norway). Vacuum recordings were downloaded and analyzed using the VaDia Suite™ software under the guidance of a milking specialist. Access to data from AMSs across various manufacturers and herds facilitated a retrospective study aimed at describing and comparing key milk emission parameters for different AMS brands while identifying potential mastitis risk factors. Using the proper statistical procedures of SPSS 29.1 (IBM Corp., Armonk, NY, USA), researchers analyzed data from 4878 individual quarter milkings from cows in 48 dairy herds. Results indicated a significant variability in milking parameters associated with quarter milk yield and AMS brand. Notably, despite AMSs standardizing teat preparation and stimulation, this study revealed a surprisingly high frequency of two major mastitis risk factors-bimodality and irregular vacuum fluctuations-occurring more frequently than in conventional milking systems. This study, one of the few comparing different AMS brands and their performance, highlights the crucial role of dynamic testing in evaluating AMS performance under real-world conditions.

Evaluating the performance of herd-specific Long Short-Term Memory models to identify automated health alerts associated with a ketosis diagnosis in early lactation cows.

The growing use of automated systems in the dairy industry generates a vast amount of cow-level data daily, creating opportunities for using these data to support real-time decision-making. Currently, various commercial systems offer built-in alert algorithms to identify cows requiring attention. To our knowledge, no work has been done to compare the use of models accounting for herd-level variability on their predictive ability against automated systems. Long Short-Term Memory (LSTM) models are machine learning models capable of learning temporal patterns and making predictions based on time series data. The objective of our study was to evaluate the ability of LSTM models to identify a health alert associated with a ketosis diagnosis (HAK) using deviations of daily milk yield, milk FPR, number of successful milkings, rumination time, and activity index from the herd median by parity and DIM, considering various time series lengths and numbers of d before HAK. Additionally, we aimed to use Explainable Artificial Intelligence method to understand the relationships between input variables and model outputs. Data on daily milk yield, milk fat-to-protein ratio (FPR), number of successful milkings, rumination time, activity, and health events during 0 to 21 d in milk (DIM) were retrospectively obtained from a commercial Holstein dairy farm in northern Indiana from February 2020 to January 2023. A total of 1,743 cows were included in the analysis (non-HAK = 1,550; HAK = 193). Variables were transformed based on deviations from the herd median by parity and DIM. Six LSTM models were developed to identify HAK 1, 2, and 3 d before farm diagnosis using historic cow-level data with varying time series lengths. Model performance was assessed using repeated stratified 10-fold cross-validation for 20 repeats. The Shapley additive explanations framework (SHAP) was used for model explanation. Model accuracy was 83, 74, and 70%, balanced error rate was 17 to 18, 26 to 28, and 34%, sensitivity was 81 to 83, 71 to 74, and 62%, specificity was 83, 74, and 71%, positive predictive value was 38, 25 to 27, and 21%, negative predictive value was 97 to 98, 95 to 96, and 94%, and area under the curve was 0.89 to 0.90, 0.80 to 0.81, and 0.72 for models identifying HAK 1, 2, and 3 d before diagnosis, respectively. Performance declined as the time interval between identification and farm diagnosis increased, and extending the time series length did not improve model performance. Model explanation revealed that cows with lower milk yield, number of successful milkings, rumination time, and activity, and higher milk FPR compared with herdmates of the same parity and DIM were more likely to be classified as HAK. Our results demonstrate the potential of LSTM models in identifying HAK using deviations of daily milk production variables, rumination time, and activity index from the herd median by parity and DIM. Future studies are needed to evaluate the performance of health alerts using LSTM models controlling for herd-specific metrics against commercial built-in algorithms in multiple farms and for other disorders.

Impact of dairy cow personality traits and concentrate allowance on their response to training and adaptation to an automated milking system.

