Dairy farms are increasingly adopting renewable energy systems to reduce emissions and energy costs, yet comprehensive economic evaluations are lacking. The objective of this study was to conduct an investment appraisal of PV, a hot water diverter (HWD), and battery energy storage (BES) on dairy farms, using three farms of varying scale, assessing economic and environmental performance. Four scenarios were evaluated (1: PV, 2: PV with HWD, 3: PV with BES, 4: PV with HWD and BES) and compared to a baseline without investment in these technologies. Three farms were simulated (Farm 1: 99 cows, Farm 2: 191 cows, Farm 3: 329 cows). Three electricity tariffs were considered (flat rate, day/night, time-of-use (TOU)). PV generation, self-consumption, electricity cost, additional profit (AP), return on investment (ROI), and electricity-related CO2 emissions were quantified over 10 years. The flat rate tariff resulted in the highest electricity costs, while the day/night tariff resulted in the lowest for all farms and scenarios. PV with BES (scenario 3) yielded the lowest electricity costs for all farms when day/night or TOU tariffs were applied. With a day/night tariff, scenario 1 resulted in the largest ROI for Farm 1 (15.67%), scenario 3 resulted in the largest ROI for Farm 2 (16.99%), and scenarios 1 and 3 resulted in the largest ROI (17.23%) for Farm 3.

This paper presents the Farm-level Renewable Microgeneration Optimisation Tool (FaRMOT), a platform designed to simulate the financial and environmental performance of integrating renewable energy sources (RES) and demand-side management (DSM) strategies on dairy and beef farms. FaRMOT incorporates two energy models: FaRMOT-Dairy and FaRMOT-Beef, which simulate farm electricity consumption at 15-minute and hourly intervals, respectively. FaRMOT-Dairy evaluated RES and DSM integration on dairy farms, including photovoltaic (PV) systems, battery energy storage systems (BESS), and energy-efficient technologies such as plate coolers and heat recovery systems. FaRMOT-Beef focused on RES integration in beef farms, assessing PV and BESS systems. The platform also included an embedded greenhouse gas (GHG) offset calculator to quantify the environmental benefits of renewable energy adoption. To demonstrate FaRMOT’s applicability, three representative Irish farms (one dairy and two beef) were simulated. For the dairy case study, a 17 kWp PV system with an 8.5 kWh BESS achieved $55\%$ self-consumption, a 3.6-year payback period, and an estimated GHG offset of ${7 5. 9 2}$ tCO2 over 20 years. For the two beef farms, simulations with a 6 kWp PV system showed that $56 \%$ and $48 \%$ of generated electricity could be consumed on-site by farm operations and the dwelling house, resulting in payback periods of 2.3 and 2.4 years. These results demonstrated that FaRMOT can help farmers to optimise the design, sizing, and deployment of PV, BESS, and DSM strategies to improve energy efficiency and economic returns. Additionally, the tool provides policymakers with a framework to evaluate scenarios for reducing electricity use, operational costs, and GHG emissions.

The objective of this study was to quantify the method-to-method variation between two widely used field indicators of the end-of-milking vacuum-exposure period (i.e., operational overmilking duration), and to identify cow- and milking-level factors associated with this variation. Operational overmilking was defined using two approaches: (i) MPC vacuum fluctuation patterns collected via VaDia™ recording devices, and (ii) milk flow curves generated from milking system data, with simulated ACR take-off thresholds ranging from 0.2 to 0.8 kg/min. Seven quarter combinations were analyzed to determine their effect on method-to-method variation. Multivariable modelling was used to investigate the factors which influenced the absolute difference in operational overmilking duration (ADOD) between methods, with larger ADOD indicating greater method-to-method variation. All quarter combinations showed large method-to-method variations. VaDiaTM-derived estimates indicated longer overmilking durations and higher milk flow at the onset of overmilking compared with the milk flow curve approach. Our findings showed that a combination of the rear quarters was significantly associated with the lowest ADOD, and that a combination of the front quarters was significantly associated with the highest ADOD. All other combinations did not differ from each other, indicating that combinations including one front and one rear quarter performed similarly, and that recording all four quarters did not improve agreement between methods within this dataset. Milk flow factors associated with increased ADOD included longer low flow times, longer high flow times, longer machine-on times, and increased yield. Vacuum values associated with increased ADOD included high short milk tube vacuum during the full milking, and high mouthpiece chamber vacuum levels during both the full milking and overmilking periods. High short milk tube vacuum during overmilking was associated with decreased ADOD. Wider teat diameters, longer teat lengths, and increased parity were associated with increased ADOD. These findings indicated that vacuum-based and flow-based indicators of operational overmilking capture different aspects of the end-milking process and should be clearly specified when measuring or reporting overmilking in research or commercial milking systems.

