Animal husbandry aims at managing the environment of animals to enable them to express their production potential. However, ruminants are often confronted with perturbations that affect their performance. Evaluating the effect of these perturbations on animal performance could provide metrics to quantify how animals cope with their environment and therefore better manage them. For dairy herds, milk production is an accessible data that has been extensively used to predict lactation curve. However, a typical milk yield time series rarely follow the ideal lactation curve. It often presents several stalls, with variable frequency, intensity and shape. In this study a lactation curve model with explicit representation of perturbations was developed. Fitting the model on milk time series generates two curves in only one run. The first represents a potential theoretical curve without perturbation and the second corresponds to the perturbed curve (explicit individual perturbations modelled by way of exponential functions). Perturbation characteristics (starting dates, intensity and shape) were estimated on complete lactation data from 10 individual goats which have undergone digestive perturbations. Results showed that 7±2.2 perturbations were detected by goat. Residual standard deviations were 52±6% lower than a classical milk production model. Loss of production due to perturbations was 16.7±12.7%. By incorporating explicit representation of perturbations, the model allowed the characterization of both potential milk production and deviations induced by perturbations, and thereby comparison of animals. These indicators are likely to be useful to move from raw data to decision solutions in dairy production.