The trend of increasing cow numbers per husbandry person makes it difficult to achieve good reproductive management by traditional means. A system to predict the reproductive status of cows would be of great value. There exist time-series models for detecting oestrus from milk traits other than progesterone and decision strategies for interpreting progesterone measures have been suggested but a system combining these aspects to predict not just oestrus but reproductive status throughout the reproductive cycle is lacking. The objective of this study was to develop such a model on the basis of automated milk progesterone sampling.