The main objective of this research is to analyze the stability of the milkproduction series throughout the use of statistical process control techniquesin a way to offer to the companies, which receive the product innatura an estimative of the periods of bigger and smaller milk production.To accomplish the main purpose, a sample was collated of somesmall producers in Santo Ângelo city. The sample was composed by fiftyweeks from July 20th, 2009 toJuly 12th, 2010. Firstly, the series behaviorwas analyzed graphically and aftera SARIMA(2,1,0)(1,0,1)7model wasfound.A control charts was established in the original variable and inthe residues data from the SARIMA model, as a way to understand theautocorrelation effect in the control chart. Making a comparison amongthese control charts is possible to understand the autocorrelation effectpresent in the original data. Based on the original charts it is possibleto verify that the series has a seasonal period represented by summerand winter months. Comparing the control charts, the original data presentedmore out-of-control sample points besides control limits than the residual series of the math model. It is necessary to emphasize thatthe producers should create alternative methods in order to maintain aminimum stock of cattle feed. In the summer is necessary to maintainan enough level of water to the cattle and to irrigate the pasture. Besidesthese factors the sanitary conditions must be maintained.
