Optimal replacement policies for dairy cows based on daily yield measurements

Markov decision processes (MDP) with finite state and action space have often been used to model sequential decision making over time in dairy herds. However, the length of each stage has been at least 1 mo, resulting in models that do not support decisions on a daily basis. The present paper describes the first step of developing an MDP model that can be integrated into a modern herd management system. A hierarchical MDP was formulated for the dairy cow replacement problem with stage lengths of 1 d. It can be used to assist the farmer in replacement decisions on a daily basis and is based on daily milk yield measurements that are available in modern milking systems. Bayesian updating was used to predict the performance of each cow in the herd and economic decisions were based on the prediction. Moreover, parameters in the model were estimated using data records of the specific herd under consideration. This includes herd-specific lactation curves.     

Optimally achieving milk bulk tank somatic cell count thresholds

High somatic cell count in milk leads to reduced shelf life in fluid milk and lower processed yields in manufactured dairy products. As a result, farmers are often penalized for high bulk tank somatic cell count or paid a premium for low bulk tank somatic cell count. Many countries also require all milk from a farm to be lower than a specified regulated somatic cell count. Thus, farms often cull cows that have high somatic cell count to meet somatic cell count thresholds. Rather than naïvely cull the highest somatic cell count cows, a mathematical programming model was developed that determines the cows to be culled from the herd by maximizing the net present value of the herd, subject to meeting any specified bulk tank somatic cell count level. The model was applied to test-day cows on 2 New York State dairy farms. Results showed that the net present value of the herd was increased by using the model to meet the somatic cell count restriction compared with naïvely culling the highest somatic cell count cows. Implementation of the model would be straightforward in dairy management decision software.     

Dairy farmers with larger herd sizes adopt more precision dairy technologies

An increase in the average herd size on Australian dairy farms has also increased the labor and animal management pressure on farmers, thus potentially encouraging the adoption of precision technologies for enhanced management control. A survey was undertaken in 2015 in Australia to identify the relationship between herd size, current precision technology adoption, and perception of the future of precision technologies. Additionally, differences between farmers and service providers in relation to perception of future precision technology adoption were also investigated. Responses from 199 dairy farmers, and 102 service providers, were collected between May and August 2015 via an anonymous Internet-based questionnaire. Of the 199 dairy farmer responses, 10.4% corresponded to farms that had fewer than 150 cows, 37.7% had 151 to 300 cows, 35.5% had 301 to 500 cows; 6.0% had 501 to 700 cows, and 10.4% had more than 701 cows. The results showed that farmers with more than 500 cows adopted between 2 and 5 times more specific precision technologies, such as automatic cup removers, automatic milk plant wash systems, electronic cow identification systems and herd management software, when compared with smaller farms. Only minor differences were detected in perception of the future of precision technologies between either herd size or farmers and service providers. In particular, service providers expected a higher adoption of automatic milking and walk over weighing systems than farmers. Currently, the adoption of precision technology has mostly been of the type that reduces labor needs; however, respondents indicated that by 2025 adoption of data capturing technology for monitoring farm system parameters would be increased.     

Management Guides for Insemination and Replacement Decisions

Decisions to inseminate and replace cows should be based on expected future income. Two management guides are described to help dairy farmers make decisions 1) profit after conception for decisions on insemination and 2) future profitability for decisions on immediate replacement and veterinary treatment. Management guides are obtained by comparing expected cash flow of a cow currently in the herd with that of a replacement heifer entering the herd at different times. Besides optimum decisions, management guides indicate expected loss when making an nonoptimum decision. This allows farmers to weigh additional information on cow or herd and to rank cows on expected profitability.Dynamic programming can be used to calculate management guides for all possible states of a cow for a given price and production situation. Results can be used to derive management guides for individual cows in a herd, characterizing each cow by state variables. Most information needed on cows can be taken from milk recording data and breeding records. Lactation number, milk production, and stage of lactation clearly affect the management guides.     

