The Automatic Milking Systems (AMS) are well studied in dairy cows applications, but not on Buffaloes (Bubalus bubalis). Though, a few applications to the species can be found in literature. This study was carried out in a farm located in Paestum (SA), Southern Italy. It has about 230 lactation animals and to the best of our knowledge it is the first farm in the world that introduced the AMS in the case of buffaloes. The objective of this study was to optimize the AMS occupation rate in order to use it with Buffaloes. The experimental trial has been two years long. During the first year, a functional screening of barn with 4 AMS has been done. During this year, operational and functional aspects, such as number of buffaloes, milkings and rejections per AMS/year have been analysed. The working time and milking downtime per AMS/year have been calculated and additional number of buffaloes per AMS/year has been evaluated. It has been estimated the number of buffaloes that could be added and could be milked during the year per AMS (19 ± 3). During the second year, the occupation rate of one AMS has been increased. 70 buffaloes, chosen randomly among primiparous animals, have been added in the same AMS where the year before there were 51 primiparous buffaloes. The random chosen buffaloes, with DIM of 5 ÷ 10 days, have been bred with similar alimentation rates (TMR) and management proprieties as the previous year, as well as DeLaval AMS robot parameters (42 kPa vacuum; 60 cycles/min; 7 h between two milkings). In order to evaluate AMS functionality to an increase of buffaloes number, they have been determinated: 1) functional parameters of AMS during the observation year such as working times, occupation rate, number of milkings made by AMS and milkings distribution during 24 h; 2) productive parameters such as milking frequency e milk production rate per AMS during the observation year. Furthermore, milk quality during the second year has been compared with milk quality of previous year in order to evaluate if the buffaloes number increase could affect stress and uneasiness within the group, and then milk quality. Milk composition parameters have been evaluated: % fat; % protein; % casein; % lactose. Results have been validated using an analysis of variance (ANOVA), setting the significance level (α) to 0.05. Buffaloes number increase in AMS during one year caused an increase (+18 %) of milkings compared to previous year; an increase of total milk production (+55 %), a decrease (-27 %) of AMS not working time during a year; an increase (+15 %) of system occupation rate. Statistical analysis showed no significative difference of this increase on daily milking frequency in AMS during one year and daily milk production per buffalo in AMS during one year.