The objectives of this study were to determine: 1) if dairy cow personality traits and concentrate allowance are associated with the behavior and performance of cows during training to use an automated milking system (AMS); and 2) if these factors were associated with the behavior and performance of cows after AMS training. Twenty-nine mid- to late-lactation Holstein cows (218 ± 49 DIM), who were milking on a rotary parlor and had never previously been milked in an AMS, were enrolled in this study. Cows were assigned to 1 of 2 dietary treatments, consisting of a basal partial mixed ration (PMR) common to both treatment groups, with a concentrate allowance (on dry matter basis) of: 1) 2.0 kg/d in the AMS (L-Tx); or 2) 6.0 kg/d in the AMS (H-Tx). Cows were trained to use the free-traffic AMS, with supervised milkings, over 72 h and were milked in this system for 63 d after training was complete. Variables relating to feeding behavior, milking activity, and production were measured from the start of AMS training until the end of the study. Between 42 and 63 d after AMS introduction, each cow was assessed for personality traits using a combined arena test consisting of exposure to a novel environment, novel object, and novel human. Principal components analysis of behaviors observed during the personality assessment revealed 2 factors (interpreted as boldness and activeness traits) that together explained 85% of the variance; each cow received a score for each trait. Associations between dietary treatment and personality traits with feeding behavior, milking activity, and production were analyzed using mixed-effect linear and logistic regression models. Cows with greater scores for the active trait produced less milk during the 3 d of AMS training compared with cows with lower scores. Within the H-Tx, more active cows had a 3.92 times greater risk of kicking off teat cups during AMS training than less active cows. However, during the 8 wk after training, more active cows had a 1.37 times lesser risk of teat cup kickoffs than those that were less active. Cows on the H-Tx produced 4.4 kg/d more energy-corrected milk compared with cows on the L-Tx in the 8 wk after training. During the 8 wk after AMS training the cows on the H-Tx consumed an average of 21.4 kg/d of PMR and were delivered 4.6 kg/d of AMS concentrate, while the L-Tx cows consumed 23.4 kg/d PMR and were delivered 2.0 kg/d of AMS concentrate. The results indicate that both dairy cow personality traits and AMS concentrate allocation influence their response to AMS training and subsequent feeding and milking behavior and production.

Milking system and diet forage type effect on milk quality of Italian Holstein-Friesian.

Moving from conventional (CMS) to automatic (AMS) milking systems could affect milk quality. Moreover, the type and preservation methods of the forages used in the TMR, such as alfalfa hay (HTMR) or corn silage (STMR) have been demonstrated to modify milk composition. Thus, this study investigated the effect of implementing AMS and different diet forage types on the quality of Italian Holstein-Friesian bulk milk. Milk samples (n = 168) were collected monthly from 21 commercial farms in northern Italy during a period of 8 mo. Farms were categorized into 4 groups according to their milking system (CMS vs. AMS) and diet forage type (HTMR vs. STMR). Milk quality data were analyzed through the mixed procedure for repeated measurement of SAS with the milking system, diet forage type, and sampling day as fixed effects. Milking through the AMS led to lower milk fat, freezing point, and ß-LG A; longer coagulation time; and higher K content, pH, and ß-LG B than CMS. Cows fed STMR produced milk with greater fat, protein, casein, Mg content, titratable acidity, and ß-LG A, but with reduced curd firming time, freezing point, and ß-LG B than those fed HTMR. In conclusion, milk quality is not only altered by the diet's forage type and characteristics but also by the milking system.

Randomized-controlled study assessing the effect of milking permission settings and concentrate supplementation on milking frequency and milk yield in a pasture-based automatic milking system.

This study aimed to verify the effect of milking permission (MPE) and concentrate supplementation (CS) on milking frequency (milkings per cow per day) and milk yield (kilograms per cow per day) in a farm using a pasture-based automatic milking system (AMS). Sixty-eight cows milked using this AMS unit were randomly assigned to 1 of 4 groups homogeneous for parity, DIM, and milk yield. Treatments used were frequent or restricted MPE, that granted cows permission to milk after 6 to 8 h or 9.6 to 14 h of the previous milking, respectively; and low (LC) or high (HC) CS of 0.5 kg or 3.5 kg/cow per day, respectively. The combination of the 2 levels of MPE and the 2 levels of CS resulted in the 4 treatment combinations (frequent HC [FHC], restricted HC [RHC], frequent LC [FLC], and restricted LC [RLC]). This study was designed as a 2 × 2 factorial arrangement with treatment crossover: each of the 4 cow groups was randomly assigned to 1 of the 4 treatment combinations for a 5-wk experimental period (1 pretreatment week and 4 treatment weeks), and after each 5-wk period groups crossed over to another treatment combination until they experienced all. Statistical analysis assessed the effect of MPE, CS, and their interaction on milk yield, milking frequency, box time, milking time, and average milk-flow rate. This was done using a mixed model analysis with repeated measures to account for repeated observations on the experimental unit (cow). Milk yield per cow per day and milkings per cow per day were significantly higher with the frequent compared with the restricted MPE (1.5 kg and 0.65 milkings, respectively). Milk yield per cow per day and milkings per cow per day were significantly higher with the HC compared with the LC CS (3.1 kg and 0.25 milkings, respectively). Additionally, milk yield per cow per day was affected by the interaction of MPE and CS and it was highest with the FHC (20.1 kg) treatment combination, followed by RHC (18.2 kg) treatment combination. The number of milkings per cow per day were also affected by the interaction of MPE and CS. The highest estimated number of milkings per cow per day was recorded for the FHC (2.12) and the FLC (1.77) treatment combinations, followed by the RHC (1.38) and RLC (1.23) treatment combinations. Similarly, milking interval was 2.5 h longer for the RLC treatment combination compared with RHC. The shortest milking interval was observed for the FHC (11 h) and FLC (12.8 h) treatment combinations. In conclusion, the study showed that allowing access to the robot between 6 to 8 h after the previous milking was sufficient (even with a minimal level of CS) to achieve acceptable milk production and milking performance in a pasture-based AMS.

