Authors of abstract: C. Kamphuis, J. Burke, J. Jago

Affiliation: DairyNZ, Hamilton, New Zealand

Body of abstract:

The hypothesis tested was that sensor data from milk meters, pedometers and weigh scales would help farmers in detecting lame cows earlier. Sensor data from ~5,500 cows were transferred automatically to a central database each evening, starting in November 2010. To date, 362 cases of lameness have been recorded with sensor data available for analyses. Each case was randomly matched to a non-lame cow. Compared with non-lame cows, lame cows have lower activity and weight, and present themselves later for milking several days before clinical detection. Activity and milking order showed significant differences between lame and non-lame cows on the day the lame cow was detected.