Validation of a daily automatic routine for dairy robotic milking and concentrates supply.

The feasibility and short term consequences of a robotic milking and concentrates supplementation routine in an automatic milking system dairy were tested. The system comprised a selection unit and a milking unit equipped with a milking robot. In a two-phase 34 day experiment with 16 Friesian-Holsteins, cows reported voluntarily to the selection unit throughout the day. Concentrates could be allocated in both the milking and selection unit. During the last 11 days, the milking robot was available for milking for 24 h a day. A set of variables and mathematical equations describing the visiting pattern and concentrates supplementation was devised. The selection unit occupation time (about 10 h/d), together with the number of visits (about 95 visits/d), provided a good performance measure of the system capacity. It proved possible to maintain daily automatically controlled milking and concentrates supplementation. Not all concentrates (about 90% of planned) were supplemented as planned. To achieve the planned allocation of all concentrates the system must allow for revision of allocation decision during the day based on individual consumption and visiting patterns. The use of a selection unit enables control of cow traffic and the concentrates allocation, but might slow down the traffic between the selection and milking units by a period of up to 5 min passage time. A mean shorter passage of 3?8 min/visit between the selection and milking unit, as compared with the exit time when cows are referred to the feeding area might indicate that cows prefer the milking unit to the feeding area. With a twice daily milking regime 95% of the milking visits to the selection unit were voluntary, which means that cows were brought to the unit only in 5% of all milkings. It is predicted that about 10% of all cows would be unable to adapt to the automatic milking system routine, and would have to be culled from the herd