Unlocking insights: text mining analysis on the health, welfare, and behavior of cows in automated milking systems.

Automated Milking Systems (AMS) have undergone significant evolution over the past 30 yr, and their adoption continues to increase, as evidenced by the growing scientific literature. These systems offer advantages such as a reduced milking workload and increased milk yield per cow. However, given concerns about the welfare of farmed animals, studying the effects of AMS on the health and welfare of animals becomes crucial for the overall sustainability of the dairy sector. In the last few years, some analysis conducted through text mining (TM) and topic analysis (TA) approaches have become increasingly widespread in the livestock sector. The aim of the study was to analyze the scientific literature on the impact of AMS on dairy cow health, welfare, and behavior: the paper aimed to produce a comprehensive analysis on this topic using TM and TA approaches. After a preprocessing phase, a dataset of 427 documents was analyzed. The abstracts of the selected papers were analyzed by TM and a TA using Software R 4.3.1. A Term Frequency-Inverse Document Frequency (TFIDF) technique was used to assign a relative weight to each term. According to the results of the TM, the ten most important terms, both words and roots, were feed, farm, teat, concentr, mastiti, group, SCC (somatic cell count), herd, lame and pasture. The 10 most important terms showed TFIDF values greater than 3.5, with feed showing a value of TFIDF of 5.43 and pasture of 3.66. Eight topics were selected with TA, namely: 1) Cow traffic and time budget, 2) Farm management, 3) Udder health, 4) Comparison with conventional milking, 5) Milk production, 6) Analysis of AMS data, 7) Disease detection, 8) Feeding management. Over the years, the focus of documents has shifted from cow traffic, udder health and cow feeding to the analysis of data recorded by the robot to monitor animal conditions and welfare and promptly identify the onset of stress or diseases. The analysis reveals the complex nature of the relationship between AMS and animal welfare, health, and behavior: on one hand, the robot offers interesting opportunities to safeguard animal welfare and health, especially for the possibility of early identification of anomalous conditions using sensors and data; on the other hand, it poses potential risks, which requires further investigations. TM offers an alternative approach to information retrieval in livestock science, especially when dealing with a substantial volume of documents.

Milking robots have revolutionized the cow milking, reducing dependence on human labor and increasing milk yield per cow. However, addressing concerns about farmed animal welfare and overall sustainability is crucial. This paper presents a text-mining analysis of the scientific literature to explore the effects of robotic milking on cow health, welfare, and behavior. The analysis revealed a growing body of research studies on these subjects, highlighting the complex nature of the relationship between automated milking, welfare, health, and cow behavior. Robotic milking has the potential to enhance animal health and living conditions, but the associated risks require further investigation.

Farm-level risk factors associated with increased milk ß-hydroxybutyrate and hyperketolactia prevalence on farms with automated milking systems.

The objectives of this study were to determine the farm-level hyperketolactia (HKL) prevalence, as diagnosed from the milk BHB concentration, on dairy farms milking with an automated milking system (AMS) and to describe the farm-level housing, management, and nutritional risk factors associated with increased farm-average milk BHB and the within-herd HKL prevalence in the first 45 DIM. Canadian AMS farms (n = 162; eastern Canada, n = 8; Quebec, n = 24; Ontario, n = 75; western Canada n = 55) were visited once between April and September 2019 to record housing and herd management practices. The first test milk data for each cow under 45 DIM were collected, along with the final test of the previous lactations for all multiparous cows, from April 1, 2019, to September 30, 2020. The first test milk BHB was then used to classify each individual cow as having or not having HKL (milk BHB ≥0.15 mmol/L) at the time of testing. Milk fat and protein content, milk BHB, and HKL prevalence were summarized by farm and lactation group (all, primiparous, and multiparous). During this same time period, formulated diets for dry and lactating cows, including ingredients and nutrient composition, and AMS milking data were collected. Data from the AMS were used to determine milking behaviors and milk production of each herd during the first 45 DIM. Multivariable regression models were used to associate herd-level housing, feeding management practices, and formulated nutrient composition with first test milk BHB concentrations and within-herd HKL levels separately for primiparous and multiparous cows. The within-herd HKL prevalence for all cows was 21.8%, with primiparous cows having a lower mean prevalence (12.2 ± 9.2%) than multiparous cows (26.6 ± 11.3%). Milk BHB concentration (0.095 ± 0.018 mmol/L) and HKL prevalence for primiparous cows were positively associated with formulated prepartum DMI and forage content of the dry cow diet; however, they were negatively associated with formulated postpartum DMI, the major ingredient in the concentrate supplemented through the AMS, and the partially mixed ration to AMS concentrate ratio. Multiparous cows' milk BHB concentration (0.12 ± 0.023 mmol/L) and HKL prevalence were positively associated with the length of the previous lactation, milk BHB at dry-off, prepartum diet nonfiber carbohydrate content, and the major forage fed on farm, while tending to be negatively associated with feed bunk space during lactation. This study is the first to determine the farm-level risk factors associated with herd-level prevalence of HKL in AMS dairy herds. The results may help to optimize management and guide diet formulation and thus promote the reduction of HKL prevalence.