Symposium review: Use of multiple biological,management, and performance data for the design of targeted reproductive management strategies for dairy cows

As the reproductive efficiency of dairy cattle continues to improve in response to better management and use of technology, novel reproductive management approaches will be required to improve herd performance, profitability, and sustainability. A potential approach currently being explored is targeted reproductive management. This approach consists of identifying cows with different reproductive and performance potential using multiple traditional and novel sources of biological, management, and performance data. Once subgroups of cows that share biological and performance features are identified, reproductive management strategies specifically designed to optimize cow performance, herd profitability, or alternative outcomes of interest are implemented on different subgroups of cows. Tailoring reproductive management to subgroups of cows is expected to generate greater gains in outcomes of interest than if the whole herd is under similar management. Major steps in the development and implementation of targeted reproductive management programs for dairy cattle include identification and validation of robust predictors of reproductive outcomes and cow performance, and the development and on-farm evaluation of reproductive management strategies for optimizing outcomes of interest for subgroups of cows. Predictors of cow performance currently explored for use in targeted management include genomic predictions; behavioral, physiological, and performance parameters monitored by sensor technologies; and individual cow and herd performance records. Once the most valuable predictive sources of variation are identified and their effects quantified, novel analytic methods (e.g., machine learning) for prediction will likely be required. These tools must identify groups of cows for targeted management in real time and with no human input. Despite some encouraging research evidence supporting the development of targeted reproductive management strategies, extensive work is required before widespread implementation by commercial farms.     

Understanding the adoption of smartphone apps in dairy herd management

The number of decision support tools available to farmers, including dairy herd management smartphone apps, has been steadily increasing. The existing literature does not cover topics concerning the adoption and use of herd management smartphone apps or which specific functions of such apps are perceived as most useful by dairy farmers. It is unclear whether technology adoption can only be explained by economic reasoning, because the beliefs about a technology also play a role in decision-making. Therefore, this study seeks to determine whether an extended technology acceptance model can explain adoption and use of herd management smartphone apps. Results about the adoption and use of dairy herd management smartphone apps are derived from an online survey conducted in 2018 with 280 German dairy farmers. To model farmers' frequency of use of herd management smartphone apps, we applied partial least squares structural equation modeling and an ordered logit model. Our results show that 93% of the dairy farmers in our sample use a smartphone and 61% already use a herd management smartphone app. Daily use is reported by 38% of the adopters. Dairy farmers rated functions related to the observation of animal health, reproduction management, and data gathering as most useful, which should be in focus by developers and providers for future development. The key attitudinal components of the technology acceptance model, namely perceived ease of use and perceived usefulness, both positively influence the intention to use such apps. This ultimately has a positive effect on the actual usage behavior. Besides other factors, dairy farmers' education and knowledge of herd management smartphone apps have a positive effect on perceived ease of use. Our model explains 33% of the variance in the actual usage behavior related to herd management smartphone apps. Because perceived ease of use and perceived usefulness positively influence the intention to use such apps and ultimately the actual usage behavior, developers and providers should highlight the benefits of using herd management smartphone apps and also keep the interface of the apps as simple as possible.     

Dairy cow longevity: Impact of animal health and farmers' investment decisions

A dairy farmer's decision to cull or keep dairy cows is likely a complex decision based on animal health and farm management practices. The present paper investigated the relationship between cow longevity and animal health, and between longevity and farm investments, while controlling for farm-specific characteristics and animal management practices, by using Swedish dairy farm and production data for the period 2009 to 2018. We used the ordinary least square and unconditional quantile regression model to perform mean-based and heterogeneous-based analysis, respectively. Findings from the study indicate that, on average, animal health has a negative but insignificant effect on dairy herd longevity. This implies that culling is predominantly done for other reasons than poor health status. Investment in farm infrastructure has a positive and significant effect on dairy herd longevity. The investment in farm infrastructure creates room for new or superior recruitment heifers without the need to cull existing dairy cows. Production variables that prolong dairy cow longevity include higher milk yield and an extended calving interval. Findings from this study imply that the relatively short longevity of dairy cows in Sweden compared with some dairy producing countries is not a result of problems with health and welfare. Rather, dairy cow longevity in Sweden hinges on the farmers' investment decisions, farm-specific characteristics and animal management practices.     

Use of herd solutions from a random regression test-day model for diagnostic dairy herd management

In a random regression test-day model, environmental effects in addition to individual animal factors can be included and analyzed. Moreover, instead of herd-year classification of the management groups, the herd-test-day classification within the model better accounts for month-to-month short-term environmental variation in production and somatic cell count (SCC) traits. The herd management levels of milk yield (milk deviation from whole-country mean, kilograms/day), protein and fat concentration (protein and fat deviation, %), and SCC (SCC deviation, 1,000 cells/mL) are used in the dairy herd management Web application “Maitoisa” (in English, “Milky”). This management tool helps to recognize several management problems. For recognition of systematic patterns and single unusual test-days, a monthly time-trend analysis was developed to smooth the random fluctuations and display the yearly production pattern. In addition to analyzing single test-day deviations from the mean, modeled herd solutions assist users in identifying repeated phenomena and enable the forecasting of the management pattern for the subsequent year. The solutions are displayed either as tables or graphs plotted by calendar months. In addition to management effects of the farmer's own herd, he or she can request country or region percentiles to be displayed in the graphs. The Web service has been offered to farmers and dairy advisors since 2001, and it has proved to be a powerful tool for herd monitoring and planning.     