The current challenges associated with increasing global demand for dairy products, developing herd structures, and declining availability of labor have resulted in the need for improvements in contemporary dairy farm infrastructure and management practices. Internationally, across different production systems, the milking process accounts for between 33% and 75% of annual labor requirements on commercial dairy farms. Because milking efficiency is multifactorial and highly context-specific, a clearer understanding of the factors that influence it-as evaluated through key performance indicators-is required to support future optimization strategies. This study conducted a systematic review of the milking efficiency literature to identify knowledge gaps and evaluate how previous research approached this topic. Five research questions (RQ) guided the review, relating to contributions by country and production system (RQ1), problem types and research objectives (RQ2), methodologies used (RQ3), milking efficiency values across parlor systems (RQ4), and the effect of operator involvement (RQ5). Fifteen publications were analyzed, representing 23 sample groups and 315 parlors. Outcome variables included cows milked per hour (cows/h), cows milked per operator per hour (cows/h per operator), cows milked per cluster per hour (cows/h per cluster), and clusters per operator (clusters/operator). Four parlor types were represented: herringbone double-up, parallel double-up (P-DU), herringbone swing-over (H-SO), and rotary (RO). The United States and confinement-based systems contributed the most studies. Comparative research was the dominant problem type, and time-and-motion analysis was the most frequently used methodology. The P-DU parlors achieved the greatest overall throughput (352 cows/h), RO parlors achieved the highest operator-level efficiency (143 cows/h per operator) and clusters handled per operator (27 clusters/operator), and H-SO parlors achieved the highest throughput per cluster (5.6 cows/h per cluster). Across all parlor types, increasing operator number did not significantly increase hourly cow throughput or cows/h per cluster, but significantly reduced cows/h per operator and clusters/operator at milking. This systematic review synthesizes the structure, methods, and empirical findings of the milking efficiency literature. It provides a summary understanding of milking efficiency levels across a range of parlor types operating in different production systems and quantifies the effect of operator on milking efficiency levels relative to parlor type. For researchers in this field, the authors of this review have highlighted key directions for future investigation into milking efficiency optimization through the alignment of parlor infrastructure, automation specification, and management practices at milking.

BACKGROUND: This study describes associations between bulk tank somatic cell count (BTSCC) and farm management practices, parlour management practices and implemented technologies, milking management practices, somatic cell count (SCC) control strategies, and farmer demographics and attitudes around SCC management using a sample of Irish dairy farms. RESULTS: This paper utilised a pre-existing dataset from a farm management and technology survey of 376 commercial Irish dairy farms conducted in 2022. Five mixed models were used to examine associations between variables in each of the five survey sections and log-10 transformed BTSCC (log10BTSCC). Seasonal calving patterns, family members milking alongside survey respondents, and keeping of mastitis treatment records were associated with lower log10BTSCC. Parlour technologies such as automatic cluster removers and automatic washers on the milking machine were associated with significantly reduced log10BTSCC, whereas the presence of backing gates and straight breast rails were associated with increased log10BTSCC. Fore-milking, pre-milking udder preparation and post-milking teat disinfection contributed to lower log10BTSCC. Advice sought from veterinary professionals regarding SCC, multi-faceted approaches to selective dry cow therapy decisions, and utilisation of results from multiple milk recordings were also associated with significantly decreased log10BTSCC. CONCLUSIONS: In this study, we successfully established associations between log10BTSCC and farm management practices, parlour management practices and implemented technologies, milking management practices, SCC control strategies, and farmer demographics and attitudes around SCC management. We identified scope for further research on many of the aspects found to be associated with log10BTSCC in this study, particularly in the areas of cow positioning within parlours, fore-milking practices, milk recording, and means of disseminating SCC advice to farmers, particularly around the topics of parlour hygiene and selective dry cow therapy.

