Lameness is a major problem in modern dairy husbandry and has welfare implications and other negative consequences. The behavior of dairy cows is influenced by lameness. Automated lameness detection can, among other methods, be based on day-to-day variation in animal behavior. Activity sensors that measure lying time, number of lying bouts, and other parameters were used to record behavior per cow per day. The objective of this research was to develop and validate a lameness detection model based on daily activity data. Besides the activity data, milking data and data from the computerized concentrate feeders were available as input data. Locomotion scores were available as reference data. Data from up to 100 cows collected at an experimental farm during 23 mo in 2010 and 2011 were available for model development. Behavior is cow-dependent, and therefore quadratic trend models were fitted with a dynamic linear model on-line per cow for 7 activity variables and 2 other variables (milk yield per day and concentrate leftovers per day). It is assumed that lameness develops gradually; therefore, a lameness alert was given when the linear trend in 2 or more of the 9 models differed significantly from zero in a direction that corresponded with lameness symptoms. The developed model was validated during the first 4 mo of 2012 with almost 100 cows on the same farm by generating lameness alerts each week. Performance on the model validation data set was comparable with performance on the model development data set. The overall sensitivity (percentage of detected lameness cases) was 85.5% combined with specificity (percentage of nonlame cow-days that were not alerted) of 88.8%. All variables contributed to this performance. These results indicate that automated lameness detection based on day-to-day variation in behavior is a useful tool for dairy management.

De brochure bevat innovatieve ideeen, die aantonen dat natuurgebieden geen bedreiging hoeven te zijn voor veehouders, maar juist mogelijkheden bieden voor verdere bedrijfsontwikkeling. Daarvoor is samenwerking tussen veehouders en onderzoekers, provincies, gemeenten, natuur- en milieuorganisaties en ketenpartijen essentieel.

Dynamic feeding is an innovative application for concentrate feeding of dairy cows. Daily individual settings are derived from the actual individual milk yield response to concentrate intake. This response is estimated using an adaptive dynamic linear model. Optimal daily individual settings for concentrate supply are directed to achieve the maximum gross margin milk returns minus concentrate costs. This response curve plays a key role in the application. The response curve is derived from a mechanistic model for milk production and can also be established empirically from daily milk yield development during early lactation when concentrate supply increase is linear. A test application for dynamic feeding ran for several months in 2008 and results from 145 cows at one farm on 17 December 2008 have been used to demonstrate the variation in individual response. The gross margin, milk returns minus concentrate costs, varied from 2.52 to 26.32 €/day. The estimated response parameters provide insight in variation between individuals concerning the effects of concentrate and base ration intake on daily milk yield. Economical and nutritional aspects can be evaluated for each individual. Individual dynamic feeding towards an economic optimum indicates that excessive changes in individual bodyweighti can be prevented.

The trend of growing farm sizes is expected to continue in the coming decades. Increased herd sizes should not lead to less attention for the health and welfare of the individual animal. Moreover societal requirements for food safety and quality and for the production environment increase. Current developments in wireless sensor technology provide good possibilities for real-time acquisition of information about the production environment (climate, weather, housing) and production factors (animal, food, pasture). Localization and tracking of individual animals also contains interesting information for operational management decisions. Examples are data about social interactions in a herd (ranking, grouping and seclusion behaviour) and frequency and duration of visits to interesting locations in barn or pasture (milking robot, feeding station, water trough and pasture border). An overview of available localization technologies is presented. The application of a wireless sensor network using received signal strength (RSS) for a simple localization protocol is discussed based on experimental data. Finally possible development directions for a future localization and tracking system are suggested.

Daily concentrate allowances for individual dairy cows are usually based on empiric models. These models are generally based on regression equations derived from population data and do not take into account individual and temporal variation. An application was implemented in common practice which consists of an adaptive model for estimating the actual individual response in milk yield on concentrate intake using individual real time process data. Before the application was implemented, a prototype was developed by a team consisting of biometricians, animal nutritionists and ICT application specialists. It was tested in an animal experiment and further developed into a proof of principal, which was implemented for testing in a common practical setting on a research farm. Because the results were very promising, a workshop was organised to introduce the concept to software, hardware and feed industries where they were challenged to participate. In the next collaborative phase with industry involvement the further implementation into a management system was stepwise: (1) technical documentation of algorithms, (2) programming, (3) verification of algorithms, (4) on-farm implementation of the integrated software, and (5) on-farm evaluation. During the implementation it became clear that steps 1 to 3 were not difficult to perform and did not take much time. Steps 4 and 5 were more complicated because: (1) correct data must be generated from the management system as an input for the model, and (2) the output of the model has to be interpreted correctly for calculating concentrate allowances in the management system. However, not only technical aspects of an implementation process are important, also the communication with end users and stakeholders requires particular attention, for successful implementation of a new concept. While testing and implementing the application it became clear that end users and stake holders were willing to accept and use the innovative concept but interpreted the outcome based on traditional population knowledge and paradigms.

Met de nieuwe generatie draadloze sensoren, die momenteel in ontwikkeling is en langzaam voor de praktijk beschikbaar komt, kunnen steeds meer signalen van het individuele dier automatisch worden vastgelegd, zoals een naderende geboorte

The daily behaviour of dairy cows reflects the health and well-being status. The behaviour can be monitored with accelerometers (used as a tilt sensor to measure the angle) in a wireless sensor network. The angle of a leg reflects the lying or standing behaviour, the angle of the head might reflect the eating behaviour. An experiment was carried out at an experimental farm during 50 days. The cows were indoors during the first 36 days and had access to a pasture on the last 14 days. Six cows were equipped with two 2D accelerometers, one attached to the neck and one attached to the right hind leg. The accelerometers were attached to wireless sensor nodes. The acceleration of the neck and leg was recorded every halve minute (average of seven measurements with 1 Hz measuring frequency). Based on calibration measurements, the acceleration of the leg and the neck were both transformed to the angle. A cow was standing when the angle of the leg was more than 45°, otherwise lying. The method to transform the acceleration to angle and behaviour appears to be appropriate, it is possible to monitor the cow's behaviour with a wireless sensor network equipped with accelerometers.

Op 5 juni 2008 maaide een trekker, met maaimachine met daarop het prototype van een GPS-gestuurd nestenalarm, een perceel gras in Nijkerk. Daarmee werd de 1e stap gezet naar automatisch om weidevogelnesten heen maaien

Precision Livestock Farming (PLF): dat is de naam, waaronder een internationale groep van landbouwwetenschappers aan nieuwe informatie- en communicatietechnologie werkt voor de veehouderij. Deze technologieen, die buiten de landbouw al veelvuldig worden toegepast, moeten het mogelijk maken om individuele dieren beter te volgen op de steeds grotere bedrijven