Modeling the lactation curve is an important step for assessing the true milk potential of dairy animals. The present study aims to investigate the use of four different mathematical models (Wood, Cobby and Le Du,Cappio-Borlinoand and Dhanoa) to describe camel milk lactation curves, to estimate the potential of dairy she-camels and to identify different factors that could influence produced milk quantity and quality of Maghrebi she-camel. A total of 813 records from one experimental farm were used. Data collections were daily made with three milking per day. The complete milking was performed on two quarters (one posterior and one anterior). The other two were reserved for the calf and the volume collected was doubled. Among the four used mathematical models, the Wood model appeared the most appropriate according to mean square prediction error (MSPE), coefficient of determination (R2= 83.56). The differences in estimated total milk yields between the models were not statistically significant. All models were adequate in describing total milk yield, although total milk yield estimated using the Wood model was very close to total milk yield. The quantities of daily produced milk differed among individuals. Milk production peaked approximately at 3rd months postpartum and then decreased. Daily production was 6.72 ± 2.46 L. Milk yield decreased with lactation. Daily milking order as well as stage of lactation affects milk yield and its composition. These constituents became concentrated as lactation proceeded, and protein was substituted by fat. Calving date had a similar concentrating effect on fat whereas it decreased protein. This study showed that among the population of camels in Tunisia, improving environment and management of camel can be a way to improve milk production.