Effects of pellet starch levels in automatic milking systems on rumen fermentation, plasma metabolites, and milk production in mid-lactation cows.

The aim of this study was to evaluate whether the starch levels in pellets fed to cows in automatic milking systems (AMS) affect subacute ruminal acidosis (SARA) occurrence and metabolite parameters. Twenty-four lactating cows (124.4 ± 49.9 days in milk) were studied in a crossover design with two periods of 21 days each and two treatment groups-a control group fed AMS pellets containing 30.0% of starch dry matter (DM) and an experimental group fed AMS pellets containing 23.5% of starch DM. All cows received the same partial mixed ration (PMR). The 1-hr mean ruminal pH in both groups decreased over 4 hr after feeding on PMR but recovered by the next morning. The ruminal pH was unaffected by either treatment, and both groups developed SARA. The groups had no significant differences in the concentrations of ruminal volatile fatty acids, lipopolysaccharides, plasma acute-phase proteins, other metabolites, and hormones. The milk yield and composition were not different in both groups. Feeding low-starch pellets in the AMS did not contribute to the risk of SARA occurrence in cows and had no additive effects on rumen fermentation, plasma metabolites, or milk production.

Behavioural variability, physical activity, rumination time, and milk characteristics of dairy cattle in response to regrouping.

In the commercial dairy industry worldwide, it is common practice to periodically regroup cows as part of their management strategy within housed systems. While this animal husbandry practice is intended to improve management efficiency, cows may experience social stress as a result of the social environment changes, which may have an impact on their behavioural patterns, performance, and welfare. We investigated whether regrouping altered dairy cows' behaviour and impacted their cortisol concentration (a physiological marker of stress), oxytocin, milk yield, and quality in a robotic milking system. Fifty-two lactating cows (17 primiparous; 35 multiparous) were moved in groups of 3-5 individuals into established pens of approximately 100 cows. Behaviour of the regrouped cows was directly observed continuously for 4 h/day across 4-time blocks (day-prior (d-1), day-of regrouping (d0), day-after (d + 1), and 6-days after (d + 6) regrouping). Cows were categorised as being with others, alone, or feeding every 2.5 min prior to the assessment of behavioural dynamics. Milk yield (MY) and composition, total daily activity, and rumination time (RUM) data were extracted from the Lely T4C management program (Lely Industries, Maassluis, the Netherlands), and milk samples were collected for cortisol and oxytocin concentration analyses; data were analysed using linear mixed-effect modelling. Primiparous cows were less likely to be interacting with others on d + 1 than d-1 compared with multiparous. However, average bout duration (minutes) between being alone and feeding activity states were similar on d-1, d + 1, and d + 6, for both primiparous and multiparous cows. A reduction in the average alone and feeding bout duration was observed on d0. Multiparous cows spent significantly more total time being alone on d0 compared to d-1. Neither regrouping nor parity statistically influenced milk DM content, energy, or cortisol concentration. Primiparous cows produced 3.80 ±â€¯2.42 kg (12.2%) less MY on d + 1 compared to their d-1, whereas multiparous cows did not change MY. A significant decrease of 0.2% fat was found in both parity groups following regrouping and remained low up to d + 6. Daily activity in both parity groups increased significantly and RUM reduced after regrouping. A significant decrease in oxytocin concentration was observed in all cows on d + 1. The results, specifically for primiparous cows, indicated a negative impact of regrouping on social interactions, due to changes in the social environment which may lead to short-term social instability. Multiparous cows may benefit from previous regrouping experiences.

Effect of pre- and postpartum supplementation of a pure glycerol product to dairy cows on feeding behavior, lying behavior, and reticulorumen pH.