The objective of this study was to investigate the effect on milking duration and teat condition after milking with dynamic pulsation settings that increased the open phase and reduced the closed phase of pulsation during the peak milk flow period, together with increasing the milk flow rate switch-point for cluster detachment. The present study filled current gaps in knowledge by informing on the effects of both milk flow rate switch-points and dynamic pulsation together in one experiment, along with presenting data on milking performance, strip milk, teat condition and vacuum levels in the cluster during milking. To this end, 4 treatments consisting of different milk flow rate switch-points and pulsator settings combinations were deployed across 4 groups of 24 cows for 8 weeks. Treatments consisted of 2 levels of milk flow rate switch-point (0.2 kg/min and 0.8 kg/min) and 2 pulsator settings (static and dynamic). The static pulsator settings had a pulsator ratio of 65:35. For the dynamic settings, a pulsator ratio of 63:37 below a milk flow rate of 2 kg/min was used, which changed to 73:27 above 2 kg/min. Milking parameters and teat scoring data were analyzed using mixed models. The effect of milk flow rate switch-point on milking duration was significant, whereas the effect of dynamic pulsation was not. The reductions in milking duration were larger for p.m. milking (for static pulsation settings, the average reduction was 21% between switch-point of 0.2 kg/min and 0.8 kg/ min) compared with a 12% reduction for a.m. milking. In addition, we found a significant effect of treatment on teat-barrel congestion, with reduced odds of teat-barrel ringing for treatments with a milk flow rate switch-point of 0.8 kg/min relative to 0.2 kg/min due to significantly reduced over milking time and hence reduced exposure to higher mouthpiece chamber vacuum during the over milking period. On average across all treatments, cows spent 32% of the milking above 2 kg/min, but this varied from 37% for a.m. milking to 23% on average for p.m. milking; hence, an uneven milking interval impeded the ability of dynamic pulsation to contribute, especially for p.m. milking. We concluded that milk flow rate switch- point settings are more impactful than dynamic pulsation settings for reducing milking duration. Furthermore, we found no evidence of interactions between the effects of switch-point and pulsator setting on milking duration, milk yield or milk flow rate.

Rotary milking systems have potential to reduce the labor requirement of the milking process on dairy farms. However, there is a need to identify the most effective strategies that lead to consistently efficient milkings. We developed a mechanistic, process-oriented model that accurately simulates the milking efficiency of rotary parlors operating under a diverse range of conditions. The rotary parlor model (RPM) was developed using milking efficiency data collected from a sample of commercial Irish dairy farms (n = 10) and the Teagasc Moorepark Dairy Research Farm (Teagasc, Ireland). To validate the performance of RPM, simulated milking process times (MPT, s) and efficiency metrics (cows milked per hour [cows/h], liters of milk harvest per hour [L/h], MPT per cow [s/cow]) were compared with empirical data recorded on the Moorepark Dairy Research Farm across 2 recording periods. Model validation produced mean absolute percentage error values of 3.3%, 3.5%, and 2.9% for MPT, cows/h, and L/h metrics, respectively. We defined rotation time as the time taken for the rotary platform to pass 1 bail position (s/bail); this value decreases as the platform rotates faster. The functionality of RPM was demonstrated by simulating the milking efficiency (cows/h) of parlor sizes ranging from 40 to 120 clusters, operating at rotation times of 5 to 25 s/bail. We found that the effect of decreasing rotation time (≤15 s/bail) on milking efficiency was greater for larger parlors (50, 60 clusters) than lower-sized parlors (40 clusters). For example, decreasing rotation time from 15 to 10 s/bail increased milking efficiency by 3% for a 40-cluster parlor, compared with 26% for a 60-cluster parlor. As rotation time decreased for all parlor sizes, there was an increased occurrence of go-around cows at milking (i.e., cows with a milking time longer than the platform time, thereby requiring additional rotations). A sensitivity analysis investigated the effect of automatic cluster remover (ACR) threshold change from 0.2 to 0.8 kg/min on the milking efficiency of 40-, 50-, and 60-cluster parlors operating at rotation times of 6, 8, 10, 12, 16, and 20 s/bail for a 300- and 600-cow herd. Using increased ACR thresholds reduced the milking time duration of individual cows. This lowered the number of go-around cows, and thereby rotations required at milking, as a result, milking efficiency increased. The ACR threshold increase effect was largest among lower-sized parlors with lower rotation times. For example, increasing the ACR threshold from 0.2 to 0.8 kg/min for a 40-cluster parlor with a rotation time of 12 s/bail lowered MPT by 17%. In contrast, for a 60-cluster parlor operating in the same conditions, the increased ACR threshold reduced MPT by only 5%. Optimal go-around cow occurrence ranged between 2% to 20%, depending on herd size, parlor size, rotation time and ACR threshold. Through the development and application of RPM, this study provides greater understanding into the effects of parlor size, rotation time, rotation numbers, herd milking characteristics, ACR thresholds, and go-around cows on rotary milking efficiency.

