Timely and accurate oestrus detection is crucial for optimizing reproductive efficiency in dairy herds. This study evaluated the potential of monitoring rumination time and body activity using sensor technology to identify oestrus events in lactating cross bred H.F cows. Rumination time, body activity data were collected from 100 cows from Chahal dairy farm for 180 days. Gold-standard oestrus events were identified by visual observation and confirmed by veterinary examination. Machine learning models were developed to predict oestrus events based on deviations in sensor data patterns. Results showed that both rumination time and activity index could detect oestrus with reasonable accuracy. Rumination time decreased by an average of 19.27% (95.83 minutes) compared to the baseline during estrus period (P<0.05). Activity index increased by 53.93% compared to non-oestrus days. This study demonstrates the value of monitoring rumination time alongside activity metrics for oestrus detection. The proposed automated oestrus alert system using sensor data can improve reproduction management on dairy farms. Further research with larger cow populations and additional farm management factors is warranted to optimize prediction models.