The objective of this study was to quantify the effects of supplementing a low level of dry glycerol product pre- and postpartum on the feeding behavior, lying behavior, and reticulorumen pH of dairy cows. Multiparous Holstein dairy cows (n = 60) were enrolled in a 2 × 2 factorial design study. Twenty-one days before expected parturition, cows individually received a dry cow diet with (1) 250 g/d glycerol supplementation (GLY; 66% pure glycerol, United States Pharmacopeia grade), or (2) no supplementation (CON). Following parturition, cows were individually assigned to either (1) 250 g/d glycerol product (GLY; 66% pure glycerol), or (2) no supplementation (CON) to their partial mixed ration (PMR) for the first 21 d in milk (DIM). All cows were milked by an automated milking system and offered a target of 5.4 kg/d pellet (23% of target total dry matter intake [DMI]). For both treatment periods, cows were individually assigned to automated feed bins to measure PMR feeding behavior. Rumination time and lying behavior were monitored with electronic sensors for the whole study (-21 to 21 DIM). Reticulorumen pH boluses were administered to a subset of cows (n = 40) where pH was recorded every 10 min from 21 d prepartum to 21 d postpartum. Prepartum, cows fed GLY had fewer, larger meals and spent 20.2% more time feeding than CON while consuming feed at a similar rate. Cows on the CON diet prepartum spent more time lying down in more frequent bouts in the 21 d before calving. Following parturition, cows that received GLY prepartum continued to devote more time to eating, while tending to spend less time ruminating per kilogram of DMI. Cows receiving CON postpartum had larger meals with longer intervals between meals. In the first 21 DIM, cows receiving CON prepartum tended to have shorter, but significantly more frequent, lying bouts than cows fed GLY prepartum. Glycerol supplementation pre- and postpartum resulted in less time spent lying down following parturition. Minimal differences between treatments were observed for pre- and postpartum sorting behavior or reticulorumen pH. Overall, supplementation of glycerol pre- and postpartum altered cow time budgets, with cows spending more time eating pre- and postpartum, less time lying pre- and postpartum, and having fewer, larger meals prepartum when receiving glycerol prepartum, and with cows having slower feeding rates and smaller meals following parturition with postpartum glycerol supplementation.

Effects of 2 gradual debonding strategies on machine milk yield, flow, and composition in a cow-driven cow-calf contact system.

This controlled study compared the effects of 2 different gradual debonding strategies on machine milk yield, flow, and composition in a cow-driven cow-calf contact (CCC) system with automatic milking. Cows had 24 h/d access to their calves during the first weeks of lactation. In the long debonding (LDB) treatment (n = 16), a gradual reduction of cows' access to their calves was initiated 4 wk after calving over a total period of 28 d; first to 12 h/d (14 d), and then to 6 h/d (14 d). In the short debonding (SDB) treatment (n = 14), gradual reduction was initiated 6.5 wk after calving over a total period of 10 d; first to 12 h/d (5 d), and then to 6 h/d (5 d). From 6 h/d, access was finally reduced to 0 h/d for 7 d for both treatments. Machine milk yield, somatic cell count, and peak and average milk flow were automatically registered at milking. During the 9-wk study period, composite samples were analyzed for milk composition. Data were analyzed with linear mixed effect models. Results showed that machine milk yield during 24 h/d access varied between cows (range 1.2-49.9 kg/d, average ± standard deviation 13.2 ± 7.82 kg/d). The LDB cows had a higher daily machine milk yield than SDB cows at the end of and after access reduction was completed (+5.0 ± 1.63 and +5.1 ± 1.55 kg during the last 5 d of 6 h/d access, and 0 h/d access, respectively). Somatic cell count was on a healthy level, with no difference between treatments. Milk fat content increased with reduction in access, regardless of treatment. Short debonding cows tended to show higher milk protein content and lower milk lactose content than cows with a longer debonding. This study has shown that a longer debonding initiated earlier may give a higher milk yield in the short term. The variation in machine milk yield may indicate differences in milk ejection, suckling, and visiting patterns and preferences among cows.

Planned behavior, social networks, and perceived risks: Understanding farmers' behavior toward precision dairy technologies.

Precision dairy tools (PDT) can provide timely information on individual cow's physiological and behavioral parameters, which can lead to more efficient management of the dairy farm. Although the economic rationale behind the adoption of PDT has been extensively discussed in the literature, the socio-psychological aspects related to the adoption of these technologies have received far less attention. Therefore, this paper proposes a socio-psychological model that builds upon the theory of planned behavior and develops hypotheses regarding cognitive constructs, their interaction with the farmers' perceived risks and social networks, and their overall influence on adoption. These hypotheses are tested using a generalized structural equation model for (a) the adoption of automatic milking systems (AMS) on the farms and (b) the PDT that are usually adopted with the AMS. Results show that adoption of these technologies is affected directly by intention, and the effects of subjective norms, perceived control, and attitudes on adoption are mediated through intention. A unit increase in perceived control score is associated with an increase in marginal probability of adoption of AMS and PDT by 0.05 and 0.19, respectively. Subjective norms are associated with an increase in marginal probability of adoption of AMS and PDT by 0.009 and 0.05, respectively. These results suggest that perceived control exerts a stronger influence on adoption of AMS and PDT, particularly compared with their subjective norms. Technology-related social networks are associated with an increase in marginal probability of adoption of AMS and PDT by 0.026 and 0.10, respectively. Perceived risks related to AMS and PDT negatively affect probability of adoption by 0.042 and 0.16, respectively, by having negative effects on attitudes, perceived self-confidence, and intentions. These results imply that integrating farmers within knowledge-sharing networks, minimizing perceived risks associated with these technologies, and enhancing farmers' confidence in their ability to use these technologies can significantly enhance uptake.

