The utility of sensor technology to support reproductive management on dairy farms

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. Some sensors like activity meters, electrical conductivity, weight floors and somatic cell count sensors are commercially available. Adoption has in general been low and mainly driven by the AMS, with a clear exception for estrus detection. In practice, the economic benefits of using sensor systems has not been proven. So, to make sensors live up to their full potential there is a need for research to shift from technical development towards practical applications and integration with operational farm management. Estrus detection sensors can have a good detection performance and are currently applied by farmers in practice, therefore this thesis focusses on sensors that support reproductive management. The main objective of this thesis is to study the utility of sensor technology to support reproductive management on dairy farms. This main objective was split in five sub objectives that each study a part of the main objective and were discussed in the separate chapters of this thesis. We demonstrated that utility of sensors for reproductive management can be found in economic benefits (estrus and calving detection), reduction of labor (calving and estrus detection) and more detailed management information (prognosis of insemination success). So, automated estrus detection aids reproductive management. From this thesis the following conclusions can be drawn: The developed theoretical framework describes four levels of sensor development, which should all be included in proper development of sensor systems. The literature review showed that no studies developed sensor systems with regard to management and decision support. It was possible to improve the prediction of the start of calving compared to a model that only uses the expected calving date. However, predicting the start of calving within an hour was not possible with a high sensitivity and specificity. There was financial merit in the use of calving detection, because the sensor system enables more timely intervention by the farmer. The uncertainty about the positive effects was large, which caused a wide range in the simulated financial benefits. Investment in a sensor for estrus detection was on average profitable with a return on investment of 11%. Profitability was influenced most by the heuristic culling rules and the expected increase of the estrus detection rate between detection by visual observation and the sensor. Routinely collected farm data can be used to estimate a prognosis on insemination success and be used to determine whether an individual cow has a higher or lower than average likelihood of insemination success. Integration of this prognostic model with an estrus detection sensor has potential. Currently farmers only adopt sensors for estrus detection or because they were standard with an AMS. A reason for this is that sensor systems do not produce clear information for farmers. Sensor technology should be focused on management support of applications. Labor benefits of sensors are important for adoption of sensors by farmers, farmers value flexibility, increased family time and less physical workload as benefits. However, economic evaluations of technical solutions are unable to quantify these benefits. Sensor research should consider the preference of farmers regarding labor. For the appraisal of sensor technology new methods to value labor benefits of sensor are needed. Furthermore, in sensor development societal acceptance should be an important consideration. Animal rights activists may frame the use of sensors as a form of industrialized farming. Only using technical arguments and considerations to explain the benefits of sensors will hamper the societal acceptance of modern dairy farming. Application of sensors on dairy farms should be communicated smartly to society in terms that relate the values of citizens.