Since the 1980s, efforts have been made to develop sensors that measure a parameter from an individual cow. The development started with individual cow recognition and was followed by sensors that measure the electrical conductivity of milk and pedometers that measure activity. The aim of this review is to provide a structured overview of the published sensor systems for dairy health management. The development of sensor systems can be described by the following 4 levels: (I) techniques that measure something about the cow (e.g., activity); (II) interpretations that summarize changes in the sensor data (e.g., increase in activity) to produce information about the cow's status (e.g., estrus); (III) integration of information where sensor information is supplemented with other information (e.g., economic information) to produce advice (e.g., whether to inseminate a cow or not); and (IV) the farmer makes a decision or the sensor system makes the decision autonomously (e.g., the inseminator is called). This review has structured a total of 126 publications describing 139 sensor systems and compared them based on the 4 levels. The publications were published in the Thomson Reuters (formerly ISI) Web of Science database from January 2002 until June 2012 or in the proceedings of 3 conferences on precision (dairy) farming in 2009, 2010, and 2011. Most studies concerned the detection of mastitis (25%), fertility (33%), and locomotion problems (30%), with fewer studies (16%) related to the detection of metabolic problems. Many studies presented sensor systems at levels I and II, but none did so at levels III and IV. Most of the work for mastitis (92%) and fertility (75%) is done at level II. For locomotion (53%) and metabolism (69%), more than half of the work is done at level I. The performance of sensor systems varies based on the choice of gold standards, algorithms, and test sizes (number of farms and cows). Studies on sensor systems for mastitis and estrus have shown that sensor systems are brought to a higher level; however, the need to improve detection performance still exists. Studies on sensor systems for locomotion problems have shown that the search continues for the most appropriate indicators, sensor techniques, and gold standards. Studies on metabolic problems show that it is still unclear which indicator reflects best the metabolic problems that should be detected. No systems with integrated decision support models have been found.

The technical performance of activity meters for automated detection of estrus in dairy farming has been studied, and such meters are already used in practice. However, information on the economic consequences of using activity meters is lacking. The current study analyzes the economic benefits of a sensor system for detection of estrus and appraises the feasibility of an investment in such a system. A stochastic dynamic simulation model was used to simulate reproductive performance of a dairy herd. The number of cow places in this herd was fixed at 130. The model started with 130 randomly drawn cows (in a Monte Carlo process) and simulated calvings and replacement of these cows in subsequent years. Default herd characteristics were a conception rate of 50%, an 8-wk dry-off period, and an average milk production level of 8,310 kg per cow per 305 d. Model inputs were derived from real farm data and expertise. For the analysis, visual detection by the farmer ("without" situation) was compared with automated detection with activity meters ("with" situation). For visual estrus detection, an estrus detection rate of 50% and a specificity of 100% were assumed. For automated estrus detection, an estrus detection rate of 80% and a specificity of 95% were assumed. The results of the cow simulation model were used to estimate the difference between the annual net cash flows in the "with" and "without" situations (marginal financial effect) and the internal rate of return (IRR) as profitability indicators. The use of activity meters led to improved estrus detection and, therefore, to a decrease in the average calving interval and subsequent increase in annual milk production. For visual estrus detection, the average calving interval was 419 d and average annual milk production was 1,032,278 kg. For activity meters, the average calving interval was 403 d and the average annual milk production was 1,043,398 kg. It was estimated that the initial investment in activity meters would cost €17,728 for a herd of 130 cows, with an additional cost of €90 per year for the replacement of malfunctioning activity meters. Changes in annual net cash flows arising from using an activity meter included extra revenues from increased milk production and number of calves sold, increased costs from more inseminations, calvings, and feed consumption, and reduced costs from fewer culled cows and less labor for estrus detection. These changes in cash flows were caused mainly by changes in the technical results of the simulated dairy herds, which arose from differences in the estrus detection rate and specificity between the "with" and "without" situations. The average marginal financial effect in the "with" and "without" situations was €2,827 for the baseline scenario, with an average IRR of 11%. The IRR is a measure of the return on invested capital. Investment in activity meters was generally profitable. The most influential assumptions on the profitability of this investment were the assumed culling rules and the increase in sensitivity of estru...

No abstract

The adoption rate of sensors on dairy farms varies widely. Whereas some sensors are hardly adopted, others are adopted by many farmers. A potential rational explanation for the difference in adoption may be the expected future technological progress in the sensor technology and expected future improved decision support possibilities. For some sensors not much progress can be expected because the technology has already made enormous progress in recent years, whereas for sensors that have only recently been introduced on the market, much progress can be expected. The adoption of sensors may thus be partly explained by uncertainty about the investment decision, in which uncertainty lays in the future performance of the sensors and uncertainty about whether improved informed decision support will become available. The overall aim was to offer a plausible example of why a sensor may not be adopted now. To explain this, the role of uncertainty about technological progress in the investment decision was illustrated for highly adopted sensors (automated estrus detection) and hardly adopted sensors (automated body condition score). This theoretical illustration uses the real options theory, which accounts for the role of uncertainty in the timing of investment decisions. A discrete event model, simulating a farm of 100 dairy cows, was developed to estimate the net present value (NPV) of investing now and investing in 5 yr in both sensor systems. The results show that investing now in automated estrus detection resulted in a higher NPV than investing 5 yr from now, whereas for the automated body condition score postponing the investment resulted in a higher NPV compared with investing now. These results are in line with the observation that farmers postpone investments in sensors. Also, the current high adoption of automated estrus detection sensors can be explained because the NPV of investing now is higher than the NPV of investing in 5 yr. The results confirm that uncertainty about future sensor performance and uncertainty about whether improved decision support will become available play a role in investment decisions.

A prognosis of the likelihood of insemination success is valuable information for the decision to start inseminating a cow. This decision is important for the reproduction management of dairy farms. The aim of this study was to develop a prognostic model for the likelihood of successful first insemination. The parameters considered for the model are readily available on farm at the time a farmer makes breeding decisions. In the first step, variables are selected for the prognostic model that have prognostic value for the likelihood of a successful first insemination. In the second step, farm effects on the likelihood of a successful insemination are quantified and the prognostic model is cross-validated. Logistic regression with a random effect for farm was used to develop the prognostic model. Insemination and test-day milk production data from 2,000 commercial Dutch dairy farms were obtained, and 190,541 first inseminations from this data set were used for model selection. The following variables were used in the selection process: parity, days in milk, days to peak production, production level relative to herd mates, milk yield, breed of the cow, insemination season and calving season, log of the ratio of fat to protein content, and body condition score at insemination. Variables were selected in a forward selection and backward elimination, based on the Akaike information criterion. The variables that contributed most to the model were random farm effect, relative production factor, and milk yield at insemination. The parameters were estimated in a bootstrap analysis and a cross-validation was conducted within this bootstrap analysis. The parameter estimates for body condition score at insemination varied most, indicating that this effect varied most among Dutch dairy farms. The cross-validation showed that the prognosis of insemination success closely resembled the mean insemination success observed in the data set. Insemination success depends on physiological conditions of the cow, which are approximated indirectly by production and reproduction data that are routinely recorded on the farm. The model cannot be used as a detection model to distinguish cows that conceive from cows that do not. The model validation indicates, however, that routinely collected farm data and test-day milk yield records have value for the prognosis of insemination success in dairy cows.