Genomic Selection for Dairy Cattle Behaviour Considering Novel Traits in a Changing Technical Production Environment.

Cow behaviour is a major factor influencing dairy herd profitability and is an indicator of animal welfare and disease. Behaviour is a complex network of behavioural patterns in response to environmental and social stimuli and human handling. Advances in agricultural technology have led to changes in dairy cow husbandry systems worldwide. Increasing herd sizes, less time availability to take care of the animals and modern technology such as automatic milking systems (AMSs) imply limited human-cow interactions. On the other hand, cow behaviour responses to the technical environment (cow-AMS interactions) simultaneously improve production efficiency and welfare and contribute to simplified "cow handling" and reduced labour time. Automatic milking systems generate objective behaviour traits linked to workability, milkability and health, which can be implemented into genomic selection tools. However, there is insufficient understanding of the genetic mechanisms influencing cow learning and social behaviour, in turn affecting herd management, productivity and welfare. Moreover, physiological and molecular biomarkers such as heart rate, neurotransmitters and hormones might be useful indicators and predictors of cow behaviour. This review gives an overview of published behaviour studies in dairy cows in the context of genetics and genomics and discusses possibilities for breeding approaches to achieve desired behaviour in a technical production environment.

A Literature Review of Modeling Approaches Applied to Data Collected in Automatic Milking Systems.

Automatic milking systems (AMS) have played a pioneering role in the advancement of Precision Livestock Farming, revolutionizing the dairy farming industry on a global scale. This review specifically targets papers that focus on the use of modeling approaches within the context of AMS. We conducted a thorough review of 60 articles that specifically address the topics of cows' health, production, and behavior/management Machine Learning (ML) emerged as the most widely used method, being present in 63% of the studies, followed by statistical analysis (14%), fuzzy algorithms (9%), deterministic models (7%), and detection algorithms (7%). A significant majority of the reviewed studies (82%) primarily focused on the detection of cows' health, with a specific emphasis on mastitis, while only 11% evaluated milk production. Accurate forecasting of dairy cow milk yield and understanding the deviation between expected and observed milk yields of individual cows can offer significant benefits in dairy cow management. Likewise, the study of cows' behavior and herd management in AMSs is under-explored (7%). Despite the growing utilization of machine learning (ML) techniques in the field of dairy cow management, there remains a lack of a robust methodology for their application. Specifically, we found a substantial disparity in adequately balancing the positive and negative classes within health prediction models.

Prevalence of udder pathogens in milk samples from Norwegian dairy cows recorded in a national database in 2019 and 2020.

BACKGROUND: Identification of aetiological agents of mastitis in dairy cattle is important for herd management of udder health. In Norway, results from mastitis diagnostics are systematically recorded in a central database, so that the dairy industry can follow trends in the recorded frequency of udder pathogens and antimicrobial resistance patterns at national level. However, bacteriological testing of milk samples is based on voluntary sampling, and data are therefore subject to some bias. The aim of this study was to examine the prevalence of udder pathogens in Norwegian dairy cows by analysing data from the national routine mastitis diagnostics and to explore how routines for sampling and diagnostic interpretations may affect the apparent prevalence of different bacterial pathogens. We also assessed associations between udder pathogen findings and the barn- and milking systems of the herds. RESULTS: The most frequently detected major udder pathogens among all milk samples submitted for bacterial culture (n = 36,431) were Staphylococcus aureus (24.5%), Streptococcus dysgalactiae (13.3%) and Streptococcus uberis (9.0%). In the subset of samples from clinical mastitis (n = 7598); Escherichia coli (14.5%) was the second most frequently detected pathogen following S. aureus (27.1%). Staphylococcus epidermidis (10.0%), Corynebacterium bovis (9.4%), and Staphylococcus chromogenes (6.0%) dominated among the minor udder pathogens. Non-aureus staphylococci as a group, identified in 39% of the sampling events, was the most frequently identified udder pathogen in Norway. By using different definitions of cow-level bacterial diagnoses, the distribution of minor udder pathogens changed. Several udder pathogens were associated with the barn- and milking system but the associations were reduced in strength when data were analysed from farms with a comparable herd size. S. aureus was associated with tiestall housing, E. coli and S. dysgalactiae were associated with freestall housing, and S. epidermidis was associated with automatic milking systems. Only 2.5% of the 10,675 tested S. aureus isolates were resistant to benzylpenicillin. Among the 2153 tested non-aureus staphylococci, altogether 34% were resistant to benzylpenicillin. CONCLUSIONS: This study presents the recorded prevalence of udder pathogens in Norway over a two-year period and assesses the possible impact of the sampling strategies, diagnostic methods and diagnostic criteria utilized in Norway, as well as associations with different housing and milking systems. The national database with records of results from routine mastitis diagnostics in Norway provides valuable information about the aetiology of bovine mastitis at population level and can reveal shifts in the distribution and occurrence of udder pathogens.