A computerized mastitis decision aid using farm-based records: an artificial neural network approach

A computer module was developed and tested that used field survey and Dairy Herd Improvement Association (DHIA) data to broadly classify bacterial causes of mastitis in dairy herds. Further development of the computer model could aid interpretation of DHIA data by dairy record processing centers and herd consultants. This diagnostic module was developed with an artificial neural network, a technology that processes complex data in a manner similar to human brain function. Information describing herd management practices, quarter milk samples, and monthly DHIA data was collected from Pennsylvania dairy herds with moderate to high somatic cell counts. This information was used to develop or train an artificial neural network model that discriminated among four categories of bacterial organisms (contagious, environmental, no significant growth, and other) associated with clinical and subclinical mastitis. After training the model, new DHIA and management data were presented to the model to assess its ability to classify bacteriological etiology. When the artificial neural network was used, the probabilities of diagnosing the bacteriologic status from three randomly selected cow groups and from new untested herds ranged from 57 to 71%. Performance of the artificial neural network model was best in herds with higher frequency of minor and contagious pathogens. Prediction results for the same test data with linear discriminant analysis were less successful, ranging from 42 to 57%.     

Comparison of different applications of automatic herd control systems on dairy farms – a review

Recent years have seen the rapid development of different devices which can be helpful in the daily work of livestock farmers. The growing size of livestock herds has led farmers to lose individual contact with their animals, while behavioral studies show that breeders can effectively and precisely monitor a herd of up to 100 cows. This was the main motivation for this study, which aims to identify and test various electronic devices which provide useful herd management data, including estrus detection, individual activity and body temperature measurement, monitoring rumen pH levels, milk quality and content as well as milk temperature and somatic cell count measurements. Some devices can detect the metabolic status of animals with a reasonable level of precision. Contemporary animal farms are offered a large number of systems for monitoring the behavior of the animals in the herd and helping to identify those that are intended for insemination or are too active or excessively apathetic. Monitoring devices support herd management and help to reduce costs through the early detection of animal diseases and nutritional problems. This review aims to compile and summarize the information currently available on the use of automatic herd control systems on dairy farms, as well as to discuss the interpretation of the results, providing a useful diagnostic tool in nutritional evaluations of dairy herds. © 2018 Society of Chemical Industry     

Prediction of dairy housing construction costs

Dairy farms in Galicia and elsewhere in Europe are going through a transition phase to adapt to modern dairy technology, improve labor efficiency, and increase in size and scale. Expanding a dairy herd and building housing for more cows can be very expensive. A poor decision during expansion can result in serious financial difficulties even to the point of making the farm economically unviable. Dairy managers must carefully evaluate existing alternatives and must select an optimal strategy. To aid this decision, a computer spreadsheet application has been developed that predicts the cost per cow and cost per unit of area of alternative designs as functions of the number of cows to be housed. The spreadsheet is, in principle, applicable to a wide variety of designs and to housing for livestock other than dairy cattle. However, the current database allows comparison among six of the dairy housing designs that have been used most widely in Galicia in recent years. From projected financial results of the developed model, it was concluded that differing designs were preferred for different farm circumstances. Preferred designs for farms with 60 to 200 cows were either four rows of facing free stalls or four rows of tail-to-tail free stalls, which have virtually the same costs. Whereas for farms with fewer than 60 cows, the preferred design was two rows of tail-to-tail free stalls, designs with three rows of free stalls were generally more costly per cow. Results of design calculations must be integrated with other farm management considerations in choosing a particular design.     

Method to monitor the precision of milk yields recorded at individual milking stalls on daily basis.

