Weight of cows affects a large number of factors. Regular weighing and data processing can detect differences that may indicate disorders requiring nursing interventions, e.g. nutritional deficiencies, incorrect fetal development and health problems. The current weighing systems operate as stationary - the animal is fixed, identified and weighed. However, the procedure is time consuming and operation, and that is way this system is used minimally. That implies the need of complete automation of all activities associated with the weighing, which enables introduction of  pass – through weight. The aim of this thesis was to develop a methodology for evaluating health and nutritional status of the herd based on data from an automated system for weighing a live weight of dairy cows. There was used in the weighing unit for milking robots Astronaut A3 (Lely company) to obtain weight data of individual cows. There were selected dairy cows with the longest period of lactation or already drying off, and especially dairy cows with various health problems for study. Limiting values of weight changes were established after assembling a general equation of mass curve. In the sphere of the diseases there was manifested only ketosis in the weight curve with a loss of 10.2 kg / day (38% weight loss). Additionally, the completion of growth during the first 2 periodes of lactations and weight gain due to advanced pregnancy were confirmed. The maximum daily weight difference recorded in healthy animals was 7 %, equivalent to 40 - 45 kg. The results of the study will be applied for compiling algorithm that will be implemented in the complete management system of cattle breeding, monitoring the dairy cows every day and highlight possible deviations exceeding of physiological changes in weight.