The objective of this study was to develop, validate, and demonstrate the herringbone parlor model (HPM). The HPM was built using empirical data collected from a sample of commercial Irish dairy farms (n = 16). The HPM is a mechanistic model that accurately simulates the milking process time and milking efficiency of herringbone swing-over parlors, where 1 operator is present at milking, accounting for variances in parlor infrastructure, management practices, and automation specification. The HPM was validated by comparing simulated outputs against empirical recordings from a commercial dairy farm across 2 typical periods during the lactation cycle. Mean absolute percentage error (MAPE) values of 9.6% and 8.4% for cows per hour (cows/h) and milking process time per cow (MPT, s/cow) were observed through the HPM validation process, respectively. The average MAPE for row times was 7.9%. HPM was then demonstrated across 3 parlor sizes (1 × 16, 1 × 20, and 1 × 24 clusters) with automatic cluster removers (ACR) and rapid exit. We found that smaller parlors experienced the largest benefits from the use of automations. For example, for a 1 × 16 parlor, we found the addition of ACR or rapid exit increased milking efficiency (cows/h) by 11% and 6%, respectively. Combined, they increased milking efficiency by 14%. In contrast, for a 1 × 24 parlor, adding ACRs increased milking efficiency (cows/h) by 7%, whereas rapid exit increased cows/h by 2%. Combined, ACR and rapid exit increased milking efficiency by only 8%. A sensitivity analysis examined the effect of an increased ACR threshold (from 0.2 to 0.8 kg/min) on the milking efficiency for the 3 different parlor sizes. Increasing the ACR threshold had a low impact on milking efficiency, with cows/h values increasing by 9%, 1%, and 7% for 1 × 16, 1 × 20, and 1 × 24 cluster parlors, respectively, when compared with values achieved at the lower threshold (0.2 kg/min). However, when an increased ACR threshold (0.8 kg/min) was used together with a rapid exit system, substantial gains in efficiency were generated. Using the increased ACR threshold together with the rapid exit system increased milking efficiency (cows/h) by 26% for a 1 × 16 cluster parlor when compared with a 1 × 16 cluster parlor with no automations. Similarly, for a 1 × 24 cluster parlor, using an ACR threshold of 0.8 kg/min with the rapid exit system increased milking efficiency (cows/h) by 16% when compared with a 1 × 24 cluster parlor with no automations. Further, we found that the use of ACR increased the operator idle time across all parlor sizes. Using ACR with a threshold of 0.2 kg/min increased the operator idle time of 1 × 16, 1 × 20, and 1 × 24 cluster parlors by 38, 45, and 50 min, respectively, when compared with parlors without ACR. This study highlights that the ability of automations to enhance milking efficiency and reduce labor requirements at milking varies across parlor size, emphasizing the need for strategic decision making in parlor configuration and operation.

The objective of this study was to document the milking efficiency of a sample of Irish dairy farms and to understand the effects of (1) seasonality, (2) management practices, (3) parlor infrastructure, and (4) parlor automations on milking efficiency metrics. A novel methodology based on empirical data from video cameras, infrastructure surveys, and milk yield data allowed for the accurate computation of milking efficiency metrics and quantification of the effects of seasonality, number of operators, and parlor automations on milking efficiency across 2 parlor types. The data for this study were collected over 2 periods: period 1 (July 28, 2020, to October 23, 2020, peak-late production) and period 2 (April 12, 2021, to May 19, 2021, early-peak production) from a sample of 16 herringbone and 10 rotary commercial Irish dairy farms. Milking efficiency was evaluated on each farm using 3 key performance indicators: (1) cows milked per hour (cows/h), (2) cows milked per operator per hour (cows/h per operator), and (3) liters of milk harvested per hour (L/h). Milking efficiency key performance indicators were calculated using "total process time," defined as the time between the first cow entering the holding yard and the end of the cleaning process. Average herd sizes for herringbone and rotary farms were 180 and 425 cows, respectively. Average system sizes for herringbone and rotary farms were 20 and 50 clusters, respectively. For herringbone farms, the average milking efficiency was 94 cows/h, 73 cows/h per operator, and 1,012 L/h, whereas rotary farms achieved an average milking efficiency of 170 cows/h, 132 cows/h per operator, and 1,534 L/h. Parlor size was strongly correlated with milking efficiency (cows/h) for herringbone parlors (0.91) but was only moderately correlated for rotary parlors (0.50). Hence, we documented the effect of parlor size on milking efficiency is relative to parlor type. Cluster utilization values on herringbone farms were 5 cows/cluster per h, 4 cows/cluster per operator per h, and 51 L/cluster per h, which were 67%, 33%, and 65% greater than rotary farms, respectively. We found for both herringbone and rotary farms hourly cow throughput (cows/h, cows/h per operator) were greatest during period 1 and that the volume of milk harvested per hour (L/h) was greatest for period 2. Thus, we documented an inverse seasonal relationship between hourly rates of cows milked and milk harvested. We observed that for herringbone farms, milking efficiency (cows/h, L/h) had a strong positive correlation (0.75, 0.74) with the levels of automation use. However, the minimal variation in automations used among rotary farms made it difficult to evaluate their effect on milking efficiency. Similarly, we found that the effect of automations on milking efficiency was dependent on parlor type. On average, a second operator at milking for both herringbone (H) and rotary (R) farms increased values for cows/h (+19%, H; +34%, R) and L/h (+21%, H; +12%, R) but lowered values for cows/h per operator (-35%, H; -12%, R). The holistic methodology applied in this study allowed us to add novel data to the literature by quantifying the effects of seasonality, the number of operators present at milking, and parlor automation use on milking efficiency across 2 parlor types.

