REAL TIME ACQUISITION AND ANALYSIS OF MILK CONDUCTIVITY DATA.

An on-line monitoring system operated in the milking parlor and collected data on milk temperature and milk yield in addition to milk conductivity data. The system had an electronic interface board with an onboard CPU. A microcomputer communicated with this interface board and collected, displayed and stored the data. In analyzing the data, a representative conductivity value for a quarter for a milking was established from several values available. Using this statistic of MHV (mean high value) ratio, a method for mastitis detection was established whereby 3 consecutive values of 1. 10 or more of this statistic would be an indication of infection. Using this approach it was possible to screen all infected quarters. There were 0. 4% false positives. This system will be very simple to implement. The microcomputer will then be able to automatically identify possible infections as soon as the third consecutive milking is completed for each cow.