The US dairy sector is facing structural changes including a geographical shift in dairy production and a tendency toward the implementation of more intensive production systems. These changes might significantly affect farm efficiency, profitability, and the long-term economic sustainability of the dairy sector, especially in more traditional dairy production areas. Consequently, the goal of this study was to examine the effect of practices commonly used by dairy farmers and the effect of intensification on the performance of the farms. We used a sample of 273 Wisconsin dairy farms to estimate a stochastic production frontier simultaneously with a technical inefficiency model. The empirical analysis showed that at a commercial level the administration of bovine somatotropin hormone to lactating cows increased milk production. In addition, we found that production exhibits constant returns to scale and that farm efficiency is positively related to farm intensification, the level of contribution of family labor in the farm activities, the use of a total mixed ration feeding system, and milking frequency.

When making the decision about whether or not to breed a given cow, knowledge about the expected outcome would have an economic impact on profitability of the breeding program and net income of the farm. The outcome of each breeding can be affected by many management and physiological features that vary between farms and interact with each other. Hence, the ability of machine learning algorithms to accommodate complex relationships in the data and missing values for explanatory variables makes these algorithms well suited for investigation of reproduction performance in dairy cattle. The objective of this study was to develop a user-friendly and intuitive on-farm tool to help farmers make reproduction management decisions. Several different machine learning algorithms were applied to predict the insemination outcomes of individual cows based on phenotypic and genotypic data. Data from 26 dairy farms in the Alta Genetics (Watertown, WI) Advantage Progeny Testing Program were used, representing a 10-yr period from 2000 to 2010. Health, reproduction, and production data were extracted from on-farm dairy management software, and estimated breeding values were downloaded from the US Department of Agriculture Agricultural Research Service Animal Improvement Programs Laboratory (Beltsville, MD) database. The edited data set consisted of 129,245 breeding records from primiparous Holstein cows and 195,128 breeding records from multiparous Holstein cows. Each data point in the final data set included 23 and 25 explanatory variables and 1 binary outcome for of 0.756 ± 0.005 and 0.736 ± 0.005 for primiparous and multiparous cows, respectively. The naïve Bayes algorithm, Bayesian network, and decision tree algorithms showed somewhat poorer classification performance. An information-based variable selection procedure identified herd average conception rate, incidence of ketosis, number of previous (failed) inseminations, days in milk at breeding, and mastitis as the most effective explanatory variables in predicting pregnancy outcome.

Our objective was to compare the economic and reproductive performance of programs combining timed artificial insemination (TAI) and different levels of AI after estrus detection (ED) using a daily Markov-chain model. A dairy herd was modeled with every cow following daily probabilistic events of aging, replacement, mortality, pregnancy, pregnancy loss, and calving. The probability of pregnancy depended on the combination of probability of insemination and conception rate (CR). All nonpregnant cows had a probability of pregnancy between the end of the voluntary waiting period and days in milk cutoff for AI. After the cutoff, cows were labeled as do not breed and replaced when milk production was below a minimum milk threshold. A similar model was created to represent a replacement heifer herd to simulate and adjust the supply and demand of replacements. The net value (NV) of a program was the sum of milk income over feed cost, replacement and mortality cost, income from newborns, and reproductive costs. The model was used to compare the NV of 19 programs. One program used 100% TAI (42% CR for first TAI and 30% for second-and-later services), whereas the other programs combined TAI with ED. The proportion of cows receiving AI after ED for the combined programs ranged from 30 to 80%, with levels of CR of 25, 30, and 35%. As the proportion of cows receiving AI after ED increased, the CR of cows receiving TAI decreased. The combined programs with CR of 35% for cows receiving AI after ED had the greatest NV and reproductive performance at all levels of ED. The program using 100% TAI had greater NV and better reproductive performance than all programs with 25% CR after ED inseminations, whereas it had very similar performance to combined programs with up to 60% of cows receiving AI after ED and 30% CR. The factor with the greatest relative contribution to the differences among programs was income over feed cost, followed by replacement and reproductive costs. Adjusting the days in milk cutoff for AI to match the supply and demand of heifer replacements improved the NV of all programs except for those with 25% CR after ED, which had either no change or a decrease in NV. In summary, the economic value of reproductive management programs combining TAI and ED depended on the proportion of cows receiving AI after ED and the resulting CR. Adjusting the heifer supply and demand increased the NV of programs with heifer surplus and decreased the NV of programs with heifer deficit.

Because the reproductive performance of lactating dairy cows influences the profitability of dairy operations, predicting the future reproductive and economic performance of dairy herds through decision support systems would be valuable to dairy producers and consultants. In this study, we present a highly adaptable tool created based on a mathematical model combining Markov chain simulation with partial budgeting to obtain the net present value (NPV; $/cow per year) of different reproductive management programs. The growing complexity of reproductive programs used by dairy farms demands that new decision support systems precisely reflect the events that occur on the farm. Therefore, the model requires productive, reproductive, and economic input data used for simulation of farm conditions to account for all factors related to reproductive management that increase costs and generate revenue. The economic performance of 3 different reproductive programs can be simultaneously compared with the current model. A program utilizing 100% visual estrous detection (ED) for artificial insemination (AI) is used as a baseline for comparison with 2 other programs that may include 100% timed AI (TAI) as well as any combination of TAI and ED. A case study is presented in which the model was used to compare 3 different reproductive management strategies (100% ED baseline compared with two 100% TAI options) using data from a commercial farm in Wisconsin. Sensitivity analysis was then used to assess the effect of varying specific reproductive parameters on the NPV. Under the simulated conditions of the case study, the model indicated that the two 100% TAI programs were superior to the 100% ED program and, of the 100% TAI programs, the one with the higher conception rate (CR) for resynchronized AI services was economically superior despite having higher costs and a longer interbreeding interval. A 4% increase in CR for resynchronized AI was sufficient for the inferior 100% TAI to outperform the superior program. Adding ED to the 100% TAI programs was only beneficial for the program with the lower CR. The improvement in service rate required for the 100% ED program to have the same NPV as the superior 100% TAI program was 12%. The decision support system developed in this study is a valuable tool that may be used to assist dairy producers and industry consultants in selecting the best farm-specific reproductive management strategy.

The objective of this review paper was to summarise the latest findings in dairy cattle reproductive economics with an emphasis on high yielding, confined total mixed ration systems. The economic gain increases as the reproductive efficiency improves. These increments follow the law of diminishing returns, but are still positive even at high reproductive performance. Reproductive improvement results in higher milk productivity and, therefore, higher milk income over feed cost, more calf sales and lower culling and breeding expenses. Most high-yielding herds in the United States use a combination of timed artificial insemination (TAI) and oestrous detection (OD) reproductive programme. The ratio of achievable pregnancies between OD and TAI determines the economic value difference between both and their combinations. Nonetheless, complex interactions between reproductive programme, herd relative milk yield, and type of reproductive programme are reported. For example, higher herd relative milk yield would favour programme relying more on TAI. In addition, improved reproductive efficiency produces extra replacements. The availability of additional replacements could allow more aggressive culling policies (e.g. less services for non-pregnant cows) to balance on-farm supply and demand of replacements. Balancing heifer replacement availability in an efficient reproductive programme brings additional economic benefits. New technologies such as the use of earlier chemical tests for pregnancy diagnosis could be economically effective depending on the goals and characteristics of the farm. Opportunities for individual cow reproductive management within defined reproductive programme exist. These decisions would be based on economic metrics derived from the value of a cow such as the value of a new pregnancy, the cost of a pregnancy loss, or the cost of an extra day open.