The objective of this paper was to define, validate and demonstrate a model capable of accurately simulating dairy farm electricity consumption across varying herd and parlour sizes, to facilitate research investigating renewable energy systems (RES) and demand side management (DSM). The Farm Electricity System Simulator (FESS) was developed using grey -box modelling techniques utilizing empirical data for parameter tuning. Empirical data were gathered from nine spring calving, pasture based dairy farms located in the Republic of Ireland. A k -means clustering analysis was conducted, separating the farms into three, near homogenous groups, from which representative farms were selected. FESS was trained using 12 months of data from three representative farms using the repeat hold out method for data partitioning with 75 % of data used for training and 25 % used for validation. An optimisation algorithm was used to minimize the error during model training. Through cross -validation, FESS achieved a root mean squared error (RMSE) of 7.65 kWh, mean absolute percentage error (MAPE) of 7.10 %, mean percentage error (MPE) of -0.86 % and a relative prediction error (RPE) of 7.56 % for total daily electricity consumption. Across the three farms, the simulated outputs of FESS achieved an average R 2 value of 0.72, demonstrating good agreement with observed data. FESS ' s utility was demonstrated by analysing the effects of different electricity pricing structures and on -site solar photovoltaic electricity generation on total farm energy costs. We concluded that FESS simulated on -farm electricity consumption with sufficient accuracy for the intended application. FESS accurately simulated dairy farm electricity consumption across three dairy farms of different herd and parlour sizes while evaluating the effects of demand side management and renewable generation on farm electricity consumption and costs.

Livestock production is getting increased attention due to its impact on natural resources, and freshwater is one such limited resource. To reduce the pressure on freshwater use and develop sustainable livestock systems from farm-to-fork we need to study the whole production cycle, and look for hotspots of major freshwater use. Considering this, we chose intensive pork production as our focal livestock system, since it is one the most eaten meats globally. We focused on pork production in Ireland and studied the freshwater use (green and blue) from cradle-to-farm gate using the water footprint (WFP) method. Detailed farm data (e.g. diet composition, production data) were combined with on-farm water meter data to explore variations in water consumption between farms, and potential explanatory variables for differences in consumption between farms. So far, there have been no WFP studies in pork production that explored this, and insight into variation could help to identify options for improvement. We analyzed the direct (on-farm) and indirect (off-farm) green and blue water footprint of 10 Irish pig farms. Our results show that the average total WFP, including the direct and indirect water footprint, was 2537 L/kg pork, which is at the low end of previously published studies. The indirect green water footprint related to the production of purchased feed was responsible for the largest share (99 %) of the total WFP. The direct blue water footprint formed only a minor component of the total WFP (14 L/kg pork), with drinking water playing the major role. We can conclude from this study that variation in WFP between the least and most efficient farms was small (Q3-Q1 = 181 L/kg pork); nevertheless, this indicates that efficiencies of around 7 % could be gained by the least efficient cohort of farms by adjusting on-farm management practices. We also found a weak negative correlation between WFP and farm size, and WFP and meat produced. Overall, this study suggests that to reduce the burden on freshwater resources and reduce the pork WFP, future research should focus on the feed related impacts.

Background This cross-sectional study describes a survey designed to fill knowledge gaps regarding farm management practices, parlour management practices and implemented technologies, milking management practices, somatic cell count (SCC) control strategies, farmer demographics and attitudes around SCC management on a sample of Irish dairy farms.Results We categorized 376 complete responses by herd size quartile and calving pattern. The average respondent herd was 131 cows with most (82.2%) operating a seasonal calving system. The median monthly bulk tank somatic cell count for seasonal calving systems was 137,000 cells/ml (range 20,000 - 1,269,000 cells/ml), 170,000 cells/ml for split-calving systems (range 46,000 - 644,000 cells/ml) and 186,000 cells/ml for 'other' herds (range 20,000 - 664,000 cells/ml). The most common parlour types were swing-over herringbones (59.1%) and herringbones with recording jars (22.2%). The average number of units across herringbone parlours was 15, 49 in rotary parlours and two boxes on automatic milking system (AMS) farms. The most common parlour technologies were in-parlour feeding systems (84.5%), automatic washers on the bulk tank (72.8%), automatic cluster removers (57.9%), and entrance or exit gates controlled from the parlour pit (52.2%). Veterinary professionals, farming colleagues and processor milk quality advisors were the most commonly utilised sources of advice for SCC management (by 76.9%, 50.0% and 39.2% of respondents respectively).Conclusions In this study, we successfully utilised a national survey to quantify farm management practices, parlour management practices and technology adoption levels, milking management practices, SCC control strategies and farmer demographics on 376 dairy farms in the Republic of Ireland. Rotary and AMS parlours had the most parlour technologies of any parlour type. Technology add-ons were generally less prevalent on farms with smaller herds. Despite finding areas for improvement with regard to frequency of liner changes, glove-wearing practices and engagement with bacteriology of milk samples, we also found evidence of high levels of documentation of mastitis treatments and high use of post-milking teat disinfection. We discovered that Irish dairy farmers are relatively content in their careers but face pressures regarding changes to the legislation around prudent antimicrobial use in their herds.