The Combined Use of Automated Milking System and Sensor Data to Improve Detection of Mild Lameness in Dairy Cattle.

This study aimed to develop a tool to detect mildly lame cows by combining already existing data from sensors, AMSs, and routinely recorded animal and farm data. For this purpose, ten dairy farms were visited every 30-42 days from January 2020 to May 2021. Locomotion scores (LCS, from one for nonlame to five for severely lame) and body condition scores (BCS) were assessed at each visit, resulting in a total of 594 recorded animals. A questionnaire about farm management and husbandry was completed for the inclusion of potential risk factors. A lameness incidence risk (LCS ≥ 2) was calculated and varied widely between farms with a range from 27.07 to 65.52%. Moreover, the impact of lameness on the derived sensor parameters was inspected and showed no significant impact of lameness on total rumination time. Behavioral patterns for eating, low activity, and medium activity differed significantly in lame cows compared to nonlame cows. Finally, random forest models for lameness detection were fit by including different combinations of influencing variables. The results of these models were compared according to accuracy, sensitivity, and specificity. The best performing model achieved an accuracy of 0.75 with a sensitivity of 0.72 and specificity of 0.78. These approaches with routinely available data and sensor data can deliver promising results for early lameness detection in dairy cattle. While experimental automated lameness detection systems have achieved improved predictive results, the benefit of this presented approach is that it uses results from existing, routinely recorded, and therefore widely available data.

Multivariable time series classification for clinical mastitis detection and prediction in automated milking systems.

In this study, we developed a machine learning framework to detect clinical mastitis (CM) at the current milking (i.e., the same milking) and predict CM at the next milking (i.e., one milking before CM occurrence) at the quarter level. Time series quarter-level milking data were extracted from an automated milking system (AMS). For both CM detection and prediction, the best classification performance was obtained from the decision tree-based ensemble models. Moreover, applying models on a data set containing data from the current milking and past 9 milkings before the current milking showed the best accuracy for detecting CM; modeling with a data set containing data from the current milking and past 7 milkings before the current milking yielded the best results for predicting CM. The models combined with oversampling methods resulted in specificity of 95 and 93% for CM detection and prediction, respectively, with the same sensitivity (82%) for both scenarios; when lowering specificity to 80 to 83%, undersampling techniques facilitated models to increase sensitivity to 95%. We propose a feasible machine learning framework to identify CM in a timely manner using imbalanced data from an AMS, which could provide useful information for farmers to manage the negative effects of CM.

Herd and animal factors affect the variability of total and differential somatic cell count in bovine milk.

The aim of this study was to quantify some environmental (individual herds, herd productivity, milking system, and season) and animal factors [individual animals, breed, days in milk (DIM) and parity] on the variability of the log-10 transformation of somatic cell count (LSCC) and differential somatic cell count (DSCC) on individual bovine milk. A total of 159,360 test-day records related to milk production and composition were extracted from 12,849 Holstein-Friesian and 9,275 Simmental cows distributed across 223 herds. Herds were classified into high and low productivity, defined according to the average daily milk net energy output (DMEO) yielded by the cows. Data included daily milk yield (DYM; kg/d), milk fat, protein, lactose, SCC, and DSCC, and information on herds (i.e., productivity, milking system). The daily production of total and differential somatic cells in milk was calculated and then log-10 transformed, obtaining DLSCC and DLDSCC, respectively. Data were analyzed using a mixed model including the effects of individual herd, animal, repeated measurements intra animal as random, and herd productivity, milking system, season, breed, DIM, parity, DIM × parity, breed × season, DIM × milking system and parity × milking system as fixed factors. Herds with a high DMEO were characterized by a lower content of LSCC and DSCC, and higher DLSCC and DLDSCC, compared to the low DMEO herds. The association between milking system and somatic cell traits suggested that the use of the automatic milking systems would not allow for a rapid intervention on the cow, as evidenced by the higher content of all somatic cell traits compared to the other milking systems. Season was an important source of variation, as evidenced by high LSCC and DSCC content in milk during summer. Breed of cow had a large influence, with Holstein-Friesian having greater LSCC, DSCC, DLSCC, and DLDSCC compared to Simmental. With regard to DIM, the variability of LSCC was mostly related to that of DSCC, showing an increase from calving to the end of lactation, and suggesting the higher occurrence of chronic mastitis in cows toward the end of lactation. All the somatic cell traits increased across number of parities, possibly because older cows may have increased susceptibility to intramammary infections.

This study investigated factors affecting the variability of somatic cell traits in bovine milk. Animal had greater influence on somatic cell score (SCS) and differential somatic cell count (DSCC) compared to herd factors. Herds producing high average of daily milk energy were characterized by lower SCS and DSCC compared to the low average daily milk energy herds. The SCS and DSCC were higher in Holstein-Friesian than in Simmental, and during summer with respect to the other seasons. Older cows at the end of lactation showed the highest content of somatic cell traits. These results are helpful for the management of somatic cell traits at herd and animal levels.