Automated systems to record individual milk weights of cows at each milking are now common on dairy farms. These data are used by Dairy Herd Improvement Association (DHIA) for testing purposes and by farmers to monitor herd performance. Despite the importance of these data, little effort has been made to monitor the function and reliability of milking stalls and recording equipment on a daily basis. Described is an algorithm that allows for continuous monitoring of the function of individual milking stalls. The basis of the algorithm is the comparison of milk yield of individual cows over a period of days. For each cow, milk production at a selected stall is compared with production of that same cow when milked at any other stall. These ratios are averaged across all cows milked at least once in the selected stall and once at any other stall. Ideally, the mean ratio for each stall should be 1. Stalls where the mean ratio deviates from 1 by more than 5% are assumed to be malfunctioning. This algorithm was implemented as a computer program and was used to identify 3 malfunctioning stalls out of 12 at the USDA dairy facility (Beltsville, MD). The algorithm can easily be incorporated into existing management programs or used as a separate program. In either case, routine use of this algorithm can improve the reliability of DHIA test programs and potentially reduce the incidence of new mastitis infections by enabling detection of malfunctioning milking equipment.     

Invited review: Methods to determine body fat reserves in the dairy cow with special regard to ultrasonographic measurement of backfat thickness

As the dairy cow uses body energy reserves in early lactation, body condition scoring has become an integral part of dairy herd management. Several methods based on visual and tactile evaluation have been developed. Problems caused by the subjectivity of these techniques have been reported. Alternative approaches to predict energy reserves or energy balance in dairy cattle include metabolic profiling and measurement of live weight, heart girth, or skinfold thickness. A less common method to assess fat reserves in body tissues is measuring backfat thickness (BFT) by using ultrasound. An ultrasound technique has been established to predict carcass quality in beef cattle. A new aspect is the application of ultrasound as a monitoring tool in dairy herd management where another location has to be evaluated. This technique has been validated by relating BFT to total body fat (TBF) content and carcass BFT. Backfat thickness also has been related to other methods of body condition scoring. Target values for the development of BFT throughout lactation are available. The relationship between BFT and TBF content is highly significant although biased by multiple factors. A change in BFT of 1 mm equates to approximately 5 kg of TBF content. Measuring BFT by ultrasound is of added value compared with other body condition scoring systems because it is objective and precise. Changes in body condition can be detected and evaluated properly.     

Invited review: Culling: nomenclature, definitions, and recommendations

Replacing cows on a dairy is a major cost of operation. There is a need for the industry to adopt a more standardized approach to reporting the rate at which cows exit from the dairy, and to reporting the reasons why cows are replaced and their destination as they exit the dairy. Herd turnover rate is recommended as the preferred term for characterizing the cows exiting a dairy, in preference to herd replacement rate, culling rate, or percent exiting, all of which have served as synonyms. Herd turnover rate should be calculated as the number of cows that exit in a defined period divided by the animal time at risk for the population being characterized. The terms voluntary and involuntary culling suffer from problems of definition and their use should be discouraged. Destination should be recorded for all cows that exit the dairy and opportunities to record one or more reasons for exiting should be provided by management systems. Comparing reported reasons between dairies requires considerable caution because of differences in case definitions and recording methods. Relying upon culling records to monitor disease has been and will always be an ineffective management strategy. Dairies are encouraged to record and monitor disease events and reproductive performance and use this information as the basis for management efforts aimed at reducing the need to replace cows.     

Invited review: associations between variables of routine herd data and dairy cattle welfare indicators

As farm animal welfare is high on the political and societal agendas of many countries, considerable pressure exists to establish audit programs in which farm animal welfare is routinely monitored. On-farm assessment of animal welfare, however, is time-consuming and costly. A promising strategy to monitor animal welfare more efficiently is to first estimate the level of animal welfare on a farm based on routine herd data that are available in national databases. It is not currently known which variables of routine herd data (VRHD) are associated with dairy cattle welfare indicators (WI). Our aim was to identify VRHD that are associated with WI in a literature review. The 27 VRHD used in this review included the main types of data that are currently collected in national herd databases of developed countries, and related to identification and registration, management, milk production, and reproduction of dairy herds. The 34 WI used in this review were based on the Welfare Quality Assessment Protocol for Cattle. The search yielded associations in 146 studies. Twenty-three VRHD were associated with 16 WI. The VRHD that related to milk yield, culling, and reproduction were associated with the largest number of WI. Few associations were found for WI that referred to behavioral aspects of animal welfare, nonspecific disease symptoms, or resources-based indicators. For 18 WI, associations with VRHD were not significant (n = 5 WI) or no studies were found that investigated associations with VRHD (n = 13 WI). It was concluded that many VRHD have potential to estimate the level of animal welfare on dairy farms. As strengths of associations were not considered in this review, however, the true value of these VRHD should be further explored. Moreover, associations found at the animal level and in an experimental setting might not appear at the farm level and in common practice and should be investigated. Cross-sectional studies using integrated welfare scores at the farm level are needed to more accurately determine the potential of VRHD to estimate levels of animal welfare on dairy farms.     