Increasing levels of data are routinely collected on modern dairy farms. These include multiple variables measured by milking machine sensors and software and cow-attached sensor data, used predominantly for fertility and health monitoring. Following milking efficiency principles, including milking gently, quickly, and completely, there is utility in investigating how various milking machine settings affect gentleness of milking through a proxy measurement of cow comfort during milking. The use of leg-mounted accelerometers was investigated as a noninvasive labor-efficient means of estimating cow comfort on different automatic cluster remover (ACR) milk flow-rate switch-point settings. Accelerometer step count measurements during milking were collected from 37 cows divided into 2 groups allocated to either an ACR milk flow-rate switch-point setting of 0.2 kg/min or 0.8 kg/min for a 2-wk period and then crossed over to the other setting. Significantly more rear leg stepping occurred during daily milking (combined step count during a.m. and p.m. milkings) where the ACR activated at 0.2 kg/min (11.7 steps) compared with 0.8 kg/min (10.1 steps). Shorter milking interval between a.m. and p.m. milkings resulted in lower udder fill and reduced milk flow-rate. Under these lower udder fill conditions, rear leg movement, as an indicator of cow comfort, reduced when milk flow-rate switch-point for cluster removal increased from 0.2 kg/min (5.75 steps) to 0.8 kg/min (4.96 steps). There was no significant difference between stepping rates on both cluster removal settings during a.m. milkings. Similarly, no significant differences were noted in assessed postmilking teat condition, which was conducted after a.m. milking. The 0.2 kg/min setting extended total daily milking time by 70 s, resulting in lower mean flow-rates while producing similar milk yield. Higher vacuum levels at the teat-end were also recorded on this milking setting. This provides further incentive to consider cluster removal settings above 0.2 kg/min.

Mastitis is a significant disease on dairy farms and can have serious negative animal performance and economic consequences if not controlled. While clinical mastitis is often easily identified due to visibly abnormal milk, subclinical mastitis presents a more insidious challenge. Somatic cell count (SCC) is commonly used to monitor and detect subclinical mastitis, however, SCC is not available at a high sampling frequency rate at the cow level on most farms due to the manual effort involved in collecting it. With the rise of precision dairy farming technologies such as milk meters, however, there is increasing interest in using data-driven approaches (especially approaches using machine learning) for detecting subclinical mastitis based on indicators more easily collected by modern sensors. In this article we introduce milk flow profiles, a new, easy-to-collect data type that can replace more difficult-to-collect data sources (e.g., those that require laboratory tests or manual measurements) in precision dairy farming. The results of our experiments demonstrate that milk flow profiles, combined with other easily accessible milking machine data, can be employed to train machine learning models that accurately detect subclinical mastitis (as evidenced by high SCC measurements), with an AUC of 0.793. Moreover, these models perform better than models trained using features from milk characteristic data that are expensive to collect and are only collected at low frequency on commercial farms. Our experiments used data from 16 weeks of milking events from 285 cows on Irish farms, and their results demonstrate the value of milk flow profiles as an easily accessible and valuable data source for precision dairy farming applications.