The costs of chronic mastitis: A simulation study of an automatic milking system farm.

Mastitis is a production disease in dairy farming that causes economic losses. Especially chronic mastitis (i.e., mastitis cases continuing longer than 28 days) can substantially affect the risk of transmission of intramammary infections (IMI) and total milk production losses. Insights into the impact of chronic mastitis on production and farm economics are needed to guide chronic mastitis decision-making. We aimed to estimate the costs of chronic mastitis with a Monte Carlo simulation model in which the costs of chronic mastitis were estimated as part of the total mastitis costs. The model simulated milk yields, IMI dynamics, somatic cell count (SCC), and pregnancy status on an average Dutch dairy farm with 100 cow places over 9 years. The model was parameterized using information from the literature and actual sensor data from automatic milking system (AMS) farms. The daily subclinical milk production losses were modeled using a generalized additive model and sensor data. Transmission of IMI was modeled as well. The model results indicated median total costs of mastitis of € 230 per generic IMI case (i.e., a weighted average of all pathogens). The most substantial cost factors were the extra mastitis cases due to transmission, culling, and milk production losses. Other significant costs originated from dry cow treatments and diverted milk. The model also indicated median total costs due to chronic mastitis of € 118 (51 % of the total mastitis costs). The share of chronic mastitis relative to the total mastitis costs was substantial. Transmission of contagious bacteria had the largest share among the chronic mastitis costs (51 % of the costs of chronic cases). The large share of chronic mastitis costs in the total mastitis costs indicates the economic importance of these mastitis cases. The results of the study point to the need for future research to focus on chronic mastitis and reducing its presence on the AMS dairy farm.

Sensitivity and Specificity for the Detection of Clinical Mastitis by Automatic Milking Systems in Bavarian Dairy Herds.

In automatic milking systems (AMSs), the detection of clinical mastitis (CM) and the subsequent separation of abnormal milk should be reliably performed by commercial AMSs. Therefore, the objectives of this cross-sectional study were (1) to determine the sensitivity (SN) and specificity (SP) of CM detection of AMS by the four most common manufacturers in Bavarian dairy farms, and (2) to identify routinely collected cow data (AMS and monthly test day data of the regional Dairy Herd Improvement Association (DHIA)) that could improve the SN and SP of clinical mastitis detection. Bavarian dairy farms with AMS from the manufacturers DeLaval, GEA Farm Technologies, Lely, and Lemmer-Fullwood were recruited with the aim of sampling at least 40 cows with clinical mastitis per AMS manufacturer in addition to clinically healthy ones. During a single farm visit, cow-level milking information was first electronically extracted from each AMS and then all lactating cows examined for their udder health status in the barn. Clinical mastitis was defined as at least the presence of visibly abnormal milk. In addition, available DHIA test results from the previous six months were collected. None of the manufacturers provided a definition for clinical mastitis (i.e., visually abnormal milk), therefore, the SN and SP of AMS warning lists for udder health were assessed for each manufacturer individually, based on the clinical evaluation results. Generalized linear mixed models (GLMMs) with herd as random effect were used to determine the potential influence of routinely recorded parameters on SN and SP. A total of 7411 cows on 114 farms were assessed; of these, 7096 cows could be matched to AMS data and were included in the analysis. The prevalence of clinical mastitis was 3.4% (239 cows). When considering the 95% confidence interval (95% CI), all but one manufacturer achieved the minimum SN limit of >80%: DeLaval (SN: 61.4% (95% CI: 49.0%−72.8%)), GEA (75.9% (62.4%−86.5%)), Lely (78.2% (67.4%−86.8%)), and Lemmer-Fullwood (67.6% (50.2%−82.0%)). However, none of the evaluated AMSs achieved the minimum SP limit of 99%: DeLaval (SP: 89.3% (95% CI: 87.7%−90.7%)), GEA (79.2% (77.1%−81.2%)), Lely (86.2% (84.6%−87.7%)), and Lemmer-Fullwood (92.2% (90.8%−93.5%)). All AMS manufacturers' robots showed an association of SP with cow classification based on somatic cell count (SCC) measurement from the last two DHIA test results: cows that were above the threshold of 100,000 cells/mL for subclinical mastitis on both test days had lower chances of being classified as healthy by the AMS compared to cows that were below the threshold. In conclusion, the detection of clinical mastitis cases was satisfactory across AMS manufacturers. However, the low SP will lead to unnecessarily discarded milk and increased workload to assess potentially false-positive mastitis cases. Based on the results of our study, farmers must evaluate all available data (test day data, AMS data, and daily assessment of their cows in the barn) to make decisions about individual cows and to ultimately ensure animal welfare, food quality, and the economic viability of their farm.