Factors influencing dairy calf and replacement heifer mortality in France

Herd-level risk factors for dairy calf and heifer mortality in France were identified by calculating herd-level variables (including mortality risk or rate) using the National Bovine Identification Database (2005 and 2006). Eleven dairy production areas representing different livestock systems were also included. Statistical analyses were based on a probit model (mortality risk or rate = 0 or >0) and a linear model (mortality risk or rate >0) corrected by the sample bias Heckman method. The same associations were reported for 2005 and 2006. The mortality risks or rates for calves and heifers were positively associated with the proportion of purchased cows or being a Milk Control Program member and negatively associated with adhering to the Good Breeding Practices charter and having an autumn calving peak. The associations between mortality and the breeds or the production areas were positive or negative, depending on the classes of animal. Mortality and having a beef herd in addition to the dairy herd were negatively associated for noncrossed birth to 2-d-old calves, noncrossed 3-d- to 1-mo-old calves, and 3-d- to 1-mo-old heifers. Having a beef herd probably provides specific know-how related to newborn and young calf management that makes it easier to attain low mortality in pure-breed dairy calves. The proportion of males born was positively associated with mortality for the birth to 2-d-old calves (all classes) and for the 3-d- to 1-mo-old beef-crossed calves, but negatively for all classes of heifers. This indicates that heifer management was improved when the availability of newborn heifers decreased, resulting in low mortality. This lower mortality is apparent for all classes of heifers present on the farm during the year when the proportion of males was low, and demonstrates an anticipatory effect. In conclusion, this study shows that the presence of a beef herd in addition to the dairy herd within a farm is associated with decreased dairy calf mortality. It also shows that heifer mortality decreases when the proportion of heifers born decreased. These determinants of dairy calf and heifer mortality are of great importance for farmer advisors, the dairy industry, and the political decision makers.     

Performance of automated activity monitoring systems used in combination with timed artificial insemination compared to timed artificial insemination only in early lactation in dairy cows

Identifying cows in estrus remains a challenge on dairy cattle farms, and tools and technologies have been developed and used to complement or replace visual detection of estrus. Automated activity monitoring (AAM) systems and timed artificial insemination (TAI) are technologies available to dairy farmers, but many factors can influence their relative performance. The objective of the present study was to compare reproductive performance of cows managed with an AAM system combined with TAI, or with a TAI program (Double Ovsynch) for insemination before 88 DIM. From April 2014 to April 2015, 998 cows from 2 herds were randomly assigned either to be inseminated at 85 ± 3 DIM exclusively using the Double Ovsynch protocol for TAI, or to be inseminated based on estrus detection by AAM without hormonal intervention between 50 and 75 DIM; if no alarm was detected by 75 DIM, cows were inseminated following the single Ovsynch protocol (AAM + Ovsynch). The herds used different AAM systems. Parity, hyperketonemia at wk 1 and 2 postpartum (PP), purulent vaginal discharge at wk 5 PP, body condition score at wk 7 PP, and anovulation to wk 9 PP were recorded. These health indicators did not significantly differ between treatments, but did between herds. The effect of treatment on pregnancy at first insemination and by 88 DIM were assessed using logistic regression models. Time to pregnancy was assessed using survival analysis. Results are reported from intention-to-treat analyses. Treatment did not affect pregnancy at first insemination or pregnancy by 88 DIM, but we found significant interactions between treatment and herd for both outcomes. In herd 2, marginal mean pregnancy at first AI was greater with Double Ovsynch (38%) than AAM + Ovsynch (31%), but no difference was observed in herd 1 (Double Ovsynch = 31%; AAM + Ovsynch = 34%). By 88 DIM, a smaller proportion of cows in herd 1 were pregnant in Double Ovsynch (31%) than AAM + Ovsynch (49%), but there was no difference in herd 2 (Double Ovsynch = 38%; AAM + Ovsynch = 38%). We observed a treatment by herd interaction for median (95% confidence interval) time to pregnancy, which were, in herd 1, 110 (106 to 129) and 98 (88 to 113) d, and, in herd 2, 126 (113 to 139) and 116 (105 to 131) d for the Double Ovsynch and AAM + Ovsynch treatments, respectively. The relative performance of AAM-based reproductive management compared with TAI only is likely influenced by herd-specific variables, in particular related to insemination rate when estrus detection is employed.