International trends of increasing dairy herd sizes coupled with scarcity of labor have necessitated the enhancement of labor efficiency for dairy production systems. This study quantified the effects of infrastructure, automation, and management practices on the milking and operator efficiency of herringbone and rotary parlors used on pasture-based farms in Ireland. Data were used from 592 milkings across 26 farms (16 herringbones and 10 rotaries). The metrics of cows milked per hour (cows/h), cows milked per operator per hour (cows/h per operator), and liters of milk harvested per hour (L/h) described milking efficiency. The metrics of total process time per cow (TPT, s/cow), milk process time per cow (MPT, s/cow), work routine time (WRT, s/ cow), cluster time (CT, s/cluster), and attachment time per cow (ATC, s/cow) described operator efficiency. Automations investigated were backing gates, cluster flush, plant wash, cluster removers (ACR), feeders, entry gates, rapid exit, and teat spray. Additional operator presence at milking was also investigated. Herringbone and rotary parlors were assigned to quartiles from their cows/h per operator values to examine variations in infrastructure, automations, and management practices. Fourth quartile (Q4) herringbones based on cows/h per operator values averaged 93 cows/h per operator using average system sizes of 24 clusters with 5 parlor automations. The Q4 rotaries averaged 164 cows/h per operator using average system sizes of 47 clusters and an average CT of 13 s/ cluster. Cows/h per operator values for Q4 herringbone and rotary parlors were 82% and 54% higher, respectively, than values observed on first quartile parlors, indicating the considerable potential to improve efficiency. To determine if infrastructure, automations, or additional operators at milking significantly affected operator efficiencies, general linear mixed models were developed. For parlor infrastructure, additional clusters had greater significance on operator efficiencies (MPT) for herringbones (-1.3 s/cow) as opposed to rotaries (-0.2 s/cow). Hence, increases in system size were likely to result in improved efficiencies for herringbones but less so for rotaries. For automations, ACR significantly reduced herringbone TPT, CT, and WRT values by 13.3 s/cow, 18.9 s/cluster, and 32.6 s/cow, respectively, whereas rapid exit significantly lowered CT by 18.6 s/cluster. We found no significant effect on rotary TPT, MPT, CT, or WRT values from the use of automatic teat sprayers. An additional operator at milking was found to significantly reduce herringbone TPT but not MPT or CT. For rotaries, a second operator had no significant effect on TPT, MPT, CT, or WRT values. We documented strong negative correlations between operator efficiencies (TPT, MPT) and milking efficiency (cows/h) for both herringbone (-0.91, -0.84) and rotaries (-0.98, -0.89). Strong negative correlations between the herringbone automation count and TPT (-0.80), MPT (-0.72), and CT (-0.75) suggested highly automated parlors were likely to achieve greater operator efficiencies than less automated parlors. The strong negative correlation (-0.81) between rotary milking efficiency (cows/h) and CT suggested that lower CT values (i.e., rotation speed) resulted in increased milking efficiency. Overall, our study quantified the effects of parlor infrastructure, automation, and management practices on the milking and operator efficiency of herringbone and rotary parlors.

The United Nations Sustainable Development Goals aim to double the productivity of small -medium food producers (2015-2030), while food demand is estimated to increase by 60 % by 2050. The objectives of this paper were to identify and quantify the relationship between energy efficiency and milking efficiency, identify the main energy consuming processes associated with milking, and investigate whether milking efficiency, energy efficiency or the relationship between them varies depending on parlour type. Energy and milking efficiency data from 26 pasture-based dairy farms in the Republic of Ireland were analysed (17 herringbone, nine rotary). Energy consumption was monitored continuously on the herringbone farms and for two distinct, seven-day periods (observation periods 1 and 2) for the rotary farms. Milking performance was moni-tored for all 26 farms during these periods. During the observation periods, the rotary farms achieved superior energy efficiency (29.85 Wh kgMilk-1) and milking efficiency (152 cows/hour) than the herringbone farms (32.83 Wh kgMilk-1, 97 cows/hour). Moderate correlations existed between milking efficiency (cows/hour) and energy efficiency (Wh kgMilk -1 ) for rotary (r = -0.58, R2 = 0.34) and herringbone (r = -0.44, R2 = 0.19). These results indicated that higher levels of milking efficiency were moderately correlated with improved energy efficiency.

This study documents the effect of mechanical prestimulation on the milking duration of pasture-based cows in late lactation to better harness increased capacity of automation in the milk harvesting process. Premilking stimulation, provided via manual or mechanical means, has been shown to promote the milk letdown reflex and assist in achieving quick, comfortable, and complete milk removal from the udder. The literature is lacking knowledge on the effect of mechanical premilking stimulation on milking duration, especially in late lactation and in pasture-based systems, and many pasture-based farms do not practice a full premilking routine because of a lack of labor availability. The current study addresses this gap in knowledge. In this study, we tested 2 treatments: (1) the No Stim treatment used normal farm milking settings with no premilking preparation and (2) the Stim treatment used 60 s of mechanical premilking stimulation, with a rate of 120 cycles per minute and a pulsator ratio of 30:70 on cluster attachment. Once the 60 s of stimulation had elapsed, normal milking settings resumed for the remainder of the milking. Sixty cows were enrolled in the study, which ran for 20 d. The effect of treatment on a.m. milking duration was significant, a.m. milking duration for Stim was 12 s shorter than that of No Stim. The effect of treatment on p.m. milk duration was not significant. Treatment had no effect on a.m./p.m. milk yields, average milk flowrates or peak milk flowrates. Significant differences emerged between treatments on a.m. and p.m. dead time (time from cluster attachment to reach a milk flowrate of 0.2 kg/min). The a.m. and p.m. dead times were 6 s shorter for Stim compared with No Stim. The time taken to achieve peak milk flowrate (time to peak) at morning milking was 7 s shorter for Stim compared with No Stim, and treatment yielded no significant effects on time to peak at p.m. milkings. Treatment also had no significant effect on log10 somatic cell count. Although the percentage of congested teat-ends and teat-barrels was numerically lower for Stim versus No Stim, no statistical differences were detected across these measures. Based on the results of the study, we found merit in applying 60 s of mechanical pre-stimulation at a.m. milking from a milking duration perspective. However, the strategy was not as successful for the p.m. milking. Analysis of the milk flowrate profiles recorded during the study suggest potential utility in employing different machine settings for various milkings based on anticipated yield and level of udder fill.

The United Nations Sustainable Development Goals aim to double the productivity of small-medium food producers between 2015 and 2030 [1], while food demand is estimated to increase by 60% by 2050 [2]. The objectives of this paper were to identify and quantify the interactions between energy efficiency and milking efficiency, identify the main energy consuming processes associated with milking, and investigate whether milking efficiency, energy efficiency or the relationship between them varies depending on parlour type. This paper analysed energy and milking efficiency data from 26 pasture-based dairy farms in the Republic of Ireland. Of the 26 farms, 17 used herringbone parlours while nine used rotary parlours. Energy consumption was monitored continuously on the 17 herringbone farms and for two distinct, seven-day periods (observation periods 1 and 2) for the rotary farms. Milking performance was monitored for all 26 farms during these periods. Moderate correlations existed between milking efficiency (cows/hour) and energy efficiency (Wh/kgMilk) for rotary (r = -0.58) and herringbone (r = -0.44) farms. Energy efficiency was recorded for rotary (29.85 Wh/kgMilk) and herringbone (32.83 Wh/kgMilk) farms for the observation periods. These results indicated that higher levels of milking efficiency correlated with improved levels of energy efficiency.

Livestock feed production is one of the primary users of freshwater and arable land, and it is also in competition with human food production. Therefore, we require reconsideration of the way we use freshwater in livestock feed production. The objective of this study is to assess the impact on freshwater use of pork production by using alternative pig diets based on local feed ingredients, or by-products. We used a lifecycle approach to analyse the freshwater use associated with feed production to produce one kg of pork. We explored three feeding scenarios (STANDARD: diets commercially used in Ireland; LOCAL: diets based on ingredients grown in Ireland; and BY-PRODUCT: diets based on by-products only). We calculated the freshwater use, using the water footprint (WFP) method, and the competition for water use between food and feed production using the water use ratio (WUR) for each scenario. The WUR quantifies the maximum amount of human digestible protein (HDP) derived from food crops that could be produced on the same land, and using the same water resources, that were used to grow the feed ingredients needed to produce 1 kg of pork. The WFP of the scenarios was 2,470 L/kg pork for STANDARD, 2,492 L/kg pork for LOCAL, and 2,205 L/kg pork for BY-PRODUCT. When we considered the WUR, none of the scenarios had a value < 1 (i.e. in all scenarios, more HDP can be produced from direct cultivation of food crops rather than pork). However, the BY-PRODUCT scenario (1.4) performed better than STANDARD (1.9) and LOCAL (2.9). Beet pulp and bakery by-products had zero WFP and no edibility and were thus considered promising ingredients. Moreover, rapeseed meal had a low WFP and rapeseed meal and sunflower seed meal are not considered human edible and were considered fit for future inclusion in diets. We also concluded that both the WFP and WUR methods have separate strengths and limitations, and should thus be used in conjunction; the ideal diet is one with the minimum WFP and WUR. Consideration of human edibility of feed ingredients is an important approach which should be included in future studies. Moreover, the entire food system including dairy, beef, poultry and other competitive uses should be taken into account when considering which feed ingredients to use in pig diets.

The objective of this study was to document the level of milking efficiency on a sample of Irish dairy farms with respect to the effects of seasonality, infrastructure, automation and management. A multifaceted methodology involving the use of cameras, milk yield databases and infrastructure surveys was developed to achieve this objective. Three milking efficiency metrics were used; cows milked per hour (cows/h), cows milked per operator per hour (cows/op/h) and liters of milk harvested per hour (L/h). A sample of 17 herringbone and 10 rotary commercial dairy farms were selected. The herringbone farms had an average of 180 milking cows and an average milking parlor system size of 20 clusters. The rotary farms had an average of 425 milking cows and an average milking parlor system size of 50 clusters. Data were collected over two separate one week periods to cover the seasonal effect on milking efficiency. Four video cameras were installed on each farm during the recording periods. Camera positioning was informed by milking parlor type and milking process KPIs. The mean milking efficiency of herringbone farms was 94 cows/h, 70 cows/op/h and 1,015 L/h. The mean milking efficiency of rotary farms was 170 cows/h, 132 cows/op/h and 1,534 L/h. Hence, owing to detailed data collect via the method described here, we found that the average milking efficiency of rotary farms was 51% and 89% more efficient for L/h and cows/h respectively. This study highlights the benefits of collecting data on milking efficiency especially when farms are designing new milking systems or planning for future labor requirements.